If you work on a computer, you likely spend a good chunk of your week looking at charts, numbers, graphs, and spreadsheets. But if you’re anything like me, a lot of this time can feel wasted. We all want to be more data driven, but few of us are ever taught how to do it well.

Understanding Variation by Dr. Donald J. Wheeler is the best book I’ve read on how to use data to make more effective decisions. It’s no surprise, Dr. Wheeler has been teaching people how to use data for more than 40 years, and was a student and colleague of Dr. W. Edwards Deming2 for over 20 years.

Or to put another way, the work of Dr. Wheeler and Dr. Deming gives you a the path toward Process Power like the Toyota Production System.

Traditional status reports are bad

The way data is typically communicated in status reports compares the current period with the previous period or the same period in the prior quarter or year. In general, status reports tend to include rows that look like this:

While the numbers shown above are an improvement on last year, they’re worse than they “should” be according to the plan. If you were the metric owner, you’d have to explain why and outline steps to get things back on track.

This sounds like the right thing to do. The number is worse than you want, so you need to make it better. The issue is that you can’t learn from this data without its context and what is provided isn’t enough.

Data must be presented in context

Comparing the current value to a prior value has two issues:

1. The amount of data is small

2. Both numbers are subject to variation

These two issues make it hard to know how much of the difference is due to variation, and how much (if any) is due to real change. You should only compare two values as part of the broader context.

If two data points aren’t enough, should we present every value?

Tables of data are better than direct comparisons, but they provide too much detail. Humans struggle to process them. We need to present data in a human readable form.

Time series provide context in a human readable form

Times series graphs (or running records) typically have days, weeks, months, quarters, or years on the horizontal axis and values on the vertical. Scanning from left to right is the passage of time.

Time series communicate data faster and more completely than tables. They make it easy for humans to compare the current value to previous values, and to see if the current value is unusual.

In general, graphs and other visualizations are the way to provide context without information overload. They include all values in a visual—rather than digital—way.

The table above is graphed in a time series below:

Histograms compress data and make comparisons easier

A histogram (or tally plot) is an accumulation of the different values as they occur without the time component. A mark is placed on a value each time it’s observed. Tally plots compress data which make them useful for side-by-side comparisons.

If we plot possible values of our data set on the horizontal axis, the vertical axis represents frequency. The two histograms below were constructed from the table above, grouping values into groups of ten thousand:

Stacking time series can make comparison easier

Another way to compare two time series is to stack them on top of each other:

Average and range: Two numerical summaries of data

In addition to time series and histograms, numerical summaries of data can be useful:

1. Average is the sum of the values divided by the number of values, typically known as x̄ (X-bar). Think of the average as a measure of location of the set of values. Averages are typically used to compare two data sets. The set with the greater average “exceeds” the other set. Be careful with averages, in most data the majority of values are not equal to the average.

2. Range is a measure of dispersion of a set of values defined as the maximum value minus the minimum value, typically known as R.

Numerical summaries supplement graphs, but never replace them. Other measures of location and dispersion exist but average and range are sufficient for most uses.

The first principle for understanding data

Dr. Walter Shewhart’s3 two rules for presenting data form the basis for honest statistics. You should follow them:

1. Present data in a way that ensures all predictions that may be made can be made. Provide a table of data, graph, and context. Data divorce from context is in danger of distortion.

2. Never mislead people into taking actions they wouldn’t take if data were presented in a different form. Averages, ranges, and histograms summarize data, but they also obscure the time-order. If the time-order shows a definite pattern, don’t obscure it. Always include time series graphs when summarizing data.

These two rules can be summarized into one principle: No data have meaning without context. Don’t trust anyone who cannot, or will not, provide context for their metrics.

The most common methods of analysis are flawed

Before you can interpret data, you must have a method of analysis. Sadly, the two most common methods—comparison to specification and comparison to average—are flawed.

These methods of analysis encourages a binary worldview and don’t consider the impact of variation. They simple ignore variation and treat every change as signal.

Comparison to specification

Plans, goals, budgets, and targets are specifications. If you’ve read Andy Groves’ High Output Management, the specification approach to management will sound familiar. Andy was famous for applying the principles and discipline of manufacturing to management. And the tradition continues with objectives and key results (OKRs).

If you haven’t read Andy’s book, the idea of a goal being a specification may sound odd. OKRs come from manufacturing. Manufacturers compare product measurements to specification limits to ensure production quality. Managers compare outcomes to OKRs in an attempt to do the same.

Comparison to specification compares the current value to a target value, and deems it as acceptable or unacceptable (in or out of spec), resulting in a binary outcome. Those who are in spec are given a pat on the back. Those who aren’t are reprimanded or fired.

In other words, as long as you hit your target, you’re okay. OKRs have their place, but they should only be used for things that can be measured, and ideally, things that are activities rather than outcomes. You want people to feel like they have control over whether they can hit their target(s).

Another common mistake is setting unrealistic targets. Specifications should be based upon careful analysis of past data, current actions, and likely future conditions. They should not be based on what you want to be possible. Arbitrary targets aren’t helpful, and are often detrimental.

If you pressure people to meet an unrealistic target, they can:

1. Improve the system

2. Distort the system

3. Distort the data

Improving a system is hard. People need time to understand which inputs affect which outputs, and the power to change the inputs to achieve the desired result. If they don’t have the power to make changes, it’s always easier to make data look more favorable.

The specification approach can tell you where you are, not how you got there, or how you’ll get somewhere better.

Comparison to average

There are times where you must hit specification. You want no fraud, downtime, or data breaches. In these situations, it’s common to compare the current value to the average. Like the specification approach, comparing to the average has a binary outcome: above or below average.

Whenever the current value varies too far from the average, someone will ask for an explanation. The issue is the average is generally near the midpoint of the data, so you should expect to be above the average half the time and below the other half.

If you (or your manager) fail to appreciate this variance and apply too much pressure to the metric owner to fix it, they will distort the system or the data to make it look better than it is.

While comparison to the average is similar to comparison to specification, the average at least comes from the process rather than a (potentially) arbitrary target. Unfortunately, the average is only part of the process and doesn’t fully convey context.

Control charts: An approach that accounts for variation

The goal of data analysis is to gain knowledge so our predictions become more accurate over time. We look to data to detect changes, so we can uncover when something we’ve done is impactful. This sounds simple, but anyone who has tried knows how hard it is to do in reality.

A large part of this is due to variation. According to Shewhart, there are two types of variance:

1. Routine variation is expected even when the process hasn’t changed.

2. Exceptional variation is outside the bounds of routine and is interpreted as a signal of change.

The noise introduced by routine variation is what confuses and clouds comparisons between two values. Until you know what is noise, you can’t truly understand what may be indicated by a value.

To separate routine and exceptional variation, Shewhart created the control chart (or process behavior chart). A control chart begins with the data plotted in a time series with an average line and limits computed from the data to detect changes. Control charts provide a better approach to data analysis. They overcome the shortcomings of the two methods above because they explicitly factor in variation.

The control chart shown below includes 122 consecutive points that measure the daily conversion rate of a signup process. The fact all values fall within the computed limits with no obvious trend suggests the process is in control. This means the daily conversion rate should continue to fall between 0% and 9.78% with a long term average of 4.89%, unless the process changes in some fundamental way.

That’s a mouthful, so let’s break it down. The power of control charts lie in their limits. The limits are placed symmetrically on either side of the average at a distance which filters out virtually all routine variation4. By visualizing the extent of routine variation, the limits allow you to quickly see any exceptional points. If all points on a control chart fall between its limits, the process is said to be in control or predictable.

You can expect an in control process to stay in control going forward unless something changes. This means you can define how much movement in a metric you should expect to see in the future (e.g. how far the next point may differ from the average).

When a point (or points) fall outside the limits of a control chart, you know it’s worth investigating the cause so you can better understand how to improve your process. The presence of exceptional variation signals that a change you made has had an impact or that there are inputs which affect your process that you don’t understand.

Note how much value control charts provide. By characterizing variation as routine or exceptional, control charts shift the focus from the result and toward the behavior of the system. When a system is predictable (like the signup process above), it’s already performing as well as it can under its current design. Looking for improvement without making changes is a waste of time. We must focus on making changes.

If the system were unpredictable, it would be futile to try improve it. We would need to understand what is causing the unpredictability, stabilize it, then try to improve.

The second principle for understanding data

All data contains noise, some data contains signal. But before you can detect signal, you must filter out noise. This is why control charts are a more powerful form of data analysis than the traditional approaches.

Control charts show you the signal and stop you from attaching meaning to every value. They give you a way to know when it’s safe to extrapolate, and if it is, define the range of values you should expect to see.

By filtering out routine variation (noise) and highlighting exceptional variation (signal), control charts help you distinguish between signal and noise which is the foundation of meaningful data analysis. If you don’t have a good way to separate signal from noise, you will:

1. Interpret noise as signal: At best this leads to actions which are inappropriate and at worst, completely contrary to what you should do.

2. Fail to detect when signal: This often appears hand-in-hand with the specification approach to analysis. For example, the underlying process changes but the values are within spec, so no one digs into why.

You can avoid the first mistake by never reacting to any value, but doing so guarantees many mistakes of the second kind. And you can avoid the second mistake by treating everything as signal but this guarantees the first mistake.

You will make one of these mistakes if you can’t differentiate between signal and noise, and you’ll never properly analyze data. Control charts strike a balance between these errors and minimize both errors.

If you don’t use control charts to analyze data, you’ll always be at a disadvantage. Effective data analysis begins with separating signal from noise and control charts are the simplest way to do it.

What you choose to analyze is important

We collect data to drive decisions, and as we’ve learned proper use of data requires the separation of signal from noise. If we can’t, our actions may be inconsistent with the data. But how do we choose what to analyze?

For the purpose of comparison, let’s use the traditional approaches to data analysis—comparison to specification and averages—to pick a metric to analyze. If you’re anything like me, when faced with a table of data like the monthly marketing report shown below, you’ll hone in on the percentage difference columns. The goal is to find what has changed most dramatically and then investigate why.

There are three problems with this approach:

1. The size of a change partially depends upon the base number. A one unit change from 10 to 11 is a 10% change. A one unit change from 100 to 101 is a 1% change. Percentages show the relative size of a change, not the actual amount. Comparing one change to another is not a reliable way of finding interesting data.

2. Comparing rows in a report by comparing the size of the percentage differences assumes all lines should show similar variation month-to-month. In reality, each row has its own inherent variation. Some lines show larger percentage differences, others smaller. Choosing what to focus on based on which has the biggest difference guarantees some lines get more attention than they deserve while others don’t get enough.

3. Period-over period changes–such as the 8% decrease in cost per lead in February compared to the previous year–don’t account for variation. The difference may be due to an unusual value in the past rather than an unusual value in the present. If last February’s cost per lead was unusually high, the reduction may not indicate a trend of decreasing cost but rather a return to baseline.

Comparing percentage differences is an unreliable way to find signal. Large percentage changes don’t necessarily indicate signal, and small percentage differences aren’t necessarily noise.

With that said, it’s common practice so let’s use it to select a metric to analyze.

Using control charts for analysis

The first metric that popped out to me was average session duration, which is 33.3% lower than the plan. First, we’ll consider the usual approach to interpreting data and then we’ll construct a control chart.

If you were responsible for increasing average session duration to meet the plan’s three minutes for the month (or five minutes for the year), what would you do? I’m guessing such a large deviation from the plan would require you to create a plan to get back on track, or at least set up a meeting to talk about what you’re doing.

This sounds great in theory, but many of these reports become works of fiction. Their only purpose is to present something that shows something is being done about the perceived problem.

The whole idea behind creating a plan to get back on track assumes the change in average session duration is exceptional. But is it? To get to an answer grounded in data, we need to filter out the signal from the noise. And to do that, we need to go beyond direct comparison and start looking at historical data:

I don’t know about you, but I don’t get a lot from that table beyond the fact that it might be decreasing. Let’s use construct a control chart to get a better view.

The average of these 26 values is 3.54. We can use this as the average line in our time series graph of average session duration:

As we guessed, February is below average and there does seem to be a small trend toward lower average session duration from 2022 to 2024, but this isn’t enough to answer whether February is exceptional.

We need limits to filter out the routine variation from exceptional variation, which means we need a way to measure variation.

Moving ranges: The measure of variation over time

To measure month-to-month variation we compute the absolute value5 between successive monthly values. These values are called moving ranges:

The absolute value between January 2022’s 4 and February’s 3 is 1 as is the difference between February’s 3 and March’s 4. Repeating this process, we get all the moving ranges from our previous graph:

The time series graph of these moving ranges is shown below:

These moving ranges directly measure the month-to-month variation. Their average is called the Average Moving Range and is included above.

The two graphs in an XmR chart: X-chart and range chart

To construct a control chart for individual values and a moving range (an XmR chart) we begin with the average session duration and moving range time series charts. The first graph is the X-chart and the second is the range chart or the moving range chart.

How to compute the limits of an XmR chart

To obtain the Upper Range Limit for the moving range chart, we multiply the Average Moving Range by 3.27. The value of 3.27 is a constant required to convert the average rating into an appropriate upper bound for ranges.

This Upper Range Limit is plotted as a horizontal line on the moving range portion of the combined graph. In the example below, we multiple our Average Moving Range of 0.88 by 3.27 to get 2.88:

This completes our moving range chart. There is no Lower Range Limit because absolute value is never negative.

Now we need to construct the limits for our X-chart. The Upper Control Limit is found by multiplying the Average Moving Range by 2.66 then adding the average line of the X-chart:

The Lower Control Limit is found by subtracting the the Average Moving Range multiplied by 2.66 from the average:

In our average session duration example, we have an Upper Control Limit of 5.88 (3.54 + 2.66 x 0.88) and a Lower Control Limit of 1.2 (3.54 - 2.66 x 0.88). If the Lower Control Limit calculation results in a negative number, and your process cannot produce negative results, use 0 instead.

The complete XmR chart is shown below:

How to interpret an XmR chart

Let’s run through how to interpret our newly made XmR chart. Month-to-month variation is on the moving range portion of the chart:

The Upper Range of 2.88 means the Average Session Duration needs to move by more than 2.88 minutes in a month to be exceptional. A change of this amount from one month to the next is excessive and likely due to something.

The actual monthly values are shown on the individual values portion of the X-chart:

The limits on the X-chart define how large or how small a single monthly value must be before it represents a definite departure from the average. Here, a value over 5.88 minutes would signal that average session duration has materially improved. Likewise, a value below 1.2 minutes would be a signal of a downward shift.

In either case, you would be justified in asking for an assignable cause for the change.

This means February’s 2 minutes is not, by itself, a signal. There is no evidence of any real change in average session duration. It is likely routine variation, which means looking for a cause is a waste of time.

When you show management a control chart, they may not be happy with the amount of variation shown between the upper and lower limits. This example shows that average session duration could differ by over four and a half minutes month-to-month. “Surely our average session duration should be more steady than that.”

It’s not, and it won’t be until you make a fundamental change to the input metrics that result in this average session duration. The limits define what average session duration will fall between as long as things stay the same.

If we bring our attention back to the plan for the year, we see it calls for an average session duration of five minutes:

Our XmR chart shows this hasn’t been done in the past and is unlikely to happen in the future unless there is a fundamental change. The goal of five minutes does nothing to change or improve the outcome.

Dissatisfaction with the limits cannot be cured by finding some alternative method of computing them. Calculating them differently6 defeats their purpose.

If management is not happy with the variation shown by the limits, they need to give you resources to improve the site to generate the desired result, rather than setting an arbitrary goal that isn’t based on past performance.

There is a second problem here, average session duration is a output metric, not a input metric. Average session duration can only be improved by improving pages per session or time on page. Both of these metrics also impact other metrics. If we aren’t careful, pressure to increase average session duration could hurt other parts of the process such as new leads or conversion rate.

Small changes can contain signal

To reinforce why using percentage differences to select data to analyze is wrong, let’s look at on-time payments. On-time payments didn’t change in February and only moved by -0.6% year-over-year:

Most people wouldn’t bother analyzing on-time payments because the percentages changes are small to nonexistent. The XmR chart shown below shows this is wrong:

As you can see, four of the individual values fall outside the limits. There is too much variation in this time series to be due to chance alone. The four values outside the limits are signal, even if the changes in raw percentages is small.

We should have been looking for an explanation as to why the percentage of on-time payments materially decreased in March 2022 and increased in March, April, and May of 2023. But if we relied on comparison to specification and averages, we would have missed it.

Percentage changes are not a reliable way to find signal

In the average session duration example, we assumed a large percentage change represented signal. Analysis revealed no signal. The change was within the limits and routine. Interpreting it as signal and directing resources to fix it would be a waste of time. In the on-time payments example, we saw how signals can be missed if we rely on the traditional approach to data analysis.

The traditional approach results in wasted effort and missed opportunity. It guarantees an excess of both types of mistakes people make when interpreting data.

The control chart filters out the probable noise and detects signal in data. And when noise is filtered out, we minimize the number of times we interpret noise as signal and we limit the number of times we miss signal.

Focusing on percentage changes alone is not enough to identify signal.

Two other patterns to look for

There are two other patterns you need look for. The first is eight or more successive points on the same side of the average line. The X-chart shown below shows on-time payments became detectably higher than the average in March 2023. However, we can see the run7 may have begun in January 2023:

So, while there was a detectable signal in March 2023, the improvement could have begun as early as January 2023. Why January? Whenever eight more more successive values fall on the same side of the average line, it’s safe to assume the time series has shifted away from the average.

You can apply this rule even when none of the individual points fall outside the computed limits. Eight or more successive values on one side of the average line is roughly the same as getting eight or more heads (or tails) in a row when flipping a coin.

Given the odds of that happening eight times in a row is less than 1 out of 128, we can interpret eight or more successive points on the same side of the average as signal.

With such strong evidence of change, you can begin to ask the right questions and hopefully identify an assignable cause for the shift in a timely manner. When you look at the graph (even without the limits), the fact there was a change is apparent. The limits just help to clarify what the graph reveals.

The second pattern is a sequence of three out of three, or three out of four consecutive points that are closer to a limit than the average line8. In an alternate world our on-time payments could have looked like this:

In the chart above, no points break the limits but we have two periods of exceptional variance. The red section has three consecutive points closer to the lower limit than the average, and the green section has four consecutive points closer to the upper control limit. In both cases, we would want to investigate the cause of these changes as they are too large to have occurred solely by chance.

You need don’t a ton of data to make control charts work. Uncertainty in limits does decrease as the amount of data used to compute them increases, but useful limits can be constructed with five or six values.

The best analysis is the simplest that gets you to the right answer

The most popular method of analysis—comparing a current value to a prior value—doesn’t filter out noise. Large differences can be all noise, and small differences can be all signal. But that doesn’t mean we should overcomplicate things. The best analysis is the simplest analysis that gets the job done.

Software can’t solve all your problems

As data analytics tools have become more popular, more managers now have access to real-time data for any routine measurements of their business. It’s never been easier to generate reports on a weekly, daily, or even real-time basis.

No more waiting for the end of the month for a surprise. More access to data is good, but software can’t solve all your problems. You still need to invest time to interpret the graphs.

Imagine you’re a Growth Product Manager and your job is to increase signups for your product. You measure signups in an analytics tool like Posthog:

Can you see anything interesting in the data?

Remember, we traditionally used the current value compared to last month’s value to understand how we were doing. This table above is much more comprehensive than a limited comparison.

But does it help you see anything of interest? Eyeballing data is analysis by osmosis—you first soak up the data then you try to understand it.

Unfortunately, this is not reliable. Different people detect different things from the same data. If your reports only show tables of numbers, you are communicating in an impoverished way. It’s always better to graph your data, even if you don’t use an XmR chart:

Look how much easier it is to detect patterns with a graph versus a table of data.

Each spike corresponds to the last week of the month. You post this graph in your company’s Slack and ask if any other teams are doing anything at the end of each month. Your Head of Marketing replies and says they’ve been sending a newsletter in the last week of the month. Given how well this newsletter is driving additional signups, you might want to test increasing its cadence.

The point of this is to show how graphs present interesting information in higher fidelity than tables of data. Tables of data tend to overwhelm people with extraneous detail. Graphs sweep away complexity and reveal structure present in the data.

In some situations, such as the example above, the signals are so obvious and easy to understand that you don’t need to compute limits. You can effectively eyeball the graph to naturally filter out the noise. For most situations, changes aren’t as easily recognized and control charts are a major quality of life improvement.

If you’re not careful, data can lead you astray

Imagine you joined a fast-growing startup as the Head of Engineering in December 2023. One of the metrics you’re responsible for is the number of voluntary leavers in your team. Attracting engineering talent is hard—it takes time for them to get up to speed—so the metric is included in the executive management report, and is one of the first topics in every monthly meeting:

February, with five leavers, was far above average. More than three times as many voluntary leavers as usual, and five more than last February. In fact, the company has never had five engineers voluntarily leave in a month. You’ve had seven people voluntarily leave since you started two months ago, 3.5x ahead of the previous year.

This isn’t great and requires some sort of action, but what? In the meeting, the Head of People Operations suggests you should aim to cut the number of voluntary leavers by 50% next year, and there is widespread agreement.

But voluntary leavers are a result, not a cause. You cannot directly manage it by goal setting. The data can be distorted by pressure to meet the goal—such as by changing the definition of a voluntary leaver to only those who have been with the company past probation—but what is causing people to leave will not be impacted by choosing a target out of thin air.

You need to study the system that results in people leaving, and the best way to do that is to use an XmR chart:

The Average Moving Range for this data is 1.68 and the Upper Control Limit is 6.08. Even though five voluntary leavers in a month isn’t ideal, it is inevitable as long as the engineering team averages 1.62 voluntary leavers per month.

There is no evidence in the XmR chart that the environment which results in voluntary leavers is any different in February than it was before you started.

The number of voluntary leavers in the engineering team cannot be said to have worsened. The X-chart shows a predictable system.

As long as the team averages 1.62 leavers per month, the monthly values will range from zero to six. You would need to have seven people voluntary leave in a month for the value to be exceptional.

With all that in mind, how could you reduce the number of voluntary leavers? Since the system is stable, you have to study all the reasons people have left in every month. It would be a mistake to single out February.

What you choose to measure is also important

Good analysis depends on good data, and how you interpret data is based on where (and how) it was gathered. Unfortunately, a lot of data is aggregated over so many different processes that the only context remaining is that it was collected over the same time period.

Aggregated data loses its usefulness

Imagine you’re the Head of Sales at a B2B SaaS company, each month you record the Average Time to Close9. Your OKR for the year is to get 95% of AEs closing deals in less than 30 days. This metric appears in the monthly management report as Average Time to Close Efficiency:

Once again, this doesn’t provide enough context for anyone to actually understand how the team is performing. Values for the past 26 months are shown below:

Notice how it’s the percentage of Account Executives closing deals in time, not the count. This is because the number of AEs is not constant over time. If it were, you could use count or percentages to tell the same story. Given the number of AEs does change, you must use percentages.

February’s value of 71.3% is the lowest on record. Has something changed or is it simply a low, but routine value? The simplest way to answer this is an XmR chart:

I set the Upper Control Limit of the X-chart at 100% as the computed value exceeds 100%, which is obviously not possible.

The last point on the X-Chart is below the lower limit and is therefore signal. Detectably fewer reps were closing deals in less than 30 days. The chart can’t tell you why this happened, but it does tell you something has happened. You should investigate. You may realize you hire a lot of new reps in February and they are not yet to fully ramp up. In this situation, it may make sense to adjust your metric to be the number of ramped up reps closing deals in 30 days.

There is something else we can learn from this control chart.

With the exception of February, the percentage of AEs closing deals in less than 30 days is predictable and the average is 89%. Six percentage points lower than your OKR.

While you may have the occasional month with more than 95% of reps closing deals in time, your team cannot average 95%. Your OKR is beyond the capabilities of the current system. If your management team wants you to meet (or exceed) the goal of 95%, something has to change.

You will not be able to bring about the desired changes by only setting a goal, you must improve the system. The issue is summary statistics—like the one shown above—are not focused enough to bring about change.

Measure individual inputs

The key thing to realize is every rep is different. Some reps may take longer to close deals than others. You need to consider each rep as their own system, with their own stream of data. These data streams then need to be analyzed separately to discover if they behave predictably.

If a given rep displays predictable variation, you will do no good setting a goal that is outside their limits. You will only get what they can deliver and setting a goal outside of that will only cause them to try to distort the system or the data.

And if a given rep doesn’t display predictable variation, there is no point setting a goal. They have no well-defined capability, so there is no way of knowing what they will produce month-over-month.

Instead of setting arbitrary goals at a high level, you need to analyze the data for each rep to uncover problem areas, opportunities for improvement, and most importantly what your best reps are doing differently.

This is extremely difficult to do at an aggregate level. Aggregated data tends to be useless when you’re trying to identify problems and solutions. Process improvement requires specific measures and contextual knowledge. These are more available when you are measuring individual activity.

In order to improve your output metric, you need to develop appropriate input metrics. You need to look back at the process of your best reps and find out what they are doing. Once you know, you can start measuring the input metrics that result in the outcome you want.

Good data is not an accident.

Finally, keep in mind closing 95% of deals in less than 30 days may not even be a good goal. If closing deals quicker means a lower Average Sales Price and increased churn, you may want reps to take more time even if it raises your sales costs.

Problems like this are common when related metrics are owned by different teams. Such as when Net Revenue Retention is owned by the Customer Success team and sales is not incentivised to sell to customers who stick around.

Continuous improvement means looking at the whole picture, not a narrow slice. Optimizing the system often requires some parts operate suboptimally.

How to measure rare events

An XmR chart is inappropriate when working with data where the average falls below 1.0. Counts for rare events have to be placed on a different control chart called a c-chart, which uses slightly different formulas than the XmR chart to account for rare events which occur independently of each other.

Imagine you’ve been running a SaaS startup for just over two years and occasionally have to taken you have unexplained downtime. It annoys customers, leads to churn, and you do everything possible to prevent it. Of course, when the app does go down unexpectedly, it has to be fixed and the root cause has to be properly documented to avoid making the same mistake again.

In the past, you’ve averaged one hour of unexpected downtime every 5ish months. The most recent downtime was in February:

As you can see from above, the single hour of downtime is 520% above average. When dealing with very small numbers, such as the count of rare events, a one unit change can result in a huge percentage difference. The rest of the “information” shown is virtually useless.

A more complete perspective is shown by a time series graph:

To create a c-chart for this data, we need to compute an average count. During this time you’ve had five hours of downtime. Five hours of downtime in 26 months gives an average of:

This average is used as the central line for the x-chart and the upper limit is computed according to this formula:

The complete c-chart is shown below:

Despite the single hour of downtime being 520% above the average, the c-chart shows it’s not outside the limits. This is not a problem with the charts, it’s a problem with the data. Rare events are inherently insensitive and weak. No matter how you analyze these counts, there is nothing to discover.

Instead of counting the number of hours each month, you should instead measure the number of days between downtime and then use an XmR chart. Doing this analysis would allow you to see if unexplained downtime is becoming more or less frequent as time goes on. In general, counts are weaker than measurements. When possible, you should always measure the activity and not merely count events.

You need to use control charts

We collect data to make better decisions. Before data can be used, you must interpret it. Proper interpretation of data requires that it is presented in context, and that noise is filtered out.

The traditional approach of using percentage differences to provide context is like using a time series graph with two points. It’s not enough to know if things have improved, worsened, or stayed the same. Limited comparison can be useful, but it generally doesn’t provide enough context to point toward the right decision. Nor does it filter out noise.

Likewise, comparisons to specifications, plans, goals, targets, budgets, or averages make no attempt to filter out noise. These comparisons show if you are or aren’t “in spec”. If you aren’t, they provide no suggestion as to how to get back into spec and if you are, they don’t predict whether you will continue to be going forward.

Graphs present data in context. Among the many graphs possible, the time series and histograms are the two most useful. As you should know by now, time series form the base of our control charts.

Control charts are the simplest way of filtering out the probable noise in any data. These charts help you concentrate on what is important and ignore what is not. Not using control charts to analyze time series is one of the best ways to increase costs, waste effort, and lower morale.

Statistics isn’t a spectator sport, and control charts are no exception. Find a process you want to measure, gather the data, compute the limits, and construct an XmR chart (or c-chart).

It’s only after you use control charts that you begin to understand how they work and why they work. And only then can you start to understand the inputs to your process.

Apply what you’ve learned

1. Write out a list of metrics you use and report on.

2. Ask yourself if you are collecting the right data. Both the data you collect and the data you report need to be useful, correct, and undistorted. Data which describes activity is more useful for process improvement than data that describes outcomes. Focus on input metrics you can control rather than output metrics.

3. Pick two or three metrics and plot them on a control chart.

4. Insist upon interpreting data within their context. This will require a transformation of the traditional reporting structure.

5. Filter out noise before interpreting anything as a potential signal.

6. Stop asking for explanations of noise. If there is no signal, the current value cannot be said to differ from preceding values. There is no amount of explanation, no matter how well worded and reasoned, that can be supported by data.

7. Understand that no matter how results compare against specifications, a process which is predictable is performing as well as it can in its current design.

8. Always distinguish between what you want and what the process can give.

9. Help others take action on newly discovered input metrics. Knowing when your output metrics change is the first step. Once you know, you can discover new detrimental input metrics that need to be measured and eliminated, and beneficial ones that need to be made part of the process.

Additional resources

Articles

• Commoncog’s series on Becoming Data Driven in Business

Books

• High Output Management, by Andy Grove

• Working Backwards, by Colin Bryar and Bill Carr

• Lessons from the Titans, by Scott Davis, Carter Copeland, and Rob Wertheimer

• Kochland, by Christopher Leonard

• The Goal, by Eliyahu M. Goldratt and Jeff Cox

• Toyota Production System, by Taiichi Ohno and Norman Bodek

Videos

Thanks to Jack Walsh for reading drafts and providing feedback that made this post much better than it would have been.

What makes a great CEO? The press tends to focus on the largest, most well-known companies like the Magnificent Seven1 and growth in revenues and profits.

What really matters is the increase in a company’s per share value relative to peers and the market. Relative returns matter because the start and end of a CEO’s tenure has a huge impact on absolute returns, as does their industry. It’s one thing to deliver strong returns as the CEO of a dot-com company from 1995 to March 2000, and quite another to deliver them through the dot-com crash.

If a CEO can generate significantly better returns than their peers and the market, they are great. This comes down to:

1. Running operations efficiently; and

2. Deploying the cash generated from those operations well.

Most CEOs (and the management advice they write and read) focus on operations, which are undeniably important. In contrast, most great CEOs partner with a strong COO whose oversight of operations allows them to focus on the latter task. Capital allocation, not operations, is what distinguishes good from great.

Capital allocation is the process of choosing where, when, and how to deploy and raise capital to earn the best possible return. CEOs have five choices for deploying capital: investing in existing operations, acquiring other businesses, issuing dividends, paying down debt, or repurchasing stock. And three choices for raising it: using internal cash flow, issuing debt, or raising equity.

Over the long term, shareholder returns are largely determined by capital allocation. Two companies with identical operating results, but different approaches to capital allocation will have very different outcomes for shareholders.

The problem is capital allocation is investing, and most CEOs are not skilled investors. To borrow from Warren Buffett:

“Most bosses rise to the top because they have excelled in an area such as marketing, production, engineering, administration, or sometimes, institutional politics. Once they become CEOs, they now must make capital allocation decisions, a critical job that they may have never tackled and that is not easily mastered. To stretch the point, it’s as if the final step for a highly talented musician was not to perform at Carnegie Hall, but instead, to be named Chairman of the Federal Reserve.”

This inexperience directly impacts investor returns. To borrow from Buffett again, “after ten years on the job, a CEO whose company annually retains earnings equal to 10 percent of net worth will have been responsible for the deployment of more than 60 percent of all the capital at work.”

The best CEOs master capital allocation. The Outsiders profiles eight CEOs whose average returns outperformed the S&P 500 by a factor of twenty. An investment of $10,000 with each of these CEOs, on average, would have been worth over $1.5 million twenty-five years later.

The eight CEOs in The Outsiders are:

1. Tom Murphy of Capital Cities Broadcasting;

2. Henry Singleton of Teledyne;

3. Bill Anders of General Dynamics;

4. John Malone of TCI;

5. Katharine Graham of The Washington Post;

6. Bill Stiritz of Ralston Purina;

7. Dick Smith of General Cinema; and

8. Warren Buffett of Berkshire Hathaway.

All eight were first-time CEOs, with half under forty when they started. Except for one, they were all new to their industries. They led companies in various sectors like manufacturing, media, defense, consumer products, and financial services, in both growing and declining markets. Despite differing sizes and maturity levels, and lacking any significant power, they all significantly outperformed their peers.

Each of the outsider CEOs acted more like investor than operator and understood:

• Capital allocation is a CEO’s most important job.

• What matters is the increase in per share value, not overall growth or size.

• Cash flow, not reported earnings, determines long-term value.

• Decentralized organizations release entrepreneurial energy and reduce costs.

• Independent thinking is essential to long-term success.

• Interactions with investors and the press can be distracting and time-consuming.

• The best investment opportunity could be their own stock.

• With mergers & acquisitions, patience is a virtue, as is occasional boldness.

A perpetual motion machine for returns: Tom Murphy and Capital Cities Broadcasting

“I’ve worked at a lot of corporate events over the years, but Capital Cities was the only company where you couldn’t tell who the bosses were.” — A bartender

“Tom Murphy and Dan Burke were probably the greatest two-person combination in management that the world has ever seen or maybe ever will see.” — Warren Buffett, Chairman and CEO of Berkshire Hathaway

Buffett often compares Tom Murphy’s Capital Cities Broadcasting to CBS under Bill Paley to illustrate the impact management has on long-term returns. When Murphy became CEO in 1966, Capital Cities had five TV stations and four radio stations that were all in small markets.

At the time, CBS was the dominant media business in the United States, with TV and radios states in the largest markets, the top-rated broadcast network, valuable publishing and music properties, and a market cap 16x larger than Capital Cities.

Thirty years later Capital Cities sold to Disney at a price three times that of CBS.

Every dollar invested with Murphy grew to $204 — a remarkable 19.9% internal rate of return over 29 years, significantly outpacing the 10.1% return of the S&P 500 and 13.2% return of an index of leading media companies over the same period. In words words, Murphy was a phenomenal 16.7x better than the S&P and fourfold his peers.

CBS spent the 1960s and 70s using its enormous cash flow and stock issuance to fund an aggressive acquisition program in new fields, including the purchase of the New York Yankees. Management built a landmark office in midtown Manhattan and developed a corporate structure with forty-two presidents and vice presidents.

CBS’ strategy was consistent with conventional wisdom in the conglomerate era, which espoused diversification and synergies to justify acquisitions of unrelated businesses that, once combined, would magically become more profitable and resilient.

The goal was always to become bigger.

Murphy’s approach was different to his peers. He avoided diversification, paid minimal dividends, rarely issued stock, regularly repurchased shares, and between long periods of inactivity, made the occasional very large acquisition funded by debt that would be paid down aggressively. His goal was always to make Capital Cities more valuable. “The goal is not to have the longest train, but to arrive at the station first using the least fuel.”

Under Murphy and his lieutenant, Dan Burke, Capital Cities grew into a streamlined conglomerate that focused on media businesses within its circle of competence. Burke was responsible for daily operations and Murphy for acquisitions, capital allocation, and occasional interaction with Wall Street.

Burke believed his job “was to create the free cash flow and Murphy’s was to spend it.”

Their formula was simple: focus on industries with attractive economic characteristics that they knew well, selectively use leverage to buy large properties, improve operations, pay down debt, and repeat.

Acquisitions were far and away the largest outlet for the company’s capital and Murphy was a master at prospecting. He knew what he wanted and spent years developing relationships with owners of desirable properties. Every major transaction was sourced via direct contact with sellers, he never participated in a hostile takeover. Much like Buffett, Murphy worked hard to become a preferred buyer by treating employees fairly, running acquired properties well, and leaving something on the table so sellers came away happy.

Capital Cities is what we now call a roll-up, a company that acquires a series of businesses, attempts to improve operations, and keeps acquiring, benefiting from scale and shared best practices. Roll-ups peaked in the mid- to late 1990s before flaming out in the early 2000s because many companies acquired too quickly, underestimated the difficulty (and importance) of improving operations, and collapsed under the burden of too much debt.

Capital Cities was different. Murphy moved slowly, relying on simple but powerful rules for evaluating potential acquisitions: a benchmark of a double-digit after-tax return over ten years without leverage. This meant focusing on a small number of large acquisitions that they knew were high-probability bets, and under Burke acquired properties became beacons of operational excellence.

In other words, Murphy and Burke combined excellence in capital allocation and operations. And while they frequently used debt to fund acquisitions, they never borrowed money to fund share repurchases, avoided dilution from equity offerings, and aggressively paid down the debt they had.

Financial resources are one of the two basic types of resources a CEO needs to allocate. The other, equally important part, is human. Murphy and Burke emphasized flat organizations, dehydrated corporate staff, and autonomy. The extraordinary autonomy Murphy and Burke gave to their operating managers is summarized inside the cover of every Capital Cities annual report:

“Decentralization is the cornerstone of our philosophy. Our goal is to hire the best people we can and give them the responsibility and authority they need to perform their jobs. All decisions are made at the local level…We expect our managers…to be forever cost conscious and to recognize and exploit sales potential.”

Headquarters only existed to support the general managers of operating units. There were no vice presidents in functional areas, no corporate counsel, and no PR department. Murphy’s own secretary fielded all media calls. Publishers and station managers had the power and prestige internally, and almost never heard from New York if they hit their numbers.

The guiding human resources philosophy was to “hire the best people you can and leave them alone.” Burke himself was the guinea pig of this philosophy. When he became general manager of WTEN in 1961, he began sending weekly memos to Murphy. After months of receiving no response, he stopped sending them and spent his time on improving operations. Burke said, “Murphy delegates to the point of anarchy.”

Decentralization gave Capital Cities another gift, frugality.

You can’t control your revenue, but you can control your costs and the best defense against the revenue lumpiness inherent to advertising-support businesses is cost control. One of the earliest and most often told corporate legends was that Murphy scrutinized costs as small as the cost to paint an office’s exterior. He believed the smallest operating expenditures, particularly those related to head count could have unforeseen long-term costs and needed to be monitored.

To run the company’s dispersed operations, Burke developed a legendarily detailed annual budgeting process. Each year, every general manager came to New York to present operating and capital budgets for the coming year. Burke and his CFO, Ron Doerfler, went through them line-by-line and there were almost always material changes, with particular focus on capital expenditure and expenses.

While managers were expected to outperform peers and great attention was paid to margins, Burke was as tough on under-resourcing as excessive costs. Outside of these meetings, managers were left alone and sometimes went months without hearing from corporate.

The combination of extreme delegation and cost consciousness is what gave Murphy his edge. He could buy properties and know that under Burke, they would become more profitable, lowering the effective price paid.

That’s not to say they cut their way to high margins. The key drivers of profitability in most media businesses is revenue growth and advertising market share, so they invested heavily to ensure local leadership in talent and technology.

Capital Cities’ hiring practices were equally unconventional. Murphy and Burke had no broadcasting experience and had a clear preference for intelligence, ability, and drive over industry experience. When they made an acquisition or entered a new industry, they inevitably designated a top Capital Cities executives, often from an unrelated division, to oversee the property. And they were equally comfortable giving responsibility to promising young talent which likely helped their exceptionally low turnover.

As Robert Price, a rival broadcaster, once remarked, “We always see lots of resumes but we never see any from Capital Cities.” The system in place corrupted talent with autonomy and authority and they couldn’t imagine working anywhere else.

An unconventional conglomerateur: Henry Singleton and Teledyne

“Henry Singleton has the best operating and capital deployment record in American business… if one took the 100 top business school graduates and made a composite of their triumphs, their record would not be as good as Singleton’s.” — Warren Buffett

“Henry Singleton was this very brilliant, intellectual type who could foresee all these problems that no one else could see, and he saw opportunities. Henry was as intellectual as anyone I had come across.” — Arthur Rock, one of the first venture capitalists and investor in Fairchild Semiconductor, Intel, Apple, and Teledyne.

Henry Singleton had an unusual background for a CEO, he programmed MIT’s first computer while earning his doctorate in electrical engineering and was a world-class mathematician. During World War II, he developed technology to help Allied ships evade radar, and in the 1950s, he created an inertial guidance system still in use in most military and commercial aircraft.

This was all before starting Teledyne in the early 1960s. Known as “the Sphinx” due to his reluctance to engage with analysts or the press, he was truly iconoclastic. Unlike most conglomerateurs, Singleton ran a famously decentralization organization with less than fifty people at headquarters, despite employing more than forty thousand.

A dollar invested with Singleton in 1963 would have been worth $180 in 1990 when he retired as chairman during a severe bear market—an extraordinary 20.4 percent return for almost three decades. That same dollar invested in a broad group of conglomerates would have been worth $27, or $15 if invested in the S&P 500.

Singleton achieved these results by constantly reinventing himself based on market conditions. He acquired aggressively when P/E ratios were high, bought back nearly 90 percent of Teledyne’s stock when P/E ratios fell, and used dividends as a last resort. He even went on to disassemble the company when he believed it was in the best interest of shareholders.

In the 1960s, conglomerates were the technology stocks of their day. They enjoyed high valuations due to lofty P/E ratios. Singleton took full advantage of this, acquiring a diversified set of 130 profitable, growing businesses with leading market positions in niche markets between 1961 and 1969. All but two were acquired using Teledyne’s stock. Singleton avoided turnarounds, never payed more than 12x earnings, and most purchases were at significantly lower multiples than that. For context, Teledyne’s P/E ratio ranged from 20 to 50 over the period.

In 1967, Singleton made his largest acquisition to date, acquiring Vasco Metals for $43 million. After the acquisition, George Roberts, Vasco Metals' president, became president of Teledyne. Once Roberts joined the company, Singleton removed himself from operations to focus on strategy and capital allocation. Shortly after, Singleton dismantled his acquisition team as the multiple on his stock fell, and acquisition prices rose. The lower P/E ratio of Teledyne, the primary currency for acquisitions was no longer attractive. Singleton would never make another material purchase nor issue another share of stock.

Once the acquisition engine stopped, Roberts and Singleton focused on the company’s existing operations: improving margins, reducing working capital, and optimizing for free cash flow. Cash generation became the basis for bonus compensation for all business unit managers. From 1970, the company generated remarkably consistent profits across a range of market conditions. Cash was sent to Singleton to be allocated.

Starting in 1972, Singleton embarked on an unprecedented share repurchasing spree, buying back 90 percent of outstanding shares over the next 12 years. This obsession with repurchases represented a change in thinking for Singleton, who had been an active and highly effective issuer of stock. These tender offers were almost always oversubscribed, and he always bought all shares offered.

Singleton believed that repurchases were a better method of returning capital to shareholders than dividends, and he effectively used them as a capital allocation benchmark, repurchasing his stock when it looked attractive relative to other opportunities. But repurchases only add value for shareholders if they are made at attractive prices and Singleton’s repurchases were very large bets for Teledyne, ranging from 4 percent to 66 percent of the company’s book value at they time they were announced. All in all, he spent $2.5 billion on buybacks, resulting in a fortyfold increase in earnings per share between 1961 and 1984. Not surprisingly, he bought extremely well, generating a 42 percent compound annual return for shareholders across tenders.

In the mid-1970s, Singleton took direct responsibility for investing the stock portfolios at Teledyne’s insurance subsidiaries when P/E ratios were at their lowest since the Great Depression. He increased the equity allocation from ten percent in 1975 to 77 percent in 1981, investing over 70 percent of the portfolio in five companies, and 25 percent in one (his former employer, Litton Industries). As with his repurchases of Teledyne stock, Singleton’s returns in these insurance portfolios were excellent, approximately an eightfold growth in book value from 1975 to 1985.

Between 1984 and 1996, Singleton’s thinking evolved again. He began dismantling the company, believing spin-offs reduced the company’s complexity, simplified successions issues, and unlocked the full value of their large insurance operations. In 1987, when acquisition and stock prices (including his own) were at historic highs, he concluded that the best option for deploying the company’s cash flow was to issue a dividend, the company’s first in its 26 years as a public company.

After a few successful spin-offs, Singleton retired as chairman in 1991 to focus on his cattle ranching operations, becoming one of the largest holders of ranchland in the United States. He returned in 1996 to personally negotiate the merger of Teledyne’s remaining manufacturing operations with Allegheny Industries and to fend off a hostile takeover by Bennett LeBow. During these negotiations, Singleton focused exclusively on getting the best possible price, netting shareholders a 30 percent premium to the company’s prior trading price.

The Turnaround: Bill Anders and General Dynamics

From Zip2, PayPal and X.com to Tesla, SpaceX, and Twitter, few have had as much success at creating new technology companies as Elon Musk. And as it turns out, there is a method, or more aptly, an algorithm behind how Musk makes decisions.

The algorithm consists of five steps, quoted in full from Elon Musk by Walter Isaacson:

1. Question every requirement. Each should come with the name of the person who made it. You should never accept that a requirement came from a department, such as from “the legal department” or “the safety department.” You need to know the name of the real person who made that requirement. Then you should question it, no matter how smart that person is. Requirements from smart people are the most dangerous, because people are less likely to question them. Always do so, even if the requirement came from me. Then make the requirements less dumb.

2. Delete any part or process you can. You may have to add them back later. In fact, if you do not end up adding back at least 10% of them, then you didn't delete enough.

3. Simplify and optimize. This should come after step two. A common mistake is to simplify and optimize a part or a process that should not exist.

4. Accelerate cycle time. Every process can be speeded up. But only do this after you have followed the first three steps. In the Tesla factory, I mistakenly spent a lot of time accelerating processes that I later realized should have been deleted

5. Automate. That comes last. The big mistake in Nevada and at Fremont was that I began by trying to automate every step. We should have waited until all the requirements had been questioned, parts and processes deleted, and the bugs were shaken out.

There are a few other corollaries, a lot of which remind me of parts of Andy Grove’s High Output Management, that Musk favors:

• All technical managers must have hands-on experience. For example, managers of software teams must spend at least 20% of their time coding. Solar roof managers must spend time on the roofs doing installations. Otherwise, they are like a cavalry leader who can’t ride a horse or a general who can’t use a sword.

• Never ask your troops to do something you’re not willing to do.

• Whenever there are problems to solve, don’t just meet with your managers. Do a skip level, where you meet with the level right below your managers.

• Comradery can be dangerous. It makes it hard for people to challenge each other’s work and there is a tendency to not want to throw a colleague under the bus.

• When hiring, look for people with the right attitude. Skills can be taught. Attitude changes require a brain transplant.

• It’s OK to be wrong. Just don’t be confident and wrong.

• A maniacal sense of urgency is an operating principle. It’s better to try and fail rather than analyze issues for months. If you make make things fast, you find out fast. And then you can fix it fast.

• The only rules are the ones dictated by the laws of physics. Everything else is a recommendation.

• Separating the design of a product from its engineering is a recipe for dysfunction. Designers have to feel the immediate pain if something they devised is hard to engineer. When redesigning the factory, Musk put the cubicles for the engineers right on the edge of the assembly lines, so they would see the flashing lights and hear the complaints whenever one of their design elements caused a slowdown.

• Design is not just about aesthetics. True industrial design must connect the looks of a product to its engineering. “In most people’s vocabularies, design means veneer,” Steve Jobs once explained. “Nothing could be further from the meaning of design. Design is the fundamental soul of a man-made creation that ends up expressing itself in successive outer layers.”

• Think about the cost of things from raw materials, not the market. When a supplier quoted a price of $120,00 for an actuator that would swivel the nozzle of the upper-stage engines, Musk declared it was not more complicated than a garage door opener and told his engineers to make it for $5000. Jeremy Hollman, one of the young engineers working for Mueller, discovered that a valve that was used to mix liquids in a car wash system could be modified to work with rocket fuel.

• Turn work into a video game with metrics and high scores. It’s funner that way. You will often lose and it’ll hurt the first fifty times, but when you get used to losing, you will play each game with less emotion.

“His results are a mile higher than anyone else… utterly ridiculous.” — Charlie Munger, vice chairman of Berkshire Hathaway

Everyone knows and studies Buffett, Munger, and Berkshire, but far fewer even know of Henry Singleton and his conglomerate Teledyne that went public in 1961. But many of the things we associate with Buffett and Berkshire were first employed by Singleton at Teledyne.

Singleton was more engineer than MBA, and like Buffett he say himself as more investor than operator. Teledyne was a notoriously decentralized operation, except when it came to capital allocation, much like Berkshire. He avoided Wall Street analysts and chose cash flow over reported earnings.

Over his tenure, Teledyne would acquire more than 120 companies across a variety of, often deeply technical, industries. But his approach to investing was dead simple: Buy back shares when they were cheap, issue them when they were highly valued, and then use the proceeds from operations and share issuance to buy more wonderful businesses.

Singleton ran Teledyne for almost thirty years, and compounded his investments at an extraordinary 20.4 percent per annum. If you had invested a dollar with Singleton in 1963, by 1990, when he retired as chairman in the teeth of a severe bear market, it would have been worth $180. That same dollar invested in a broad group of conglomerates would have been worth only $27, and $15 if invested in the S&P 500.

That’s why I feel safe saying that Singleton’s Teledyne Shareholder Letters are a must read for capital allocators and business leaders alike.

1961 Teledyne Shareholder Letter

This report presents the results of your company’s first year of operation. Sales for the year totalled $4,491,431. Consolidated net after-tax earnings amounted to $133,190, or twenty-five cents per share on the 519,550 shares outstanding at year-end. Additional income of $40,4851, excluded in the consolidation, raise earnings to $173,675, or thirty-three cents per share. Backlog of the company increased throughout the year, and at present totals more than $10,000,000.

Your company’s objective has been from the beginning to become a major developer and manufacturer of a broad line of electronic systems and equipments, and of the electronic instruments and components that go in such systems. Key elements supporting this objective are:

1. The establishment of a profitable and expanding operating base, and

2. The assembly and effective utilization of a strong research and development organization.

The company made substantial progress in both of these areas during the year.

The company’s operating based was expanded steadily throughout the year by a combination of acquisitions and internal growth. In October of 1960 Teledyne acquired 87.5% of the outstanding stock of Amelco, Inc. And all of the outstanding of Handley, Inc. and Mercury Transformer Corporation. These acquisitions gave the company eighty thousands square feet of floor space devoted to the development and manufacturing of electronic systems, data entry and readout equipment, magnetic devices and transformers, hight quality etched circuits and miniature trimmer potentiometers. Total employment in these corporations was about 200 people in November of 1960, and had more than doubled by year end. In July of 1961, Teledyne acquired Palmer Instruments of South Pasadena, California, with its excellent precision shop facilities and strong proprietary position in temperature control techniques and low frequency quartz crystals used in crystal filters and stable oscillators. In August of 1961 the acquisition of the minority interest in Amelco was completed, making all of the company’s subsidiaries wholly owned. Subsequent to the close of the fiscal year, the company acquired the assets and business of Linair Engineering, Inc., a leader in aircraft and missile pneumatic and hydraulic fittings, and through its subsidiaries a manufacturer of a variety of electronic components, including precision potentiometer business of Handley, Inc. and the quartz crystal business of Palmer Instruments have been consolidated with the electronic components business of Linair in a separate building in Hawthorne, California, under the name of Teledyne Precision, Inc.

Paralleling the growth of the company’s physical assets and capabilities have been the growth of its technical and development staff. The company now employs more than fifty engineers and scientists. During the year our technical people have been responsible for the acquisition of a variety of prime and subcontracts in the fields of electronic communications, data processing, computers, and automatic control equipment. Some of these contracts are for advanced research and development, others are for technical studies, while still others are for production of advanced equipment. Typical of these programs are the following: production of navigation computers for carrier-based antisubmarine aircraft, development of data processors for scientific satellites, an instrumentation study for design of space communication antennas, an adaptive (learning) computer for simulating neural networks, an air defense study, and an antisubmarine warfare study.

Subsequent to the close of the fiscal year the company received a $5 million contract for production of FADAC digital computers, scheduled to be used in the Army’s new artillery fire direction centers. This program, widely sought after in the industry, is expected to be of considerable significance to the company’s future.

In February of 1961, the company opened its Electron Devices Division in Mountain View, California, with the objective of developing proprietary manufacturing processes for the direct application of electronic solid state technology to the manufacture of electronic systems and equipment. Under the leadership of an outstanding group of scientists, this division has made notable progress. The division now employs fifty people. In addition to actively supporting the programs of the systems division in Los Angeles, at this writing the division is completing development of a number of new soldi state devices and special assemblies for sales to the industry.

During its first year, the company invested more than $350,000 in research and development. This program of company-sponsored research and development will be expanded in the future, as we continue with the development of proprietary products and the preparation of technical studies and proposals. The field of advanced electronics invites this kind of investment, and in return offers unlimited opportunities for growth. With the support of our shareholders and the dedicated efforts of our employees, we are confident that the company’s continued growth will be assured.

1962 Teledyne Shareholder Letter

This annual report covers fiscal year 1962, your company’s second year of operations. Sales for the year totalled $10,438,367, up 132% from the $4,491,431 of our first year. Earnings, net of applicable taxes, were $331,518, a gain of 148% over the $133,190 earned in 1961. Earnings per share advanced 100% amounting to 50 cents per share on the 654,857 shares outstanding at year end.

The company now has approximately 1,500 shareholders, the majority of whom were added during the past year. It therefore seems appropriate at this time to review certain of the company’s objectives and plans. From its inception the company has planned to become a major supplier, for military and industrial applications, of electronic end-item equipment and systems, and of the specialized electronic components that make up such systems. The various steps that have been taken to date in the company’s expansion reflect the implementation of this plan.

Systems and Equipment — At the present time, the business of our Electronic Systems Division may be classified into three principal categories:

1. communications equipment,

2. Computers and data systems, and

3. guidance and control systems.

Our work in these three fields spans the range from very advanced theoretical and experimental studies and research, through development and production. For example, in the communications area we are developing and producing advanced monopulse tracking receivers for the National Aeronautics and Space Administration, which are used in the global minitrack network for the tracking of satellites and space probes. We are also producing R-390 communication receivers and ARC-73 transmitter-receivers for the Army. In computers and data systems we are producing FADAC computers for the Army’s fire direction centers, and ASA-13A navigation computers for Navy anti-submarine patrol aircraft. For guidance and control applications we are developing high-acceleration inertial instruments for the Army Missile Command, and advanced accelerometers for the Air Force’s Aeronautical Systems Division. Systems and equipment business during the fiscal year accounted for approximately two-thirds of our total sales volume.

Precision Components — The company’s business in components increased throughout the year, and accounted for approximately one-third of total sales. Our present component activities may be classified as follows:

1. electromechnical devices, developed and produced by Teledyne Precision, Inc. This product line has now been broadened to include relays, potentiometers, switches, quartz crystals, and temperature controlled component enclosures. Particularly noteworthy in this group is our new relay, which we have been developing for over a year and which was first offered for sale by the company late in the year. Packaged in a standard transistor case, it is believed to be one of the world’s smallest and fastest relays;

2. hydraulic and pneumatic aircraft and missile fittings, a line of products added through the acquisition of Linair Engineering, one of the leading suppliers in this field;

3. transformers, an area in which our capability was considerably strengthened by the acquisition of Crittenden Transformer Works, Inc., an efficient design and manufacturing organization which has operated profitably under the same management for the past seventeen years; and

4. semiconductor devices and integrated circuits.

During the year our work in semiconductor devices and integrated circuits was greatly broadened, with employment in the Electron Devices Division of our wholly-owned subsidiary, Amelco, Inc., increasing from about 50 people to approximately 200 by year-end. An important element facilitating this unusually rapid expansion has been the development and production of high performance silicon planar transistors for our own computers and data systems. These advanced transistors and additional related types have proven to be popular with other equipment manufacturers, and we are now producing and selling them in increasing quantity. Such devices as the 2N709 and 2N917, the fastest silicon planar transistors available, are now in production in our plant; and we recently introduced the FE-200 and FE-300 field effect transistors —very high input impedance active devices, with amplification and control characteristics much like those of pentode vacuum tubes. The availability of these fundamentally new units to the electronic systems engineer provides him with a valuable element of versatility in circuit and systems design, and permits easy and advantageous solutions to many otherwise troublesome circuit design problems. We believe that our achievement of successful volume production of silicon field effect transistors represents a major step forward in device technology, and that these field effect devices may some day be as widely used as conventional transistors are today.

Simultaneously with the development of new and improved semiconductor devices, the Electron Devices Division has continued a substantial program in the development of integrated circuits. In this program, we are developing processes and techniques for applying electronic solid state technology directly to the production of complete circuits rather than individual circuit elements. The utilization of these processes will result in a significant improvement in the methods now used for the manufacture of electronic equipment. Initially, we are carrying out the development of a variety of special custom circuits for our own systems division and for other equipment manufacturers. In addition, we have developed and will shortly introduce to the market, a number of proprietary digital logic circuits such as gates and flip-flops. By providing greater reliability and reduced size and weight, together with a potential reduction in cost, such circuits as these are expected eventually to become widely used in the design and manufacture of digital computers and other digital equipment.

Facilities — To accomodate the increasing activities of our systems and components divisions, the company added substantially to its plant facilities during the year. We acquired a thirteen acre site in Hawthorne, California for the expansion of our Electronic Systems Division. We have 45,000 square feet in operation there now, and can add an additional 200,000 square feet as the occasion arises. For the expansion of the Electron Devices Division we acquired a fourteen acre site in Mountain View, California and are now constructing the initial 55,000 square foot building of a research, development and production complex at this location. Floor space now occupied by the company totals approximately 180,000 square feet.

Acquisitions — In addition to the acquisition of Linair Engineering and Crittenden Transformer, by which we broadened our component product base, our activities in electronic systems was strengthened by the acquisition of American Systems Incorporated, since renamed Teledyne Systems Corporation. This acquisition brought to us increased capability in digital computer programming know-how; in electromagnetic systems technology, including advanced antennas applicable to a variety of communications systems; and in control instrumentation as applied to the detection of propellant and oxidizer vapors at missile sites. We also made an investment during the year in Micronetics, Inc., a San Diego based company newly-formed to exploit a capability in the generation, detection and utilization of extremely short pulses of microwave energy. We acquired approximately 20% of the stock of this company, and have an option to acquire the rest under specified conditions.

Stock Options — Some of our newer shareholders may be interested in the company’s restricted stock option plan. The use of stock options to attract outstanding key personnel to our company has proven to be particularly effective. Without this ability it would have been extremely difficult, if not impossible, to have brought into the company so many highly qualified technical and management personnel in such a short period of time. The careful use of our option plan has been especially beneficial during the early stages of the company’s growth. As the company gains size and recognition, the number of options needed to serve our purpose of course diminishes. Only a small number of options have been exercised, and at October 31, 1962, there were outstanding to key employees options to acquire 85,000 shares during the next five years for a total consideration of $1,080,450. We believe that the number of shares remaining of the 150,000 originally reserved by the shareholders for stock options will be adequate to serve the company’s needs for several years to come. None of the directors have stock options.

Summary — In summary, our second year of operations proceeded substantially in accordance with the company’s original plan. Sales and profits were increased. Our proprietary product base was broadened in both systems and components. Industrial and commercial as well as government business was expanded. We are entering our third year with a record-high backlog in excess of $12,000,000, and we look forward to the further development of Teledyne as a major industrial organization.

1963 Teledyne Shareholder Letter

This report covers your company’s third year of operations. Sales for the year totaled $31,924,685, up 205% from the $10,438,367 of last year. Earnings, net of applicable taxes, were $1,280,296, as compared to $331,518 earned in 1962. Earnings per share advanced 180%, amounting to $1.40 on the 849,461 common shares outstanding at year end.

The company’s financial condition improved substantially during the year. Working capital advanced to more than $9 million, stockholders’ equity to over $8.5 million, and total assets to nearly $24 million. At year end current assets were approximately double current liabilities.

The growth Teledyne is experiencing is being accomplished within the framework of the company’s original plan, and is a direct measure of the progress the company has made toward meeting its fundamental objective. From its inception the company has planned to become a major manufacturer, for industrial and military applications, of electronic and control systems and equipment, and of the specialized components, instruments, and devices that make up such equipment. The steps we have taken in the past have all been in support of this objective.

During the three years since its formation Teledyne has achieved a significant position in the general field of electronic control systems and equipment. In addition to complete systems, we also manufacture and sell separately various subsystems used in automatic control applications. These basic elements include sensors, controllers, actuators and displays. We are producing a variety of sensors or instruments used in control systems to measure the physical quantities being controlled — devices such as probes for measuring temperature, gyros for measuring attitude, instruments for measuring acceleration and velocity, optical and electromagnetic devices for observing position, and others. In the controller area, we are active on computers for fire control, navigation, temperature control, and other purposes. Among the actuators we make for control applications are electric, hydraulic, pneumatic, and in some cases explosive. We have begun to carry out work in the display and read-out area.

We are devoting a substantial effort to our work on communications equipment to link together the elements of control systems, for applications in which the control system elements are separated by considerable distance. For example, our phase-lock tracking receivers carry out this function in the tracking of satellites and space probes.

In addition to our work on control systems as such, we develop and manufacture a substantial amount of equipment used for the check-out of control systems and their elements. Our products in this category cover the range from computer-controlled satellite and space vehicle check-out systems, to equipment to check-out the hydraulic systems of commercial transport aircraft. We manufacture equipment not only to check-out operational systems, but also to test systems during the process of their manufacture.

Teledyne’s interest in electronic and control systems extends beyond the system itself and its major elements discussed above, to include the basic components of the systems. We have established a substantial competence in control system components, and are now producing a variety of electronic, electromechanical, microwave, electro-optical, electro-explosive, hydraulic and pneumatic components — all of which are extensively employed in the systems we produce.

Now operating in twelve plant locations, Teledyne utilizes more than a half-million square feet of engineering and manufacturing floor space. This year saw the addition of plants in Florida and in Arizona, our first plants outside of California. Besides our present owned and leased plant facilities, we have earmarked for new plant construction more than twenty unused acres of company-owned land at three prime industrial locations, and we have options to purchase additional facilities.

While the greatest portion of our effort in 1963 went to support our national defense and space agencies, we continued to make substantial progress towards the development of industrial markets for our products. During the year approximately $5 million, or about 15% of our output, went to industrial customers for industrial and commercial end use.

The outlook for Teledyne is good. Your company has made significant progress in past years, and expects to do so again in 1964. The concern sometimes expressed over shifting defense department procurement policies is not shared by us — we welcome these changes, which favor the more efficient firms. Our participation in industrial markets continues to expand. We are broadly diversified within our area of interest — a field with an unlimited future. We have a strong and dedicated management group, and highly skilled employees. From the base of these advantages, we look to the future with confidence and anticipation.

Review of Operations

Electronic systems and equipment

In Teledyne Systems Corporation we continued our activities in the following fields: computers and data systems, communications, electromagnetic sensors, guidance and control, toxic vapor detection, and integrated electronic and control systems. We also initiated a program in display systems, and expanded our electronic manufacturing and service activities.

Computers and data systems. Preliminary design was completed this year on a digital computer that is triply redundant, modularized by function, and designed such that, in the event of failure in a critical area, the hardware associated with some lesser function automatically replaces that used in the critical area. The logical organization and mechanization of this machine are entirely novel, and result in a great simplification of the programming burden normally associated with digital machinery. The design also achieves truly meaningful modularity, flexibility, and growth capability. Such a computer is practicable by utilizing special microelectronic integrated circuits produced by our Amelco Semiconductor division to provide dramatic increases in piece-part reliability, coupled with great reductions in size, weight, and cost.

Among several other digital programs underway in Teledyne Systems, of particular interest is our advanced high speed digital data recording and analysis system for McDonnell Aircraft, to be used in the ground support system o the Gemini two-man spacecraft.

Work is also proceeding on the POLYLINGUA system, a computer program system to automatically produce translation programs for the problem-oriented programming languages (such as FORTRAN and COBOL) commonly used by programmers. The sale of such translation programs should prove profitable since the computers which now exist as well as those which are under development all need such programs. More important, the theoretical basis of POLYLINGUA system has wide application, and should thus lead to important new strides in a number of related areas. For example, these methods have direct application to the machine translation of languages, and to information retrieval; and they may serve as the basis of a new generation of computers, vastly more efficient in their processing of common problems than previous machines. The theoretical basis of POLYLINGUA is being expanded with these applications in mind.

Communications equipment. Teledyne Systems is continuing development of advanced monopulse tracking receivers, after successfully phasing into production this year — our first products of this type. In addition to the expansion of this product line, studies are being performed for the Navy on advanced communication systems requirements for future years. Experimental work is being carried out, directed toward the use of parametric converters in various types of communication equipment.

Electromagnetic sensor systems. Several antenna programs requiring significant advances over existing technology have been undertaken this year for the Navy and Air Force. An example representative of the advanced techniques being employed in antenna systems design is a program for the development of a combination radar and directional IFF antenna for the Navy. The radar portion of the antenna consists of an X-band waveguide slot array with an elliptical aperture. Important advances in array design are represented by the wide operational bandwidth of this array, the absence of azimuth beam squint with frequency, its high power handling capability, its modularized construction, and its extremely light weight. The IFF portion of the antenna is an L-band dipole array which is fed with a unique air-dielectric stripline corporate feed with unusually low loss and good impedance characteristics. Other areas of effort include multiple-beam forming systems, ground based navigation antennas, radar cross section enhancement techniques, as well as general systems analysis in areas of basic propagation phenomena, with emphasis on passive ranging. Production requirements for some of these programs are anticipated in future years.

Guidance and control systems. The development of high-acceleration inertial devices for the Army Missile Command is expanding to accommodate the need for flight test articles. Similarly, development work is continuing for the Air Force’s Aeronautical Systems Division on advanced solid state accelerometers. A program currently in process for the application of microelectronics to gyroscope control circuits shows much promise. Success in this effort would permit the design and integrated inertial navigation systems feature spectacular reductions in over-all size, weight and power.

In another aspect of Teledyne’s control system work, 1963 saw the completion of our electronic servo and hydraulic drive systems for Aerospace Corporation’s 15-foot radio telescope and Goldstone Tracking Station’s 30-foot parabolic antenna.Vapor detection systems and devices. Our activity in toxic vapor detection equipment expanded during the year. Teledyne toxic vapor detectors are now in use at Titan missile sites for the detection of nitrogen tetroxide (N2O4) and unsymmetrical dimethyl hydrazine (UDMH) vapors. Toxic vapor detection systems are also being supplied for the Gemini, Apollo and Lunar Excursion Module programs. A new, very lightweight portable detector has been developed and is now in production. A system for the detection of fluorine vapor, a new missile fuel, is currently in development.

Display systems. Early this year we saw the start of our Microeye television project. This project represents the first design effort ever attempted in developing a complete camera system specifically for space application. Starting with the image sensor or vidicon, Teledyne has established the necessary design criteria for a new type of ceramic planar vidicon tube that is fused together into one monolithic structure. This tube is presently under construction in the company’s ultra-clean electron tube laboratory.

The Microeye camera electronics is designed to provide for both internal and external synchronization. The system is capable of accommodating a variety of types of spacecraft telemetry systems, and makes extensive use of microelectronic elements supplied by Teledyne’s semiconductor division.

At present two Government contracts are supporting various phases of this company-originated program. One of these has as its aim the perfection of a system suitable for rocket compartment installation for engine and structure monitoring. The second program is directed toward the problems of human surveillance in the manned capsule.

Integrated systems. Later in the fiscal year Teledyne Systems began a detailed program definition study for the Navy on the Integrated Helicopter Avionics System (IHAS) program. This study makes use of the complete spectrum of our systems capabilities, and further draws heavily on the microelectronics capability of our semiconductor division.

Production programs. The significant advances made during the year in research and development were matched by increasing backlog and increasing shipments of production equipment. We are continuing production of FADAC fire direction computers, and of ARC-73 transmitter-receivers for the Army. With a total of over 7,000 R-390 communications receivers now on order, Teledyne has become the nation’s largest manufacturer of these equipments. The R-390 is utilized by all three branches of the Armed Services. Accepted this year for tri-service use were the M-21 and M-22 reefing line cutters designed and produced by our Ordnance Specialities group. These explosive actuated devices are used to cut parachute shroud lines automatically in air logistic support operations.

Service activities. We continued to provide computer programming services for a variety of applications, and expanded our field service organization for servicing our products in the field. Aerial Control Geotronics, our precision mapping group, operated in nine states this year, rendering services in geodetic engineering, hydrographic and topographic surveying, analytical protogrammetry, missile telemetry, and program development for electronic computation and data reduction. Major customers included the oil, natural gas, electric utility and communications industries, as well as various Government agencies.

Fluid control systems and power equipment

A diversified product line in the fields of hydraulic and pneumatic test equipment and precision electrical power conversion and generating systems were added when Sprague Engineering became a division of Teledyne.

Hydraulic and pneumatic equipment. Extensive product history dating from 1946 has firmly established Sprague Engineering as a major supplier for the aerospace and industrial markets in the field of aircraft service and test equipment. Standard proven proprietary equipment is supplied for system testing and service of hydraulic, pneumatic, refrigeration, supercharger, and other fluid and gas systems. Equipment for test of components prior to installation on new aircraft or after overhaul is provided for aircraft systems utilizing liquids of gases.

Sprague’s industrial product line includes its air-driven hydraulic pumps, supplied in a wide range of output pressures from 100 pounds per square inch to 50,000 pounds per square inch. These pumps are capable of performing many industrial jobs which would be impractical or too expensive by other means. Our air-driven gas booster permits boosting of air, nitrogen or helium to pressures as high as 8,000 pounds per square inch for aircraft and missile test work.

Sprague has been building jet engine starters since 1953. At that time gas turbine starting units were proving to be noisy, undependable, and expensive to maintain. Sprague developed and introduced a new concept using positive displacement blowers, gasoline engine driven. During the following years additional improvements have been made. We believe the current model is the best jet engine starter available. Simplicity; low initial and maintenance costs, and economical fast sure starts characterize Sprague jet starters. Users of various models include most of the major airlines, most of the jet aircraft manufacturers, and several manufacturers of jet engines.

The Magna-Kold compressor, a basic refrigeration unit operating on DC power, is gaining acceptance for boat and trailer refrigeration requirements. This item is also adaptable to a wide variety of uses in other mobile equipment requiring refrigeration units such as trucks, buses, aircraft, trains, and mobile vendors.

Static and Rotary Power Equipment. The Inet Power Equipment Division of Sprague has over many years proven its capability as a supplier of precise high frequency power for a variety of industrial and military ground support applications. A major item in the Inet product line is our brushless synchronous motor-generator set, manufactured in ratings from 1 to 250 KW. The design eliminates the brushes and brush friction, assuring minimum maintenance and trouble-free long life. Inherent in the design is maximum suppression of electronic interference. Unlike magnetic amplifier systems, no excitation power is drawn from the machine terminals, so that the wave form is that which is natural to the generator, a nearly perfect sine wave.

During the year we introduced a line of static frequency converters. The static converter eliminates wear, since it has no moving parts; it is free from noise and mechanical vibration; it gives high performance combined with high reliability and long life; and it permits electronic frequency control. These advantages give the static converter immediate applications to variable speed machine tools and pumps, to controlled speed processes, and to high frequency lighting.

Utilizing the latest semiconductor developments in such areas as high power silicon controlled rectifiers, our design technique has resulted in static converters superior to those previously available. This development is expected to assist materially our planned penetration of the field of industrial process control. With a present producible range of sizes through 6 KVA, significant inroads have been made into this rapidly expanding market.

Microwave components

During the year our proprietary position in precision microwave components was broadened through the acquisition of the Quantatron line. These reliable high speed devices, used for rapid switching of RF and microwave power, employ either a “make-before-break” design, or a unique fail-safe mechanism. Our microwave switches are the smallest and lightest available. They are currently utilized in a variety of operational systems, such as the Hawk and Talos missile systems, and in the airborne radar and communications subsystems employed on the A3J, A5C, and F4H carrier-based aircraft.

Fluid system components

During 1963 Teledyne’s Afco Linair Division became the nation’s leading supplier of hydraulic and pneumatic fittings for aircraft and missiles.

In addition to increasing our proprietary line of standard precision fittings, Afco Linair qualified this year as one of the only two sources approved for the production of MC precision fittings for NASA procurement requirements.

Also this year we completed the development of an entirely new patented type flareless fitting for very high pressure commercial and industrial applications. These fittings, to which we have given the trade name “Flomet”, have successfully passed rigid laboratory tests and have been released for production and sale. The Flomet fitting has a unique close coupling feature — it can be assembled easily, and is capable of operating at twice the internal pressures of equivalent competitive fittings available elsewhere.

To fulfill the stringent requirements imposed by our national space program, we have designed, constructed, and tested a precision tube-flaring machine which furnishes consistent flares in all types of tubes, within the precision tolerances necessary to assure leak-tight joints. Our machine uses a completely new technique in flaring. Flares are produced with measurements accurate to better than a thousandth of an inch. Tubing material that cannot be flared at all using conventional methods, can be quickly and precisely flared using our technique. These machines will be manufactured and introduced for sale in 1964.

Optical systems and components

In keeping with our plan of developing know-how in specialized components of significance in electronic systems and equipment, 1963 saw the formation of Teledyne Optics. As the science of optics has been extended from the range of visible light to cover more of the electromagnetic spectrum, the techniques of optical production have grown in complexity and refinement. Teledyne Optics was formed by acquiring and combining three companies with complementary capabilities in development, design, and fabrication of optical components and systems. Typical applications are to automatic navigation devices, tracking cameras and telescopes, observation satellites, visible ultraviolet and infra-red detection systems, and orbiting observatories. The Red Eye, Shillelagh, and Falcon missiles, all heat seekers, employ the Teledyne Optics Irdome, through which the missile sees the target’s infra-red radiation. Teledyne Optics also supplies the lens assembly for an optical character reader employed in a nationally used business machine system. Cylindrical, aspheric, anamorphic and multiple lens elements are fabricated on a regular production basis. In addition to the traditional glasses, Teledyne Optics has wide experience in silicon, sapphire, germanium, arsenic trisulfide, fused silica and strontium titanate. A cooperative program, involving the joint efforts of Teledyne Optics and Teledyne Systems, is planned for the application of their combined know-how to advance electro-optical systems.

Electronic components

Teledyne Precision, Inc. continued to extend its product line during the year — a line that now includes relays, potentiometers, switches, quartz crystals, temperature controlled component enclosures, thermal probes and temperature sensing devices, temperature oriented control and test equipment, and military and commercial slip ring and brush assemblies. The acquisition of Control Dynamics Corporation complemented TPI’s sub-miniature electro-mechanical relay business by adding a line of full and half-size crystal can relays, plus a number of different designs of microminiature types.

The addition of Electro Development Company enhanced our switch business and established us as a supplier of high quality slip ring assemblies. Electro Development originated the hot pressure molded method of manufacturing military and commercial slip ring and brush assemblies. Our slip rings are widely used to transfer electrical power and signals between structures rotating relative to each other, as in synchros, gyros, and various navigational devices.

In the area of more complex devices, Teledyne Precision has under development for the Martin-Marietta Company a 100-pole double-throw switch to be used for relaying telemetry signals from the Titan III missile. Each of these complex switches utilizes 100 of our ultra-miniature TO-5 relays.

Semiconductor devices and integrated circuits

The business of our Amelco Semiconductor Division may be separated into four specialized categories of proprietary products: advanced silicon planar transistors; differential amplifiers, and other special assemblies of discrete devices; integrated circuits; and field effect transistors. These products were selected because the high degree of technology currently required for their production allows us to compete effectively with other larger suppliers, and because we believe these products will eventually account for the majority of semiconductor production.

During the year our position in each of these product categories continued to improve, and we introduced to the market several new products in each area. We have become a major supplier of differential amplifiers, and our field effect transistors are, we believe, more widely used than those of any other supplier. We expect the use of field effect devices to increase very rapidly as they become better known and more generally available to the electronic circuit designer. Now utilized primarily in military applications, large industrial markets should open up as the production cost of these devices is brought into line with those of conventional transistors.

I field of integrated digital microcircuits our OMIC (optimized microcircuits) line was recently introduced. This line, in which the transistor design and interconnections have been developed from the viewpoint of the particular circuit application, numbers twelve basic logic circuits and is now being shipped from stock in production quantities.

During the year Amelco Semiconductor has continued to cooperate closely with Teledyne Systems on a variety of projects. Among these is circuit design work on the IHAS program, development of microelectronic circuits for the Microeye miniature vidicon, and the development of very low power integrated logic elements aimed at high component density systems for space applications.

1964 Teledyne Shareholder Letter

As it has each year since its inception, in fiscal 1964 your company again achieved record sales and earnings. New highs were also recorded in orders booked, number of people employed, amount of plant space utilized, and other measures of business activity. Sales were up 20 per cent to $38,187,127. Net earnings for the year increased to $2,545,388, a gain of 98 per cent over the $1,280,296 reported for 1963. Earnings per share advanced to $2.35, a 68 per cent increase over last year’s $1.40.

At year end Teledyne’s financial condition was the strongest in the company’s history. During the year working capital increased 54 per cent to $14.2 million, stockholders’ equity advanced 58 per cent to $13.7 million and total assets increased 46 per cent to $35.0 million. The ratio of current assets to current liabilities was 2.4 to 1.

Even more important than the gain in operating results was the progress made during the year in broadening and strengthening the company’s foundation for future growth. Our product base was expanded; plant facilities and equipment were increased and improved; industrial and commercial activity was stepped up; and several important acquisitions were made.

Substantial additions were made to plant facilities, with the amount of floor space in use increasing 30 per cent to a total of 650,000 square feet at year end. New production and test equipment was added throughout the company, and we continued our program of upgrading and modernizing existing machinery. Book value of company-owned facilities rose to over $7 million, a gain of 40 per cent over last year.

Teledyne’s fundamental dedication to the field of automatic control systems, equipment, and components was exemplified by the broadening of the company’s product base that took place during the year. The products in this field, including sensors, computers, actuators, displays, and communications links, are characterized by a high rate of technological obsolescence, and dictate a strong dependence on technical innovation and leadership. Your company continued to invest heavily in research and development throughout the year. As a result many new products were introduced, ranging from highly advanced integrated analog circuits to jet engine starting systems. Additional new products are under development, and scheduled for introduction in future years. Balancing our internal product development program is our systematic addition of selected products through acquisition. This year acquisition brought us such important new products as communications equipment, explosively actuated devices and aircraft display equipment.

Our participation in industrial and commercial markets continued to grow. Non-Government shipments approximated $7 million, a 40 per cent increase over last year. Subsequent to year end we further increased our industrial product base through the addition of Pacific Industrial Controls and Analytics Systems Company, acquired as part of United ElectroDynamics. The products of these two divisions, used throughout the nation for a broad range of automatic control applications, considerably enhance our position in the general field of industrial process control. Because of the prospective future growth in the industrial applications of electronic control equipment, we are proceeding with a vigorous program of further development in this area.

Subsequent to fiscal year end Teledyne acquired the business and assets of Servomechanisms, Inc. SMI has been active for many years in the development and production of aircraft flight instruments, being particularly well known for its air data computer systems. SMI’s capabilities complement our existing activities in flight control and instrumentation. Because of our conviction that air data instruments will be of growing importance as our airways become more crowded, we place a high valuation on the long-term importance of the SMI acquisition.

Also shortly after year end United ElectroDynamics, Inc. was acquired and consolidated with Teledyne. UED’s activities, other than the industrial controls business mentioned above, include airborne telemetry and seismic equipment and services. Through the UED acquisition we have taken a place among the leaders in these two fields, both of which are compatible with our present activities and long-range interests.

The product area of automatic control equipment is one of unlimited promise. If properly developed and wisely employed, machines have the ultimate capability to free man from labor, and to extend indefinitely the range of his intellect. It is our privilege at Teledyne to have achieved a strong position in this field of limitless opportunity. We are confident that our efforts will result in further progress in the years ahead.

Review of operations

Electronic systems and equipment

Computers and data systems

Progress continued this year on the detailed design of Teledyne’s unique new family of digital computers. The machines are potentially applicable to a broad range of real time control system requirements, where the modular design of the computers permits them to be precisely tailored to the intended application. Combining the simple functional organization and high speed response of analog computers with the accuracy and logical capabilities of digital computers, the machines make use of a combination of incremental and whole number computing techniques. Very little programming is required. Through the use of majority logic and built-in failure detection and isolation circuits which initiate automatic self-reconfiguration to bypass any failed element, virtually failure-proof operation is assured. The microelectronic integrated circuit modules utilized in the computer are produced by our Amelco Semiconductor division.

Work on several other important digital computer projects was carried forward this year. Among systems completed and delivered in 1964 was an advanced Loran A & C Digital Simulator developed for the Federal Aviation Agency utilizing a magnetic tape-stored position program. The simulator provides a real time input to the Loran receiver system under evaluation. It precisely simulates characteristics that are normally encountered in actual flight conditions, including aircraft velocity, rate of turn, ground-wave, sky-wave, coding, and doppler effects among others.

This year also marked delivery of the Computer Data Recording System to McDonnell Aircraft Corporation for use in the pre-launch checkout complex for the Gemini program. The equipment is currently installed at Cape Kennedy, and has been utilized in the successful test launches of Gemini Capsules I and II. Production also continued throughout 1964 on the FADAC fire detection computers for the army.

Final development is under way on the POLYLINGUA system, a computer program system to produce automatically translation programs for the special languages (such as FORTRAN and COBOL) commonly used by programmers. This system uses highly advanced techniques including a new theoretical model for language syntax. As a result, the translation programs which it produces are very competitive with respect to speed, and can be operated in all but the smallest computers. This programming system significantly advances the art of translation program production. Such programs can be automatically produced for any programming language, and for almost any computer.

Communications and antenna systems

The past year has seen a continued growth in the company’s activities in communications and antenna systems and related technologies. An outstanding new program is for the design, fabrication, and delivery of a feed subsystem for use on a 60-foot Casegrain reflector antenna. The system is being built for Navy/NASA satellite tracking applications, and will permit simultaneous operation at both L and X band frequencies. It will also provide a choice of either monopulse or conical-scan tracking. The same NASA/Navy system has led to a development program for a complete receiving subsystem to provide compatible operation with the feed subsystem. The receiving subsystem utilizes Teledyne’s monopulse tracking receiver in conjunction with a specially designed down-conversion unit to achieve high stability operation at UHF, L-band and X-band frequencies. As part of this unit, Teledyne is developing a step-recovery diode multiplier, the completion of which will materially strengthen our integrated systems capability.

Directly related to our existing capability in antenna systems is our continuing development of parametric amplifiers and broadband mixers. This product line is based on proprietary design techniques which permit achievement of large gain-bandwidth products and extremely low noise figures. With the addition of this capability, Teledyne is now prepared to offer complete receiving systems for low noise applications in telemetry tracking, satellite communications, and radar target location.

For several years Teledyne has pioneered in exploring techniques for microwave passive ranging utilizing a proprietary design concept. Recent studies performed for the Navy has culminated in the completion of an experimental system to verify the operational feasibility of the approach. The experimental equipment is currently undergoing flight tests under Navy sponsorship. Completion of this phase of the program is expected to lead to development of service test models and subsequent production.

Progress in production of communications equipment matched that achieved in research and development. Production of R-390 receivers, ARC-73 UHF transceivers and monopulse tracking receivers has continued, and further improvement in manufacturing efficiency has been achieved by the use of automated production techniques. Chassis and subsystem flow is controlled by motor-driven conveyor belts which position cycle between work stations in accordance with optimum time standards for each assembly operation. In addition, our machine shop facilities have been expanded to provide improved production support.

Dubrow Electronics, our production organization in Burlington, New Jersey, continued shipment of UPA-35 radar display consoles, and began production of GRC-19 communication sets. The addition of Dubrow Electronics has strengthened our over-all manufacturing capability, and gives us a greater degree of flexibility in making best use of our specialized talents in electronic assembly and fabrication.

Airborne telemetry systems

With the acquisition of United ElectroDynamics shortly after year end, Teledyne became one of the nation’s foremost suppliers of airborne telemetry equipment. In 1956 UED made the first application of solid state circuity to airborne telemetry in the Sergeant missile program, and since that time has been a leader in advancing the technology of airborne telemetry.

In 1964 UED was an important supplier of telemetry components and systems on several major programs. UED supplied both the RF links and the analog modulation systems for all operational test flights of the Minuteman intercontinental ballistic missile. Another major contribution was the continued supply of PAM/FM/FM telemetry systems to the Ballistic System Division of the Air Force for use on every operational training flight of the Atlas, Titan, and Thor missiles flown from Vandenberg Air Force Base.

In applications to manned space flight, 1964 saw development of the signal conditioning amplifiers which constitute the backbone of the telecommunications signal conditioning subsystem of the Apollo spacecraft. Production of the prototype hardware was completed, and production was started on flight units. We will continue production of Apollo spacecraft signal conditioning amplifiers during the first half of the year, and we anticipate continuing participation in the upcoming second Block II phase of the Apollo program.

Also in 1964 UED designed, qualified and delivered the pulse code modulation telemetry systems for use in NASA’s micro-meteoroid satellite (Pegasus). This program was followed by the design of an extremely flexible 80-800 channel high and low-level PCM system for use in the NASA Flight Research Center Lunar Landing Research Vehicle, and UED continued to supply a unique d-c amplifier for each Agena missile. A further contribution to the space effort is UED’s PCM RF transmitter system utilized by several prime contractors in various phases of the Saturn program.

Other projects include the development and production of an advanced telemetry system for ARPA’s Hibex (High Environmental Booster Experiment) missile. This system successfully performed through environmental conditions more severe than those sustained by any previous missile system. In support of the Navy’s Shrike missile program, UED designed, qualified and manufactured a substantial quantity of battery-operated PDM FM/FM telemetry systems. We have recently gone through the first successful firing of a new telemetry system for the Army’s medium anti-tank weapon, and follow-on production is anticipated.

Substantial progress is being made in research and development of solid state telemetry transmitters and oscillators—work being concentrated in the UHF region and in increasing the RF power level in the VHF spectrum. Advances are also being made in oscillator linearity and reliability. Development activity in 1964 produced a low-level commutator and an analog-to-digital converter for use in advanced PCM systems.

UED has developed and installed the telemetry communications systems for Standard Oil Company of California which link the offshore oil well platform at Ventura with the onshore operations office, and similar equipment was installed in Standard’s remote Alaskan fields.

Currently we have booked, or are in the process of negotiating, substantial continuing contracts on the Minuteman, Agena, and Saturn missile programs, as well as two classified Atomic Energy Commission programs. It is anticipated that these programs will continue throughout 1965. Also, 1965 will see production of an advanced state-of-the-art pulsed RF amplifier system for use by the AEC on a classified test program requiring substantial follow-on production.

Instrument and control systems

This year Teledyne began development of a low-cost Flight Reference Stabilization System (FRSS) for the Air Force’s Aeronautical Systems Division. The FRSS is intended for application as a flight reference device to a wide variety of aircraft and space vehicles. In addition to its low cost, this advanced system features minimum size, weight, and power consumption. The FRSS is potentially a large production item for future years. Completion of design, fabrication, and testing of the first models of the system is scheduled for next year.

Development work on inertial devices for the Army Missile Command was continued this year and work is proceeding on the fabrication of flight test units. These devices include a high-acceleration angular-rate measuring instrument, and a free rotor gyroscope that utilizes new concepts to supplement existing technology. We are also developing miniature precision accelerometers for the Air Force and the Navy.

Shortly after the close of the fiscal year, Teledyne expanded its position in the flight instrument field through the acquisition of Servomechanisms, Inc. SMI has been a major factor in the development of air data devices and systems. Air data information, a basic requirement for precision flight control of airborne vehicles, is of growing importance as the number of aircraft in use continues to increase.

During the year SMI completed the development and production design of a lightweight central air data computer for the Douglas A4E Skyhawk attack aircraft. The initial production order for this Navy aircraft is currently nearing completion, and a follow-on order has been received. An additional production order is expected in fiscal 1966, and should extend the program over a period of several years.

We are also supplying the air data computer for the A7A (VAL) aircraft. The design and development, under contract with Ling-Temco-Vought, was carried out this year, and an initial production order has been received. The first unit has been delivered and the second unit, which will be utilized for qualification, is near final test stage. The delivery of the initial production units extends into early 1966, and we are anticipating a substantial follow-on production order in that same year. The A7A (VAL) aircraft is programmed for heavy production for the next several years.

Another contract with Ling-Temco-Vought calls for design and development of a true airspeed computer for utilization on the RF8G aircraft. The initial production order is for some 50 units with delivery commencing in April 1965. Additional purchases are expected in the next fiscal year.

In addition to air data systems, SMI develops and manufactures the pressure transducers which provide the basic inputs to the air data computer. The division also supplies separate pressure transducers for a variety of applications. Many of our transducers are manufactured for NASA and are used on various space programs.

An important adjunct of the air data system is the angle of attack transmitter, or vane. SMI is a leading manufacturer of these precision devices, which detect local air flow and produce an electrical output proportional to the angle between the local airstream and an aircraft reference line. This year SMI qualified its universal vane for the Air Force, and we have just received an award from the Air Force Systems Command for several hundred production units under a multi-year procurement. We are currently providing vanes for the A4E, the F8E, and the A7A (VAL) aircraft, all for the U.S. Navy. Our angle of attack transmitter is also being used on the F-104, on the French Dassault Mirage fighter, and on the new Lear Jet executive aircraft.

The SMI ice detector units have been selected for the F-111 (TFX). The ice sensing mechanism is entirely mechanical and does not rely on any change of pneumatic pressures for its operation, and thus, is not affected by rain or dirt contamination. The independence from pneumatic pressures assures dependable icing indications at all speeds or even when the aircraft is on the ground. For helicopter or VTOL applications this becomes an added advantage since it permits the reliable indication of icing when the aircraft is hovering.

The SMI thermocouple reference junction compensator is receiving wide acceptance. This device electrically compensates for reference junction variations in a thermocouple temperature measurement system. We are supplying compensators for use on the Titan III program, the Minuteman program, and the Apollo Lunar Excursion Module.

Earth sciences

Teledyne’s entrance into the field of earth sciences was accomplished through the acquisition of United ElectroDynamics. We thus acquired an organization with an established capability in detecting, measuring, and evaluating earth movements. Major improvements in seismology have come about in recent years as a result of the 1958 Geneva Conference, which sought to establish specifications and methods for nuclear test detection. Our earth sciences organization has led in these improvements, and has been a prime contractor on project Vela since 1960. As part of the Vela program, the Earth Sciences division established the Tonto Forest Seismological Observatory, selecting the remote Payson, Arizona site because of the low level of microseismic earth vibration, after detailed study of 53 possible locations throughout the nation. Earth Sciences designed, fabricated, tested and installed the equipment under the direction of the Advanced Research Projects Agency with technical supervision provided by the USAF Technical Applications Center.

The Tonto Observatory monitors and records earth vibrations from more than 80 seismometers of various types, buried in the ground in vaults. A large number of these seismometers are placed in both linear and symmetrical array patters on 23,000 acres of Tonto National Forest. Each seismometer measures movement in the earth smaller than a ten-millionth of an inch. Twenty miles of trails and 525 miles of electrical cable connect the seismometers with the recording and control center. The control center records the seismic motions, and the records are forwarded to our Seismic Data Laboratory in Alexandria, Virginia, for evaluation and analysis. At the laboratory our scientists evaluate the information from Tonto Forest, as well as other seismic information supplied by a vast network of fixed and mobile recording stations, integrating and analyzing the many diverse seismic techniques to improve underground explosion detection methods. We are operating the Seismic Data Laboratory for the USAF Technical Applications Center as part of ARPA’s Vela Uniform project.

Subsequent to fiscal year end we were awarded a significant contract related to the Vela Uniform project which involves the installation and checkout of the world’s largest seismic array. The work is being performed in the eastern half of the state of Montana and covers a large portion of the state. Some 500 of our seismometers and amplifiers are being installed in holes drilled, cased, and cemented by our crews.

Earth Sciences recently became active in lunar and planetary sciences by teaming with Bendix Systems Division to submit a winning proposal to NASA Headquarters. Our portion of the program requires an extensive study of the potential experiments to be performed by the Apollo astronauts when they reach the surface of the moon. Various types of geophysical information will be obtained by means and instrumentation to be specified for this program as a result of the study. Hardware development in this area is expected to follow our present study program.

Integrated avionics systems

The Program Definition Phase for the Integrated Helicopter Avionics System (IHAS) has been successfully completed. Our report has been submitted to the Navy and preliminary negotiations are proceeding for the award of a prime contract to Teledyne to develop and supply the complete system. The initial operational application of the IHAS planned by the Navy is in the CH-53A assault helicopter for the Marine Corps; however, the high degree of modularity and flexibility achieved in the system design has led to the contemplation by the Services of the use of the IHAS in a number of other fixed and rotary wing aircraft as well. The planned broad usage of the IHAS has both Army and Navy, in various aircraft, is consonance with current Department of Defense emphasis on commonality to provide maximum capability per dollar, especially in new developments offering major improvements in system effectiveness.

As an outgrowth of experience gained on the IHAS program, Teledyne has been awarded a contract to study the avionics system requirements for the Army’s Advanced Aerial Fire Support System (AAFSS). This effort requires an extensive study of the fire control problems associated with a variety of weapons as well as computational and system problems pertaining to the over-all requirements. Paramount in this study, as in others, is the accurate and objective assessment of the cost effectiveness of the proposed system.

Recognizing the importance of the cost effectiveness concept in defense planning, we have developed, for use on large-scale computers, a generalized mathematical model which has proven to be a powerful tool in the evaluation and optimization of defense systems. We foresee continuing emphasis on the requirement for rigorous validation of planned systems from a cost effectiveness standpoint. By virtue of our considerable and early experience in this field we anticipate further growth of our activities in this area.

Industrial and support equipment

Analytic systems

Our analytical systems product line includes eight major types of gas and liquid analyzers applied generally in the petroleum, petrochemical, and chemical industries for on stream process analysis and control. The business is conducted through our Analytic Systems division, acquired shortly after year end as part of United ElectroDynamics.

The eight major types of analyzers can be classified into three main groups. The first group is composed of gas analyzers that include thermal conductivity, trace and per cent oxygen, flame ionization (hydrocarbons) and combustibles monitors. These instruments are used to measure the purity of, and the impurities in, various gas streams, with the majority of applications being in the air liquefaction industry. A less common application now being encountered with increasing frequency is to inert gas generators, where it is necessary to measure both total combustibles and trace oxygen. Other applications relate to secondary recovery in oil fields and to process control in plastic plants. In the latter case, traces of oxygen may change the color and properties of plastics, causing a loss of product and increased operating costs.

The second group of products consists of the near infrared, ultraviolet and infrared photometric analyzers. Here the main applications are centered around needs in the petroleum and petrochemical industries, but recent emphasis has been placed on other chemical industry applications, such as plastics production and fertilizer plant controls. One of the more critical applications of this group of instruments is their use in detecting toxic vapors, such as hydrogen sulfide and mercury vapor. The latter measurement is particularly important in industries where the use of mercury vapor lamps and large amounts of liquid mercury can pose a health hazard to personnel.

The gas chromatograph family of instruments is the newest addition to Analytic Systems product line. Process control gas chromatography was chiefly developed for the petroleum industry. This means of analysis and control also has great promise as a support method for nearly all of the applications previously mentioned for Analytic Systems’ other instruments. The market for process chromatography is expected to increase rapidly and we are planning to expand our activity in this field.

Pacific industrial controls

Our Pacific Industrial Controls division was also brought into Teledyne through the United ElectroDynamics acquisition. PIC manufactures more than 50 different models of pre-packaged controls which automate equipment and conveyor systems by automatically stopping, starting, reversing and changing operating speeds of electric motors. Through PIC we are rapidly becoming an important supplier of such controls. Marketed under the trade name of “Ratiotrol” and “Precision Speed Regulator,” PIC’s controls are designed to be merchandised as off-the-shelf packages by power transmission products dealers and distributors. The automating controls manufactured by PIC are marketed through the Boston Gear Works of Quincy, Massachusetts, the largest manufacturer and distributor of power transmission products in the United States.

PIC is also introducing an automated clutch and braking device for use on small (up to 20 horsepower) electric motors.

Power equipment

In our Inet division we continued this year to expand the use of our brushless synchronous motor-generator set, manufactured in ratings from 1 to 250KW. A commercial version of this equipment is used by major airlines at terminals throughout the United States. Another product development for airline ground service, and in use in the United States and overseas, is our truck-mounted, 140 KVA, 400 cycle diesel engine generator set complete with all necessary controls.

During 1964 Teledyne began supplying precise static d-c power equipment for the Apollo and Saturn programs. This development has provided a new family of SCR regulated, high current (50 to 500 ampere) d-c power supplies suitable for the most critical system requirements. Exceptional performance and reliability are achieved through the use of high power silicon controlled rectifiers and solid state overvoltage, undervoltage and overcurrent protective devices mounted on plug-in printed circuit boards.

Also this year, shipments were made against an Air Force contract on a new line of static power conversion equipment to be used on USAF Mobile Training Units for the C-141, C-130 and C-133 aircraft.

Another development of particular interest this year was the introduction of our Inet Transient Synthesizer (ITS). To be used for testing communications equipment, the ITS generates a full family of voltage and frequency transients which are programmed and recorded on the Synthesizer Visicorder. The Synthesizer consists of an SCR inverter, programmed by solid state devices, as the controllable sine wave power source; a surge generator for supplying simulated lightning and switching surges; and suitable instrumentation for indicating and recording the pertinent electrical transient phenomena. The ITS is fully portable, and is built in five separate modules, so that the size and weight of any one module is no greater than that of a large household application. This allows for easy transportability in order to permit use of the equipment for studying specific performance characteristics of critical communication loads at widely dispersed sites.

Sprague division

In 1964 Teledyne’s Sprague Engineering division maintained its position as the industry’s standard in aircraft hydraulic and pneumatic equipment. The Sprague line of mobile jet engine starters, in use for many years throughout the world, was supplemented this year by the introduction of a pneumatic jet engine starting system, for permanent facility installation. Used at new jet air terminals and for air terminal modernization programs, the starting equipment is located in a permanent installation in the utility area of the air terminal. From the utility area the compressed air is piped underground to as many starting pads as red.

In support of the manned space flight program we were selected by the Boeing Airplane Company to supply the remote control pumping system for testing the servoactuators on the Saturn V rocket engine. Our proven ability to handle systems for contaminant-free liquids under high pressures was a major factor in obtaining this program.

In another development this year, Sprague Engineering built and delivered its first hydrofoil passenger boat. Now in use at Sea World Marine Park in San Diego, this 34-foot excursion craft is the only hydrofoil operating on the West Coast which is Coast Guard approved for carrying passengers. A total of three of these hydrofoil boats have been delivered. We have also contracted with the Coast Guard to install a set of foils on one of their 30-foot patrol boats.

Following the success of the Sea World hydrofoils, we have begun construction of two 58-foot seventy passenger hydrofoil ferrycraft for the Woods Hole, Martha’s Vineyard and Nantucket Steamship Authority of Massachusetts. To be utilized for scheduled ferry service, it is expected that these hydrofoils will operate at approximately 40 knots, reducing travel time to about 25 per cent of its present value.

Instrumentation and components

Electronic components

During the year our precision components division placed special emphasis on advanced rotary and rectilinear switch assemblies for use in liquid oxygen and gaseous oxygen rocket engine fueling systems. Units of this type are being supplied for the Rocketdyne F1 and J2 rocket engines. These switches fully meet the stringent reliability requirements dictated by NASA Saturn specifications.

During 1964 we completed qualification and shipped production quantities of our 100-pole, double-throw switch. We developed this switch for The Martin Company for use on the Titan III missile as a telemetry signal transfer switch. Each of the more than 40 switches shipped contains 100 of Teledyne’s ultraminiature TO-5 relays.

We are continuing to expand our TO-5 relay manufacturing capability. To broaden the TO-5 product line, we have recently introduced two additional TO-5 configurations: a magnetic latching version, and a 40 milliwatt ultrasensitive model. Development is also near completion on a double-pole double-throw TO-5 relay. The TO-5 relays remain the smallest available, and with the increased number of models available constitute the industry’s broadest line in the ultraminiature class.

The Mechatrol line of synchronous motors, servo motors, gearhead motors and metal film potentiometers acquired in November 1964 further enlarged our activity in the components field. Typical users of Mechatrol components include the Atlas, Pershing, Polaris and Titan missiles; the B-52, B-58, F-104, F-105, F-106 and F-110 aircaft; and the Discoverer, Tiros and other space vehicles.

Optical components

Teledyne Optics’ first full year of operation since its formation in 1963 showed good progress. The division now has the largest infrared optics manufacturing and test capability in the western United States. Teledyne’s infrared optics are utilized in most of our armed forces’ arsenal of heat seeking missiles. Development work for the Ford Motor Company’s Aeroneutronics division on the optical requirements for the Shillelagh missile was completed late this year, and it is anticipated that production requirements will be released in 1965. Initial quantities of the windows for the Apollo space vehicle were produced this year, and production of Apollo windows is continuing.

Our phasolver discs, utilizing a complex evaporated metal pattern on a close tolerance glass disc to form a high resolution angular transducer, are in production at Teledyne Optics. Telescopes for NASA’s Orbiting Astronomical Observatory will be pointed and programmed in angle by the phasolvers.

Fluid systems components

The Linair Engineering division of Teledyne, one of the nation’s leading suppliers of hydraulic and pneumatic fittings for aircraft applications, expanded into the industrial fluid fittings market this year.

In addition to strengthening our position in the military and commercial aircraft fitting market, during the year nearly ten per cent of the fittings shipped were for industrial application, a new product area for Teledyne. A substantially increased market penetration for our industrial product line is anticipated for 1965. We intend to introduce a new and fully patented industrial-type fitting, in which significant improvements have been made over competing products. The fitting is expected to have eventual use in aircraft as well as industrial applications.

An important advance was also made in the development of fittings for space applications. We were recently awarded a large contract for a new type fitting designed for use in ground support equipment on the Saturn space vehicle. Selection of Teledyne for this program was significant, since it may well become the only fitting to be used in this application.

An agreement has been made with Lakeland Manufacturing company of Cleveland, Ohio, whereby Lakeland will manufacture our newly developed and patented precision flaring machine for space-type fittings. Sales will be made by both Lakeland Manufacturing and Teledyne to those companies requiring a precision tube flare. To the best of our knowledge, this equipment is the only unit that will meet NASA requirements for tube flares used in space vehicles.

Information display equipment

During 1964 Teledyne entered the field of visual information display equipment, through the acquisition of Radar Relay, Inc. Encouraging progress is being made at Radar Relay in working toward commercial application of its products. As an example, the division recently contracted to provide the complete warning and advisory system for the Douglas DC-9 short-range jet transport. Programs of this type are expected to produce significant increases in our commercial sales over the period of the next few years.

Among the technical developments which took place in this division during 9164 was the development of a new type of four-color miniature switch-indicator. The new switch is the smallest of its kind on the market. It has captured the interest of manufacturers of such equipment as computers, ground control equipment, and aircraft, and is beginning to produce quantity orders.

Under license from the Douglas Aircraft Company, our Radar Relay division has continued the development of a new method of providing advisory or warning information. Known as Auditory Information Display (AID), the system selects the appropriate prerecorded warning message and plays it to the operator of an aircraft or control station to alert him in the event of a malfunction. Tests have shown that human response to an audible warning is as much as six times faster than the response to a lighted indicator.

Working with the Amelco Semiconductor division of Teledyne, Radar Relay has now improved the AID design to a point which should place us in an excellent competitive position. Our in-house capability in microcircuits and integrated circuitry has helped bring about improved performance characteristics with less size and weight at lower cost. The AID development is now under consideration by the Navy for use as a retrofit on its fleet aircraft. It is also being considered for the TFX aircraft, and for use on commercial jet transports.

Explosive ordnance

During 1964 the McCormick Selph division made a number of significant moves to strengthen its position in the ordnance industry. One of the more important milestones was the division’s entrance into the field of specialized design and development of sophisticated instrumentation equipment. Our first product in this market is a spark trace velocity measuring system for shock tunnels, for use by NASA’s Ames Research Center. The system measures gas velocity at speeds up to ten times the speed of sound. A second item of instrumentation now being supplied is a thermocouple input simulator. The simulator is being used to check out recording systems used for documenting temperature measurement. It also has application to hypervelocity wind tunnels, rocket thrust test stands, and other areas in which large numbers of pressure, thrust, or temperature transducers are used. We expect to expand our efforts in this field of instrumentation for velocity and temperature measurement.

Our ordnance system research and development capability is reflected in a product improvement contract received from The Martin Company. The contract calls for major improvements in the Pershing missile ordnance system hardware. Through the introduction of heliarc welding techniques and the use of stainless steel a new and greatly improved hermetic design was evolved. The Martin Company is now testing prototype quantities, and we expect to go into production soon on the coming year’s requirements.

Work on the F-11 (TFX) aircraft contract is now under way and on schedule. The contract calls for the design, development and manufacture of explosive separation devices for the crew escape system. This involves the use of high explosives to separate the aircraft cockpit, or pod. The requirement is accomplished by explosively cutting the aircraft skin in a predetermined pattern and by simultaneously detonating explosive bolts which will sever the structural members. The operation is highly critical, and must be effected without damage to the pod, which remains sealed for the protection of the two man crew inside. This is the first such explosive system intended for use in tactical aircraft.

McCormick Selph continues to occupy a strong position in the production of high density components. During the year we produced over two hundred thousand explosive actuators alone. The quality of this volume production can be measured by the fact that we were one of the first recipients of the Gemini-Titan II NASA award. The award was made as a result of our explosive pressure cartridge being certified as “Critical” for the man-rate launch vehicle. To date there has not been one failure in the 30,00 parts fired.

Semiconductor devices and integrated circuits

Sales of our semiconductor division increased markedly during 1964. The growth took place primarily in field effect transistors and digital integrated circuits, and a strong marketing position has been established in both product areas. The increased sales have been realized in the face of a price decline of approximately a factor of two in injection type transistors, differential amplifiers, and field effect transistors, and a factor of approximately ten in digital integrated circuits. Thus our unit production of semiconductors in production efficiency the division continued to show satisfactory profits.

During the year the semiconductor sales organization was strengthened and expanded. The number of field salesmen in our own employment was doubled. The number and quality of manufacturers representatives and distributors was increased. The internal sales support organization was reorganized on a product basis to better service customers and the field sales organization.

Substantial progress was made during the year in reducing the cost of manufacturing our products. As volume expanded, yields have been stepped up through increased use of production tooling, and increased training of personnel. Improved equipment has been introduced in the diffusion area, and furnaces with more precise temperature control are being used on critical operations. New methods and equipment for semiautomatic testing and sorting of dice are now in use. In assembly, new lead bond equipment has been installed which operates at a faster rate. In the electrical testing of the product, equipment has been acquired which classifies the discrete devices with respect to all of their more important parameters in one testing operation. Similar equipment for digital integrated circuits has been designed and is now in use. The new test equipment and procedures have drastically reduced the cost of testing our products.

Amelco Semiconductor has become a leading manufacturer of highly reliable semiconductor components. This leadership has been gained through superior product designs and the employment of an advanced reliability program. Beginning with the stringent quality requirements we impose upon our suppliers, the reliability program culminates in the careful monitoring, evaluation, and control of our manufacturing process.

The product evaluation portion of our reliability program employs sampling of devices selected at random from our product lines. These samples are subjected to a thorough testing program in our complete and modern environmental laboratory, which simulates all conditions of actual use. As a result of the stringent process controls and resulting device reliability, we have been selected as a qualified supplier on the Titan, Star, Mariner, Polaris, Nimbus, OAO, Gemini, Pershing, TFX, Apollo, LEM, and other programs.

Our research and development program continued to be very active during the year. New products were introduced in each of our major product lines. Significant among these new products were:

1. To the injection-type transistor line, Amelco added its first high speed switches. This family of devises meets the specifications of such popular types as the 2N2368, 2N2369, 2N743 and 2N744, among others.

2. In the digital circuit area, a number of additions have been made to the OMIC direct-coupled transistor logic line, and the lower power DTL line has been broadened and offered for general sale in the past year.

3. A line of diffused analog circuits has been under development; the first two of these, a full two-stage differential amplifier and a high performance operational amplifier, are now in production.

4. The FE-100 and FE-400 field effect transistors were offered for sale. The performance of these devices is superior to that of the FE-200 and FE-300, which were introduced in the prior year. The FE-100 is a small geometry device designed for low leakage and low capacitance. The FE-400 is larger geometry device designed for high drain currents and high transconductance.

5. A number of field effect special assemblies were introduced for the first time in the past year. Among these are field effect differential amplifiers, in which two field effect transistors matched for gain and input impedance are mounted in the same package. Our SA2345 cascoded field effect amplifier, designed to have very low input capacitance and very high output resistance, was introduced. A special analog switching circuit was also announced. Used in multiplex switching, the circuit consists of an assembly of a field effect transistor, a pnp transistor and other components.

6. A silk screen thin-film process was successfully developed this year. The process involves silk screening resistors and interconnections on a ceramic substrate to which discrete transistors and capacitors are added. We are currently delivering a variety of custom thin-film circuits of this type. The process is particularly suitable for small volume orders which do not warrant the cost of developing a full set of diffused circuit tooling. In addition, the process is capable of achieving higher frequency performance than can be realized with fully diffused circuits.

1965 Teledyne Shareholder Letter

Growth, creating new opportunities through greater resources and technological depth, is a continuing objective of the company. Essential to and resulting from growth is profit. In each of the years of its history Teledyne and its earnings have grown significantly, continuously enhancing the company’s potential for further accomplishment. The spirit and enthusiasm essential to growth is as evident in the people of Teledyne today as it was the day the company was formed. It is with this spirit and enthusiasm that a review of the 1965 fiscal year and a report on activities is presented.

The growth of Teledyne is evident in many ways. Net earnings more than doubled to $3,402,000 compared to the $1,441,000 reported last year. Net income per common share increased to $1.98 from $1.34. Sales reached $86.5 million compared to $38.2 million for the previous year. Shareholders’ equity increased to $34.8 million from $13.7 million. Working capital rose to $30.8 million from $14.2 million. Book value of company owned facilities doubled to $14.7 million.

The number of Teledyne employees at year end reached 5,400, more than twice the number employed a year ago. The efforts of these accomplished people make it possible for others to join them at all levels of technical and managerial capability, and many new employees are currently being added throughout the company. Although facilities in use at year end expanded to 1,046,000 square feet from 650,000 square feet last year, plans are being accelerated to construct an additional 300,000 square feet of company owned owned engineering and production facilities.

In previous annual reports we have emphasized Teledyne’s fundamental dedication to the field of automatic control systems, equipment, and components. We have described the products of this field as consisting of several elements: sensors or measuring instruments for measuring the quantities to be controlled; computers for calculating from the measured quantities the kind and degree of control to be effected; actuators to carry out the control function; communication links to connect together the elements of the control system; and displays for the human beings concerned with the operation.

Automatic processing and control equipment is penetrating into every area of commerce and industry, and is making itself felt in every facet of our daily lives. In cultivating this limitless field, Teledyne applies itself to certain carefully selected market areas where the rate of growth and degree of opportunity is especially great. Thus at the present time the markets we primarily serve are three in number: aviation electronics, industrial process control, and earth and ocean sciences. Utilization of electronic equipment and devices is rapidly increasing in all three of these expanding markets.

Important developments took place in each of our key market areas during the year, and many of these developments are discussed in the following pages. Four subjects are given special emphasis: the Integrated Helicopter Avionics System (IHAS) being developed by Teledyne to provide helicopters with an all weather operational capability; the development of a miniaturized Flight Reference Stabilization System (FRSS) applicable to a wide range of aircraft; the installation and operation of the world’s first Large Aperture Seismic Array (LASA) for detection of underground nuclear explosions; and continuing progress in the development of circuits basic to all electronic equipment, as exemplified in our microelectronic technology.

In addition to internal developments such as those just mentioned, our capability was further enhanced through the acquisition of several talented organizations during the year. In each case, these acquisitions contributed to the further penetration of our chosen market areas. Thus in aviation electronics Automated Specialties brought us its outstanding capability in automatic throttle controls, instantaneous vertical speed indicators, and angle of attack instrumentation. In the field of industrial controls Glenn Pacific added to our capability with its line of controlled power supplies for automatic welding equipment. Our position in the earth and ocean sciences was strengthened by the addition of The Geotechnical Corporation with its know-how in seismic technology and its growing activities in the ocean sciences.

Supporting our general position in electronic technology was the addition of Microwave Electronics with its capability in the generation and amplification of microwave energy, embodied in its position in the traveling wave tube field. The acquisition of Crystalonics broadened our line of transistors and other semiconductor devices that are basic to the entire field of electronics.

Teledyne is entering 1966 with the brightest prospects in its history. Backlog is at record levels and business activity in our operating divisions is at an all time high. During this period of expanding activity, much of our effort will be directed toward strengthening, broadening, and deepening the technological and managerial capability of the company, in preparation for continued growth in the years ahead.

Review of operations

Microcircuits

Basic to modern automatic processing and control equipment is the electric circuit—the complex of materials in which electrical energy performs its information processing function. Advances in materials technology, and more particularly in the fundamental knowledge of and the understanding of how to control the electrical properties of materials, is simulating radical improvements in production of electric circuits. The extraordinary versatility of semiconducting solids, exhibited initially in the transistor and more recently in the more complex integrated circuit, has permitted these materials to assume a dominant role in the production of electric circuits.

Teledyne is a leader in the development and application of semiconductor technology. The technical jumps from vacuum tube to transistor, and from transistor to integrated circuit have been made, and our metallurgists, chemists, physicists and engineers are participating in further advances. The economic and technical advantages of integrated circuits are such that they will determine the configuration of a majority of electronic equipment for many years to come. The growth potential of integrated circuits is so great that a single program on which the company is at work today may ultimately use as many integrated circuits as was produced by the entire industry in 1965.

Here, vastly enlarged, is a mask used in producing wafers for integrated circuits. Each of the 167 integrated circuits in this wafer (a small portion of which is greatly magnified in the photograph on the opposite page) contains the equivalent of 17 transistors, diodes and resistors—a total of 2,839 components.

Against a backdrop of greatly enlarged engineering diagrams skilled technicians assemble tiny three-quarter square inch microelectronic modular assemblies (MEMAS) consisting of some 25 separate integrated circuits. During the year the technological leadership of Teledyne in the field of semiconductors and microcircuitry was re-emphasized through advances made by our Amelco Semiconductor and Crystalonics divisions. Our semiconductor business more than doubled in 1965 with increases in all product areas.

In the field of digital integrated circuits advances were made to keep step with industry requirements, and a new series of transistor-transistor logic elements is expected to receive wide acceptance in the lower power space systems market. Our analog integrated circuits now include a number of video amplifiers, pre-amplifiers and operational amplifiers. We are producing analog circuits as hybrids—which incorporate a number of discrete interconnected transistors, capacitors and resistors—as well as in diffused monolithic circuit form, with all elements in a single silicon chip about the size of a pinhead. We believe we have become one of the nation’s leading suppliers of integrated analog circuits, and we expect the use of such circuits to undergo a major expansion in the immediate future.

The company’s market position in high reliability discrete devices was enhanced through technical advances in injection type transistors, photosensitive field effect transistors, differential amplifier pairs and PNP transistors to complement the NPN line. A new line of P-channel field effect transistors was added during the year, as well as a series of epoxy encapsulated units to compete in the high volume commercial market. We also introduced a power field effect transistor that switches high currents faster than any other device. Our newly developed PNP transistors, with high voltage capacity at medium power, were offered to industry for the first time. Another development this year was an all epitaxial varactor that makes possible rapid voltage tuning.

A photosensitive field effect transistor was developed to replace phototubes, photomultiplier tubes, photosensitive cells and phototransistors. These new devices are the first with sufficient speed to permit realization of the full benefits of computerized character recognition, as used for example in fingerprint identification and automatic mail sorting. Teledyne’s capability in semiconductors is being applied to the IHAS computer, and we are simultaneously expanding our marketing operations to realize more of the sizeable potential in aviation, space and industrial applications.

Antennas and instrumentation comprise the essential elements of Teledyne’s microwave backscatter range which is used to predetermine the radar reflection characteristics of specific aircraft. A manufacturing and service activity of our Micronetics division, the techniques utilize scale models of the aircraft, and the results allow one to determine how the actual airplane will appear on the radar screen, from any angle and at any distance. Such knowledge is essential in designing air traffic control systems for commercial and military aircraft. It is also useful in designing new aircraft where it is desired to enhance or minimize the reflection characteristics for identification or security reasons. Production of our proprietary target augmentation devices, placed on aircraft to enhance their images when seen on radar, was increased during the year.

We also continued development of the microwave portion of a railway boxcar identification system now in experimental use on two railroads. The microwave reader identifies the boxcars by number as they pass, and the information is telemetered to a central computer, permitting the railroads to keep continuous up to the minute nation-wide control of all their rolling stock. Our microwave systems know-how is also being applied to the development of the microwave transmission and reception equipment used in the IHAS station keeping system.

Electromagnetic relays developed and manufactured by our Teledyne Precision division are the smallest available. Hermetically sealed in a package the size of a standard transistor, these high speed electronically controlled switches are ideal for aircraft and space vehicles, where size and weight are critical. During the year we added new versions to the line that was already the most extensive of any ultraminiature relay series. Exhaustive tests under severe environmental conditions have led to the approval of our relays for such advanced aircraft, satellite and missile programs as TFX, Saturn V, Titan III, F4H, Phoenix, LEM, Apollo and Voyager. Teledyne also produces larger radio frequency, coaxial, waveguide and stripline switches used in a variety of applications including radio altimeters for aircraft and satellites.

An astronaut’s first close-up view of the moon will be through one of the special windows produced by our Optical Products division for the Apollo spacecraft. Application of ultra-precise techniques and experienced production know-how has established Teledyne as a leader in the production of precision optical components. For example, one unusual product is our infrared dome assembly for missiles and aircraft. This meticulously produced dome is made of pure silicon. It protects the infrared detector, and serves as a transmitting window to the detector at the same time. Silicon infrared domes are used in many of our country’s detection and countermeasures systems, and again last year we produced most of these domes. We are also participating in a program to produce an optical beacon which may be used as a visual guide for the Lunar Excursion Module on its return from the moon’s surface to the orbiting command module.

Products created through technology of the McCormick Selph division are proving that for many applications explosive devices are more reliable and efficient than any other. Once thought too dangerous or uncontrollable for precise applications, explosive systems have demonstrated their efficiency in space and airborne systems and have potential in commercial as well as military uses. Teledyne explosive devices are used in Gemini flights from launch through stage separation. Other non-munitions services are found on missiles, rockets, torpedoes and aircraft. Exacting explosive techniques are being adapted to the crew escape system for the F-111 airplane. In an emergency, the explosive system cuts the structure and skin of the aircraft, allowing ejection of the complete pod containing the pilots and instruments, which is then dropped safely to earth by parachute.

Other Teledyne explosive products make use of both electrical stimulus, as in exploding bridgewire devices, and non-electrical stimulus where it is desired to keep explosive devices separate from electric circuits. Design was completed and production was begun this year of an exploding bridgewire system for the Pershing missile which provides for motor ignition, stage separation and thrust termination. A unique application is to airborne emergency release systems, in which the explosive actuated device severs metal cables on parachutes in equipment drops, and severs cables on nets towed by airplanes in satellite recovery operations.

Flight reference stabilization system

Representing a significant milestone in man’s ability to determine exactly his location and velocity as he pilots his aircraft at varying speeds and through changing maneuvers above the earth is Teledyne’s Flight Reference Stabilization System (FRSS). Under development for the past five years, the tiny instrument held in the blue gloved hand on the opposite page is now nearing completion in our flight control laboratory.

Smallest of its kind in the world, FRSS is far more acutely sensitive in the slightest movements of the vehicle in flight than the human body is to the forces it feels as the aircraft climbs or dives, changes direction, or brakes to a stop on the runway. The compact inertial sensing platform, computer and power supply which comprise the system together weigh less than ten pounds. Microcircuits are used extensively to achieve the unprecedented size and weight and, for the first time, are mounted inside the gyros and accelerometers. Production cost is expected to be no more than a third of that of existing systems of equivalent capability, while performance and reliability should be greatly improved. Under development for the Air Force’s Flight Dynamics Laboratory, this versatile system is applicable to virtually all types of aircraft and helicopters in commercial and military use.

Air data computers designed and produced by Teledyne are the in-flight electronic brains of many high performance aircraft, including the Navy A7A attack airplanes now in production. Primary function of the air data computer is to give the pilot immediate integrated flight information including air speed, mach number and altitude. The computer also supplies the autopilot, navigation and weapons delivery subsystems with information necessary to fulfill their functions. The air data computer for the A7A weighs only 17 pounds, about one-third the weight of computers formerly used in similar applications. Teledyne is pioneering new developments in air data computers and converters utilizing the advantages of microcircuits. These developments are expected to help maintain our position as a leading designer and manufacturer of these vital flight instruments.

Standing alone in the forefront of more effective utilization of the radio frequency spectrum is the traveling wave tube. Intricately designed, produced with precision, it exploits the great potential of the microwave frequencies. Advancing the state of the art in devices to generate and amplify microwave energy is one of the important challenges facing today’s electronic scientist. In our Microwave Electronics division we are meeting this challenge. Our traveling wave tubes are used in communication systems, radar, instrumentation and in the vital electronic countermeasures equipments which defend ships and airplanes from enemy radar. Our tubes are the key to the IHAS station keeping radar. We supply more TWTs and backward wave oscillators for instrumentation such as test equipment and laboratory amplifiers than any other manufacturer.

During the year we introduced improvements to microwave solid state delay lines which are expected to make important contributions to the advancement of the electronic countermeasures art. In the future new traveling wave tubes will be utilized in satellite and space communications systems, and in industrial applications where the advantages of microwaves are recognized but as yet relatively unapplied. The microwave portion of the radio frequency spectrum is the region where the potential for development is greatest, because of the large bandwidth and information handling capacity which is waiting there to be used.

Exposed to the wind like a jet age weather vane, this sensitive angle of attack transducer measures the angle of the aircraft wing in relation to the natural flow of air. The information it provides is vital to aircraft performance, especially in landing and weapons delivery calculations. It is one of the most critical instruments manufactured by Teledyne that provides the pilot with information essential to the control of his aircraft. This company developed transducer is standard equipment on numerous high performance aircraft.

A related product is our Approach Power Control System, which maintains the correct air speed for a prescribed average angle of attack, and improves overall flight path control by relating throttle movement to flight path variation. The system is especially valuable for landing on aircraft carriers and other short runways. Another such product is the Inertial-lead Vertical Speed Indicator which provides instantaneous rate of climb and descent information. The instrument is found in virtually all modern commercial airplanes and is widely used in military aircraft as well.

Telemetry, the automatic transmission of measured data and command and control information by electromagnetic techniques, is used extensively in space and missile applications and has great potential in industry. Teledyne produces a wide range of telemetry systems and components, including signal conditioning amplifiers, pulse code modulators, subcarrier oscillators and transmitters. We are concentrating a strong developmental effort in pulse code modulation and higher frequency equipment. Pulse code modulation allows rapid transmission of coded information and is expected to see wider and wider use. Higher frequency systems are of increasing importance because greater amounts of information can be transmitted over the wider bands available at the higher frequencies.

The attributes of speed and reliability that render telemetry the most effective technique for transmitting information and command and control signals to missiles and spacecraft make it equally useful in industrial and commercial applications. A typical telemetry system developed by the company automatically monitors water heights and actuates gates in dams and reservoirs for flood control.

Thousands of communications transmitters and receivers for military applications were delivered by our Communications and Dubrow divisions last year, and Teledyne continues to be a major supplier of mobile, airborne, ground and shipboard communications equipment. In addition to providing complete equipments, our factories maintain continuous production of spare subassemblies and component parts for the numerous units now in the field. The advanced production techniques developed by Teledyne are resulting in on time delivery of high quality equipment, and communications systems are expected to remain an important part of our manufacturing output.

Large aperture seismic array

Nestled securely 200 feet below this conical metal cover in the open spaces of southeastern Montana is one of man’s newest scientific tools—a highly sophisticated seismic ear sensitive to the faintest stirrings of the earth. This seismometer is just one of 525 similar instruments, deeply buried and carefully placed in a series of concentric patterns covering a 15,000 square mile area to form the Large Aperture Seismic Array, the most advanced seismological listening system ever assembled.

Potentially capable of detecting an underground nuclear explosion anywhere in the world, LASA was installed and is now being operated by Teledyne in cooperation with the Massachusetts Institute of Technology and the Department of Defense. This experimental array, the possible forerunner of a world-wide network of similar arrays, may offer important benefits to mankind beyond its vital capacity to police nuclear activity. From it, scientists may learn more about the behavior of foreshocks, the faint normally undetectable tremors that precede the arrival of major earthquakes. Perhaps LASA will also divulge information helpful in detecting and predetermining the course of tidal waves. In either event, more knowledge of these impending, highly destructive natural forces could be instrumental in the saving of large number of human lives.

The detection and measurement of earth movements is gaining importance as new applications for seismology are developed. In Los Angeles, for example, a building code now requires that recording seismometers be installed on the bottom, middle and top floors of virtually all new structures over fire stories high. Such instruments, manufactured by our Earth Sciences group, provide information on building motion during earthquakes that will lead to more efficient structural design. In another application Teledyne is making seismic studies related to site selection for buildings and other structures, as in Seattle’s Union Bay Project. To help determine the best structural design for an underwater vehicular tunnel we have installed a complex seismographic system in the bay bottom sediment. Recordings from the study will be used in determining the practicality of the site.

During the year Teledyne continued its participation in our country’s program for the detection and monitoring of underground nuclear explosions. For example, we are continuing to operate the Tonto Forest Seismic Observatory in Arizona, which includes 80 seismometers strategically placed over 23,000 acres. At the time of installation by Teledyne in 1963, it was the world’s largest seismic observatory. But this past year it gave way as the largest to LASA, the huge Large Aperture Seismic Array in Montana also installed and operated by Teledyne. The data being reported from these and other seismological stations around the world are digested and interpreted at the Seismic Data Laboratory in Alexandria, Virginia which we operate under government contract.

The evolution of seismic science has reached the point where not only can earth movement be detected, but the physical strain in the earth can be measured as well. Such advances have given rise to the possibility of earthquake prediction, and Teledyne, with others, is studying this possibility. The prediction of earthquakes could lead to great savings in life and property, and is one of many reasons that the scientific study of the earth, and its properties will continue to be one of man’s most rewarding endeavors.

Development and manufacture of instrumentation utilized in seismological studies around the world is an important activity of Teledyne. During the past year instruments of our design were installed at the Large Aperture Seismic Array in Montana to aid in gathering information from that great system. For storage of very large quantities of seismic data our Geotech division has developed an ultra-slow tape recorder with tape speeds ranging from 0.03 inches to 0.09 inches per second. Three of these recorders are now being marketed for industrial uses on a nation-wide basis. Two laboratory models record unattended for 33 days while the portable field version has 10 days’ capacity.

During the year we also announced, through the Geotech division, our entrance into marine geophysical exploration where seismic techniques have long been used in the search for oil and minerals. We are currently operating our own electric arc seismic section profiler system in the Gulf of Mexico. Sound waves from an electric arc periodically discharged in the water are reflected from strata in the earth beneath the ocean floor. The reflections are detected at the surface by hydrophones, and stored in recorders carried on the moving ship. After processing in a computer the reflected sound gives information about the underlying strata useful in locating oil and mineral deposits. To supplement the electric arc system, we have acquired an exclusive license to manufacture and use Shell Development Company’s deep water gas source seismic profiler system. This system permits efficient and speedy seismic profiling at depths to 15,000 feet.

Topographic maps are the visual results of surveys accomplished through the advanced techniques of Teledyne’s Geotronics division, which specializes in geodetic, route location, offshore and photogrammetric surveys. Our Monomeasurments system combining aerial photographs and computerized analytical solutions permits the establishment of precise coordinates and elevations for any point visible on the photographs. The system, pioneered by Teledyne, allows significant savings in survey costs because it requires only a few horizontal and vertical ground reference points to complete a survey from the air. So precise are the measurments using this method that, over a period of time, it is possible to determine ground movement of less than an inch. The end result of Teledyne’s computerized solution is direct control of digital plotters which produce a graphic display of the engineering information.

In 1965 we began work on the largest geodetic survey contract we have received, and under which we are surveying for the Army Map Service an area in Iran approximately the size of the state of California. Our geodetic capabilities are applied to a variety of projects from highway planning to precise positioning of offshore drilling rigs. For use in the oil industry the company has developed a computerized method that takes information from earth sample data, aerial photographs and property records, to correlate surface ownership with subsurface oil pools.

Careful control of oxygen content in storage environments provides marked improvement in quality of perishable foodstuffs over that achieved by the use of refrigeration alone. Teledyne’s Analytical Instruments division is a leading manufacturer of monitoring and control instrumentation for use in storage warehouses and on shipping trailers. Our process control instrumentation and gas detection equipment is marketed to the steel, chemical, pertrochemical and petroleum industries, for applications where the mixture and blending of gases or liquids is critical.

The division’s new portable oxygen analyzer, now being marketed for environmental testing and safety engineering, makes use of a proprietary micro fuel cell that determines the percentage of oxygen in the atmosphere and provides power to activate a meter control unit. One version of this analyzer is small enough to fit into a coat pocket and serve as a personal safety device.

During the year we completed the development of a new model of our toxic vapor detector. The new model, capable of detecting vapors arising from leaks in the fuel system of rockets employing proposed new rocket fuels, can be retrofitted to existing equipment. The increasing need for analysis and control of gases and liquids in the process industries, plus the new markets that environmental controls introduces, is expected to permit considerable growth in our analytical instruments business in the years ahead.

Integrated helicopter avionics system

Advances in automatic controls are fundamental to advances in aviation, and the Integrated Helicopter Avionics System is the most advanced control and guidance system ever conceived for helicopters and other vertical take-off and landing aircraft. The system is being developed by Teledyne for the United States Navy and is intended for initial installation in the Marine Corps’ CH-53A assault helicopter. The purpose of the system is to enable helicopters to ferry troops and supplies from Navy transports quickly and precisely to selected inland locations under adverse conditions.

IHAS incorporates an automatic flight control system which gives the helicopters all-weather blind flying capability. The forward looking radar has an automatic terrain following mode which permits the helicopters to fly close to the ground to avoid detection, and still clear hills or other obstacles in their path. To increase chances of getting past enemy defenses, the helicopters are automatically grouped into formation by the IHAS in-flight station keeping system.

All of the flight control and navigation functions of IHAS are coordinated in the computer central complex which serves as a nerve center for the system. This unusual special purpose incremental digital computer is especially suited for real time control applications. Its design is based on the use of integrated microelectronic circuits designed and built by Teledyne’s Amelco Semiconductor division.

The cockpit layout for helicopters using IHAS is displayed here. The television type displays show the pilot the terrain ahead, and permit him to evaluate the computer generated information utilized in route selection, decision making and automatic flight control.

Precision hydraulic fittings are extensively used in modern high performance aircraft. Actuation of the control surfaces which guide the airplane through the air, lowering and retraction of the landing gear and many other critical functions are carried out hydraulically. Through our Linair Engineering division Teledyne is a major supplier of these essential components to the aircraft industry. To meet the many applications we manufacture fittings in a variety of exotic metal and alloy versions, as well as in a variety of shapes and sizes. Teledyne fittings are in use on virtually every commercial and military aircraft in the air today.

In addition to supply the growing aviation markets, Teledyne also makes high reliability precision hydraulic fittings for a number of critical industrial applications. Our industrial fittings are used principally in large automatic machinery installations, where failure would be very costly. Of particular significance for industrial applications is our newly developed patented flareless fitting, which introduces a novel method of sealing without overlap. Among the advantages of the advanced new fittings are easier installation and less costly maintenance. Marketing of the new flareless components will begin this year.

Directly above and forward of the cockpit crew in the new DC-9 jet aircraft is a vital advisory and warning display developed and manufactured by Teledyne’s Radar Relay division. The display is the visual output of a sensor system that provides advisory information about all important conditions in the aircraft, from unlatched doors to oil pressure. Teledyne’s visual warning displays are soon to be joined by our newly developed audio warning systems. Miniaturized, even to the tape that tells the pilot about the exact condition, audio warning systems represent a new concept and figure prominently in future plans for both fixed wing aircraft and helicopters. As the useful of aircraft in the economy increases, our contributions to the development of visual and audio warning systems will continue to grow in importance.

Large reels spin off as much as four miles of film insulated conductive copper winding in the production of magnetic coils for each distribution transformer manufactured by our Crittenden Transformer division. Teledyne produces such transformers to meet the demands of a variety of large power users such as factories, office and residential buildings, shopping centers and institutions. Users purchase power at high voltage from the utility companies and employ transformers to reduce the voltage for internal distribution. For example, a recently shipped 2,000 kva transformer accepts 12,800 volts from the utility, steps it down to 480 volts and distributes the power economically throughout the plant for standard voltage applications.

Welding power supplies, variable voltage transformers, line voltage regulators, custom direct current power supplies and related specialized equipment is manufactured by Teledyne’s Glenn Pacific division for the industrial market. The division originated the constant potential rectifier type power supply which, through its characteristics of producing uniform welds, is especially suitable for automatic welding applications. Such power supplies are now used throughout industry, especially in the field of metallic inert gas, submerged arc and core wire welding. Our line voltage regulators are finding increasing demand as the need for stable line voltage becomes urgent, particularly where computers or other specialized instrumentation are used. The special power supply markets using DC arc furnace, anodizing, plasma arc and general purpose DC supplies present additional growth opportunities.

Many thousands of variable voltage transformers we build are used in industrial drive systems where exacting speed control and regulation is essential. The expanding market for industrial drive systems is also the primary target for Pacific Industrial Controls division, with its diversified line of motor speed control equipment. Development of new proprietary products in this area was continued throughout the year.

Mobile jet engine starters and power converters contribute to around the clock servicing and maintenance of aircraft, primary functions of equipment manufactured by our Sprague Engineering and Inet Power divisions. Our jet engine starters are used by airframe manufacturers and commercial airlines, as well as in military applications. Sprague continues to be a prime supplier of both fixed station and mobile equipment for testing hydraulic systems and supplying hydraulic power, and is now producing large numbers of flight line test equipments, both electric and turbine powered, for the Air Force. In addition to standard production models, the division develops and produces specialized items for critical applications. For instance, a unit for purging huge Saturn rocket engines to reduce explosion possibilities during fueling was recently completed.

During the year Inet increased its sale of equipment for the generation, conversion, monitoring and control of precise power. A 250 kw rotary type, battery backed, no-break power system—the largest rotary system of this type to date—was delivered in 1965. The no-break system switches to battery operation when a power failure occurs, eliminating the few moments of down-time between power failure and transfer to field generators. Inet is also developing the largest static no-break system—a 200 kw system for the Navy’s Bureau of Yards and Docks. This program will significantly enhance Inet’s position as a leader in the field. Another development of interest last year was a complex static high powered transient synthesizer. The system is the first of its kind ever developed and is capable of generating programmable voltage and frequency transients with power peaks in excess of four megawatts for synthesizing a broad range of power supply parameters. The synthesizer is invaluable in the testing of communications and other electronic equipment.

Microeye, the smallest television camera-transmitter ever produced, was conceived by Teledyne for the National Aeronautics and Space Administration. The two and one-half pound package consisting of complete camera and transmitter is a television station in itself, and can use either radio frequency or coaxial cable to relay the video signal. Microeye employs microcircuitry throughout and produces images with a resolution equivalent to that of commercial television. The development is being sponsored by NASA to test the feasibility of using the small, compact design for high quality, reliable space applications. The potential for Microeye in a multitude of military and commercial applications includes those where television was previously impossible due to size or power limitations.

1966 Teledyne Shareholder Letter

In 1966, approximately 55 per cent of Teledyne’s business was in electronics and control; some 23 per cent was in geophysics and oceanography; and about 22 per cent was in materials technology. What is the relationship between these fields, and how has it come about that Teledyne is in them?

In its early days Teledyne was known for the preeminence of its people in systems technology—the technique of combining a multiplicity of diverse elements to perform a complete mission. During the six years since those days Teledyne’s strength in systems technology has steadily increased. In 1965, this capability was given special recognition with the award to Teledyne of a contract to develop the IHAS system—a synthesis of complex electronic and control equipments designed to permit unrestricted use of helicopters under conditions of darkness or poor visibility.

During the past three years, while our capabilities in systems technology have been maturing, our work in geophysics and oceanography have been beginning. The seismic reflection technology which our crews employ for geophysical exploration is based on the science of information theory—an electronic system technology initially developed by the communications engineer, and applied, among other things, to the development of radar. Viewed in this perspective, our geophysical work becomes simply an application of our know-how in electronic systems technology. Our primary tool for geophysical exploration is in fact a kind of earth radar, which depends for its operation upon the transmission into the earth of specially shaped acoustic or seismic waves, and the subsequent reception and analysis of reflections of these waves to determine as much as possible about the distribution of material beneath the surface. In radar, of course, the energy is electromagnetic radar than seismic, and the transmission is into space, but the system theory and analysis is the same. The direction that Teledyne’s developmental activities in geophysics is taking is thus being guided by our knowledge and understanding of electronic systems technology, leading to our current emphasis on the rapid introduction of a variety of new seismic transmitters tailored specifically to the individual job, and to a rapid conversion from analogue to digital processing of the resulting acquired seismic data.

Underlying and decisively shaping future developments in both electronics and control and in geophysics and oceanography is Teledyne’s third field of interest—materials technology. During the first half of the twentieth century the great pioneers in systems technology dominated and directed the course of progress in electronics and control. Among their achievements we may note the following: In communications technology, the development of methods for transmitting many messages simultaneously over a single wire or radio link by frequency or time division multiplex; and the development of modulation and communication theory permitting the reliable transmission of information over noisy channels by the use of various forms of amplitude, frequency, and pulse modulation. In control technology, the development of stability criteria, and the elaboration of practical methods for ensuring the stability of feedback control systems. In computer technology, the development of the analogue differential analyzer; the application of Boolean algebra to the synthesis of switching circuits; and the elaboration of the stored program digital computer.

But with the invention of the transistor in 1948 a shift in emphasis began. By then the theories referred to above had been relatively well developed. Since then, and particularly today, the continuous revolution taking place in electronics and control is not primarily due to any fundamental new developments in systems theory. Instead, the electronic revolution is receiving its essential thrust from advances in material technology. These advances are radically altering the form of the hardware that performs a given job, leaving the processing performed and the function performed unchanged. Witness MEMA, Teledyne’s microelectronic modular assembly, about the size of a postage stamp, performing a task that ten years ago would have required at least a trunk full of circuitry.

From these considerations it is apparent that Teledyne’s interest in materials technology is basic and essential to continued progress in our chosen field. Our first commitment to materials technology was made six years ago, with the initiation of our work on semiconductor devices. Since then we have greatly increased our involvement with materials, and have broadened our capability to include a wide range of exotic alloys. In the same way that advances in electronics are now governed by advances in materials, in future times progress in control mechanisms will come to be limited by progress in the materials out of which the mechanisms are made. Teledyne’s strong position in materials technology will help us to lead this progress. It gives us too a potential for additional growth in other natural and related ways from our present base.

Thus we see that our three fields of interest—electronics, geophysics, and materials technology—are interrelated and dependent upon each other in an essential way; and that Teledyne’s capability in each of them enhances and lends additional significance to our activities in the others.

Electronics and control

This transcript of Inventing on Principle by Bret Victor was originally posted on GitHub1.

So, unlike the previous session, I don't have any prizes to give out. I'm just going to tell you how to live your life.

This talk is actually about a way of living your life that most people don't talk about. As you approach your career, you'll hear a lot about following your passion, or doing something you love. I'm going to talk about something kind of different. I'm going to talk about following a principle — finding a guiding principle for your work, something you believe is important and necessary and right, and using that to guide what you do.

There are three parts to this talk. I'm first going to talk about the principle that guides a lot of my work, and try to give you a taste of what comes out of that. And I'm going to talk about some other people that have lived this way; what their principles are, what they believe in. But these are all just examples, to help you think about what you believe in, and how you want to live your life.

So to begin with me: Ideas are very important to me. I think that bringing ideas into the world is one of the most important things that people do. And I think that great ideas, in the form of great art, stories, inventions, scientific theories, these things take on lives of their own, which give meaning to our lives as people. So, I think a lot about how people create ideas and how ideas grow. And in particular, what sorts of tools create a healthy environment for ideas to grow. Now, I've spent a lot of time over the years making creative tools, using creative tools, thinking about them a lot, and here's something I've come to believe: Creators need an immediate connection to what they're creating. That's my principle. Creators need an immediate connection to what they create. And what I mean by that is when you're making something, if you make a change, or you make a decision, you need to see the effect of that immediately. There can't be a delay, and there can't be anything hidden. Creators have to be able to see what they're doing. Now I'm going to show a series of cases where I noticed that that principle was violated and I'll show you what I did about that, and then I'm going to talk about the larger context in which I do this work.

So, to begin with, let's think about coding. Here's how coding works: you type a bunch of code into a text editor, kind of imagining in your head what each line of code is going to do. And then you compile and run, and something comes out. So in this case, this is just JavaScript2, drawing to a Canvas3, and it draws this little scene with a tree. But if there's anything wrong with the scene, or if I go and make changes, or if I have further ideas, I have to go back to the code, and I edit the code, compile and run, see what it looks like. Anything wrong, I go back to the code. Most of my time is spent working in the code, working in a text editor blindly, without an immediate connection to this thing, which is what I'm actually trying to make.

So I feel that this goes against this principle that I have, that creators need an immediate connection to what they're making, so I have tried to come up with a coding environment that would be more in line with this principle I have. So what I have here is I've got this picture on the side, and the code on the side, and this part draws the sky and this draws the mountains and this draws the tree, and when I make any change to the code, the picture changes immediately. So the code and the picture are always in sync; there is no compile and run. I just change things in the code and I see things change in the picture. And now that we have this immediate connection between the code and the picture, we can start thinking about ways of changing the code other than typing. So for example if this number here is the length of the branches. If I want to control that number, I just point my mouse to it, hold down the control key, and I can dial it up and down. So I can see what it looks like for big branches or small branches, and I can kind of converge on what feels right to me artistically. And this works great on any part of the code, I just point to it, and dial it up and down. Some of these numbers here, I know what they do but it's kind of surprising to see them do it. And other ones are just completely surprising.

So down here, I've got this for loop where I'm counting to sixteen, I'm putting sixteen little pink blossoms on every branch. And I can turn it down for less blossoms or turn it up for more. But, look at what I'm doing here: I'm just kind of moving the number up and down around twenty or so: and it has this really interesting shimmering effect; it kind of looks as if the wind was blowing through the tree. And the first time I saw this I immediately started thinking about how I could use this effect for an animation. How would I ever have discovered that if I had to compile and run between every change? So much of art, so much of creation is discovery, and you can't discover anything if you can't see what you're doing.

So I've shown you adjusting the code, let's add some code. I'd like to put a sun up here in the sky, so I'll go to the end of the drawSky function, and I'll want to fill a circle, so I start typing context.fillCircle, and as soon as I start typing, I get this autocomplete list of the different fill methods. So these are the things I can type there: fillCircle, fillRect, fillText. And as I move up and down this autocomplete list, I immediately see what each of them is doing. So, I don't have to imagine what it would do from the method name. I don't have to look at the documentation, I just see it, immediately.

So I want a circle, and I'm going to adjust the x coordinate and the y coordinate, change the radius a bit. That looks about right. Probably it should be yellow, so I'm going to set the fill style, context.fillStyle, same autocomplete as before, choose fillStyle, gives me white by default, and I can change that color code the same way I change any number, I hit the control key, and I get a color palette. So I can choose a nice yellow for my sun. Although, the white was kind of interesting, I thought. I kind of didn't expect that. But, with white, it now looks like the moon instead, right? Hey look, it's night time! So having this immediate connection allows ideas to surface and develop in ways that would be impossible before.

But there's still a problem here, I think, which is I've got this picture, and I've got this code over here and I have to maintain the mapping between the two in my head. So I've got all these lines of code, and just looking at this line I don't immediately know what it does. So here's what I can do. I can hold down the option key, and my cursor changes to a magnifying glass, and now as I roll over each line of code, it's highlighting in the picture what's being drawn in that line. So, if I want to know what's going on in this function, I just roll down the function and see what highlights. So here I've got two calls to drawMountain; I don't know which is which; well, that's that mountain, and that's that mountain. And this has to work the other way too; if I see part of the picture, I have to know what code was responsible for drawing it. So I do the same thing; I hold down the option key, and now as I move over each pixel of the picture, you'll see on the right it's jumping to the line of code that drew that pixel. So that drew the sky, and that drew the tree, and that drew the blossom. So this is really important for maintaining that mapping, but it's also really useful just for navigating around. So you know, I want to make the sun a little bit bigger; I jump there, and make it a little bigger. Or I want to bring up the tree a little bit; I jump there, and bring up the tree a little bit; I want to bring up the mountains a little bit, so I jump there, bring up the mountains a little bit; and I can make these changes as quickly as I think of them, and that is so important to the creative process. To be able to try ideas as you think of them. If there is any delay in that feedback loop, between thinking of something and seeing it, and building on it, then there is this whole world of ideas which will just never be. These are thoughts that we can't think.

Ideas are very important to me. And the thing about ideas is that ideas start small. Ideas start out tiny, weak and fragile. In order to develop and mature, ideas need an environment where the creator can nurture them. Kind of take care of them, feed them, and shape their growth. And to me, that's what the principle of immediate connection is all about. And because ideas are so precious to me, when I see this principle violated, when I see ideas stillborn or stunted because their creator couldn't see what they were doing, I feel that's wrong. And not wrong in the sense of violating some UI guideline or going against some best practice, but wrong in a deeper sense then that. And I'll come back to this, but I want to show you another example of following this principle.

So in this code here, there is no state, there is no persistent state, there is no time, there is no interactivity. And I was thinking about how we would handle those aspects of coding in a way that's in line with these principles I have: creators need immediate connection. So what I have here is a little platform game. Here is my little guy, he can run around, he can jump, he can die. And the code for him is over here. So this code makes him run around, this makes him jump, this makes him collide with things... and down here, I've got some code for this little turtle. And the turtle is not doing much right now because I haven't finished writing his code, so, I'm just going to do that right now. Say on each tick his x position plus equals his direction times the time interval, one sixtieth of a second times some speed, which, um, I dunno? Could be fast, could be slow, if it's negative, he walks backwards. And these are all ideas I can use for other enemies but I think turtles are supposed to be slow, so let's set that speed for our turtle. And up here, I've got some code that says, when my guy collides with the turtle, he gets some Y velocity, so he bounces into the air, and the turtle gets stomped. So that looks like that. And the turtle gets up after a bit.

The problem is, I don't want the player to be able to get up here yet. I want the player to bounce off the turtle, and go through this little passageway down here. And he'll have to go around and solve puzzles and whatnot, and then come back and get the star. So, the turtle is too bouncy right now. Now of course I can just turn that down on the code, and now I can try it but now it's not bouncy enough. So while it's nice that I can adjust the code while it's running, without having to stop and recompile and find my place again, I can't immediately see what I need to see, which is whether or not he can make that jump.

So here's what I'm going to do. I'm going to bounce off the turtle, and pause the game. So I pause the game, and now there's this slider up here, which lets me rewind through time. And now, I can rewind to back before I made the jump, and change the code, make it less bouncy, and now when I move it forward, it's going to simulate it, using the same input controls, the same keyboard commands recorded as before, but with the new code.

This is not good enough. I need to be able to see the changes immediately. I need to be able to see immediately whether or not my bounciness is correct. None of this stuff. And if you have a process in time, and you want to see changes immediately, you have to map time to space. So here's what I'm going to do. I'm going to bounce off my turtle, pause the game, and now hit this button here, which shows my guy's trail. So now I can see where he's been. And when I rewind, this trail in front of him is where he is going to be. This is his future. And when I change the code, I change his future. So I can find exactly the value I need, so when I hit play, he slips right in there.

So, creators need to be able to see what they're doing. If you're designing something embedded in time you need to be able to control time. You need to be able to see across time, otherwise you're designing blind.

As I was playing with this, I noticed it's fun to play with gravity. So I can make gravity a little negative and he starts to float up in the air. And I can kind of play with that and try to get him to stay there. And you could probably make an entire game around just the mechanic here, it's gravity manipulation. In fact, I bet I could fiddle with any part of this code and come up with an idea for a game. Even if I just comment out the first statement in the code, now my guy can't move left - he can only move right. Which sounds kinda silly, but Terry Cavanagh4 actually made a beautiful game around that concept called Don't Look Back5. Terry Cavanagh, he made another really wonderful game which you might have seen, called VVVVVV6, spelled as the letter v six times. And, the way that game works, is that you can't jump. Instead, you can only flip upside down, and you fall up instead of falling down. So it kinda works like this. You can walk on the ceiling or you can walk around on the ground. And so you have these levels which kinda look like this, and you kinda walk around... You have to learn how to navigate a terrain like this. And so if you had like a, something like that, you wouldn't be able to jump over it. You'd have to flip over and flip over; he got an amazing amount of gameplay out of this concept.

So again, being able to try ideas as you think of them. This example, and the last one with the tree, these are both very visual programs; we're able to see our changes just by seeing how the picture changes. So I was thinking about, how we could do more abstract coding that's more in line with this principle. How can we write a generic algorithm in such a way that we can see what we're doing. So as an example, let's take a look at binary search. Super quick refresher on how binary search works: you have an array of values that are in order, and you have a key, which is the value you're trying to locate within the array. And you keep track of two variables, which are the lower and upper bounds of where you think that value could possibly be; right now, it could be anywhere. And you look right in the middle of that range - if what you find is too small, then the key has to be after that. Look in the middle of the range, if what you find is too big, the key has to be before that. And you kinda keep subdividing your range until you narrow in on the value you're looking for. And in code, binary search looks like this. And from my perspective, you can't see anything here. You can't see anything. I see the word 'array', but I don't actually see an array. And so in order to write code like this, you have to imagine an array in your head, and you essentially have to play computer. You have to simulate in your head what each line of code would do on a computer. And to a large extent, the people that we consider to be skilled software engineers are just those people that are really good at playing computer. But if we're writing our code on a computer... why are we simulating what a computer would do in our head? Why doesn't the computer just do it... and show us?

So. Let's write binary search. Function "binary search" takes a key and an array. And then over here on this side, it's saying "Ok, it takes a key and an array, such as what? Give me an example; I need something to work with here." So, for instance, my array might be 'a', 'b', 'c', 'd', 'e', 'f'. And let's say for instance we're looking for the 'd'. So now let's start coding. The lower bound starts out as zero. Over here it says 'low equals zero', nothing amazing there. Upper bound starts out at the end of the array, so high equals array length minus one. And over here, it says 'high equals five'. So I have my abstract formula in the code. Over here, it's giving me the concrete value corresponding to these example arguments. So I don't have to maintain this picture in my head; it's just showing it to me.

So now I need the index in the middle of the array, so I'm going to take the average of those two. Mid equals low plus high over two, and... well, that's obviously not right. Two point five is not a valid array index. So I guess I need to round this off. So I'm going to add the floor function and it rounded it down to two. And I caught that bug literally the second I typed it, instead of writing the entire function in twenty unit tests. So now I get the value out of the array... and then I need to subdivide my range, which, so there's an if statement which I'll just paste in here. So in this case, the value I found is less than the key, so it's taking this first branch of the if statement. This is adjusting the lower bound. Of course if the key was smaller, then it would take this branch of the if statement and adjust the upper bound. Or, if the key was 'c', then we would've just happened to find it on the first shot, and we'd return the index.

So this is the first iteration of this algorithm. And now what we need to do, is loop. We've subdivided the array, we need to keep subdividing until we narrow in on what we're looking for. So, we need to loop; I will just loop. While 1, do all this. And now what we have are three columns corresponding to three iterations of this loop. So this first column here is exactly what you saw before. Low and high span the entire array, we found a 'c', it was too low, so we adjusted our lower bound, and loop up to here. Second iteration, bounds are tighter; we found an 'e'. Adjust the upper bound. Third iteration, loop up to here; low and high are the same. We've narrowed it down to a single candidate - it's indeed the key we're looking for, and we returned this index. So there's nothing hidden here; you see exactly what the algorithm is doing at every point. And I can go up to here and try different keys, so I can see how the algorithm behaves for these different input arguments.

And by looking across this data, I can develop an intuition for how this algorithm works. So I'm trying different keys here, and say I try looking for a 'g'. And this looks a little different. It's not actually returning. And the reason for this is, I'm looking for a key which is not actually in the array. And the only way of breaking out of this loop, is by finding the key. So it's kinda stuck here looping forever. So we can take a look at this and see what went wrong, where's the algorithm going off the rails. These first few iterations look fine, but this iteration looks weird, because low is greater than high. Our range is completely collapsed. So if we get to this point, then we know the key can't be found. So I see this faulty condition, and I say, "Oh, that's not right; low has to be less than or equal to high." Okay, well, I'll just put that over as the condition of my while statement. Low, less than equal to high, and then that would break out of the loop, and I would return some signal to say that it couldn't be found. So here we have three iterations of the loop, couldn't be found, we return a not found value. So that's what it might be like to write an algorithm without a blindfold on.

So I've got this principle, again, that creators need to be able to see what they're doing. They need this immediate connection with that they're making. And I've tried to show this principle through three coding examples, but that's just because this is a software engineering conference, I thought I was supposed to talk about programming. But to me, this principle has nothing to do with programming in particular. It has to do with any type of creation. So I'd like to show you a couple more demos, just to show you the breadth of what I have in mind here.

So, to begin with, let's take a look at a different branch of engineering. So here I have an electronic circuit that I drew. I'm not quite done drawing it, so let me finish up there. And we'll put 2. And now we have a working circuit. I mean I assume it's a working circuit. I don't actually see anything working here. So this is exactly the same as writing code, that we work in a static representation. But what we actually care about is the data. The values of the variables, so we can't see that here. Now in a circuit, the variables are the voltages on these different wires. So each of these wires has a voltage that's changing over time, and we have to be able to see that. Now, if I was building this circuit on a lab bench, building it physically, I could at least take an oscilloscope and kinda poke around and see what's going on in the different wires, what's going on here, or here. So at the very least, I should be able to do that. So what I have here, is a plot of the voltage on this wire over time. You can see it's high, it's low, high and low, so this is clearly oscillating. If I built this physically, also I would be able to see the circuit doing something. In this case I have these two LED's up here. These are LED's, little lights, presumably they're there for a reason. I can hit Play, and watch it simulate out in real time. So now you can see what the circuit is doing.

In order to design a circuit like this, you have to understand the voltage on every wire. You have to understand how all the voltages are changing throughout the entire circuit. And just like coding, either the environment shows that to you, or you simulate it in your head. And I have better things to do with my head than simulating what electrons are doing. So what I'm gonna do, I'm gonna spread these out a bit. So same circuit, spread out a little bit, and I'm going to add the voltage at every node. So now you can see every voltage throughout the circuit. And I can even hit Play and watch it all kind of simulate out in real time.

Although, what I prefer to do, is just move my mouse over it, and I can kind of look in areas that are interesting to me and see what the values are. I can compare any two nodes. So if you look at say the node over here, while I mouse over this one, you see the shadow of the one I'm mousing over is overlaid on that. The shadow of the one I'm mousing over is actually overlaid on all of them. And so I can compare any two nodes just by mousing over one of them and looking at the other one.

And again, I can immediately see results of my changes. So I've got this 70k resistor here. I want to change its value, I just click and drag it, and now I see the waveforms change immediately. And you'll notice that when I click and drag, it leaves behind the shadow of the waveform before I started dragging, so I can compare. I can immediately see the results of my changes.

Two golden rules of information design: Show the data, show comparisons. That's all I'm doing here. But even this isn't quite good enough. What we're seeing here are the voltages, but in electronics there are actually two data types. There is voltage and there is current. And what we're not seeing is the current, flowing through each of these components. And in order to design a circuit, you need to understand both the voltage and the current. You need to understand the interplay between the two. That's what analog design is.

So what I'm gonna do is spread these out a little bit more. And now I'm gonna replace each of these components with a plot of the current going throw it over time. So each of these blue boxes represents a component. And you can see which component it is, because it has a little badge in the corner, a little icon, but now you can see everything that's going on in the circuit. You can see how the current changes, you can see how the voltage and the current changes. There's nothing hidden, there's nothing to simulate in your head.

So what we have here is a different way of representing the circuit. Just in general, you could draw any circuit with these blocks and instead of being made out of little squiggly symbols, it's made out of data. And I think it's important to ask: Why do we have these squiggly symbols in the first place? Why do they exist? They exist because they're easy to draw with pencil on paper. This is not paper. So when you have a new medium, you have to rethink these things. You have to think how can this new medium allow us to have a more immediate connection to what we're making. How can this new medium allow us to work in such a way that we can see what we're doing.

And it's really the same situation with programming. Our current conception of what a computer program is — a list of textual definitions that you hand to a compiler7 — that's derived straight from Fortran8 and ALGOL9 in the late '50's. Those languages were designed for punchcards. So you'd type your program on a stack of cards, and hand them to the computer operator (it's the guy in the bottom picture), and you would come back later. So there was no such thing as interactivity back then. And that assumption is baked into our current notions of what programming is.

C10 was designed for teletypes. That's Ken Thompson11 and Dennis Ritchie12 up there. Ritchie made C. And there are no video displays in this picture. Ritchie is basically typing on a fancy typewriter that types back to him. Any time you use a console or a terminal window, you're emulating a teletype. And even today, people still think of a REPL or an interactive top-level as being interactive programming. Because that's the best thing you could do on a teletype.

So I have one more demo I want to show because I want to emphasize that this principle, immediate connection, is not even about engineering, it's about any type of creation. So I want to move to a different field entirely, so let's think about animation.

So I've got this painting here, of a tree and a leaf on it and I want to make a little video with the leaf kinda drifting down the tree. And the normal way of doing this in a conventional animation package like Flash13, is through keyframes. So you basically say where you want the leaf to be at different points in time, and then you hit Play and see what it looks like. So, I'm gonna say: ok, at frame 20, I'm gonna create a keyframe and the leaf should be there. And at frame 40, create a keyframe and the leaf should be there, and I'm just totally guessing here. I can not see the motion. I can not feel the timing, I'm just throwing things in time and space.

So I've got this leaf at different points in time, and I'm gonna add a tween, which tells Flash to connect the dots. And then I'm gonna hit Play and see what it looks like. And it looks ridiculous, it looks like billiard balls bouncing back and forth.

And the thing is I kind of know what I want, right? It's a leaf. I want a leaf drifting down from a tree. And I can even perform that with my hand: leaf drifting down from a tree. But Flash doesn't know how to listen to my hand. But maybe there's a new medium that does know something about listening to my hand.

So what I'm gonna show you here is a little app I made for performing animation. And we're not really set up to do a live demo off the iPad so I'm just gonna play you a video of me making a video. The way this scene is going to play out is the leaf is gonna kind of drift down from the tree, and the scene is gonna pan over and the rabbit is gonna do something. And two things: one, this is going to move pretty quickly, and second, I'm going to be using both hands at almost all times. So I've got these different layers, the background, the mid-ground and the foreground. I'm choosing which layer to move using my left thumb. I'm gonna move my leaf to its position. I'm gonna move my bunny off stage and start time rolling. Now I'm gonna perform the leaf drifting down from the tree. Run it back, check out how that looked. The motion looks pretty good but the leaf kinda needs to rock back and forth. So I'm gonna pull out a rotation controller, run it back, find where the leaf is about to break off, and record the rotation. And I added a little flip there just because it felt right at the moment. It wasn't even planned. Stop, because I want to pan over. So I'm gonna drag a whole bunch of layers at once, I grab all the layers into a list, I turn down the sensitivity of the background layers so they'll move slower for a kind of parallax effect. I only want to move horizontally so I pull out a horizontal dragger and check out how it looks. I don't quite like the parallax so I adjust the sensitivities just a little bit, try it out again, I like that better, so I get ready to go, I run it back to the beginning so I can get back into the rhythm of the piece. The leaf hits, I wait a beat, and I start panning. And I don't know how many frames I waited, I don't know how long it was, I went when it felt right.

So I panned over this winter scene and kind of slowed down to a stop. And then I run it back, because I want to do something with my bunny. Throw away these tools because I'm done with them. And wait until I think my bunny should move and he hops away. And I have got a few different poses for my bunny. So I pull those out. And then I find the point where the bunny is about to take off the ground. Which is right there. I switch his pose and I kind of toggle between the poses as he hops away. And then I run it back because I wanna check out how it looked and I'm just gonna bring that up full screen for you. This is the piece.

So I made that in 2 minutes, performing with my hands like a musical instrument. Very immediate connection between me and what I was trying to make.

One of the inspirations for this tool was an animation that I tried to make several years ago. Not that one but it also began with a leaf drifting down from a tree. And I spent all day in Flash trying to keyframe that leaf. Couldn't do it. And so that was the end of that. I still have my storyboards. Sometimes I play the music I wrote for the piece. But the piece itself is locked in my head. And so I always think about the millions of pieces that are locked in millions of heads. And not just animation, and not just art, but all kinds of ideas. All kinds of ideas including critically important ideas, world-changing inventions, life-saving scientific discoveries. These are all ideas that must be grown. And without an environment in which they can grow, or their creator can nurture them with this immediate connection, many of these ideas will not emerge. Or they'll emerge stunted.

So I have this principle that creators need an immediate connection and all of those demos that I just showed you simply came from me looking around, noticing places where this principle was violated, and trying to fix that. It's really all I did. I just followed this guiding principle and it guided me to what I had to do.

But I haven't said much about the most important part of the story, which is why. Why I have this principle. Why I do this.

When I see a violation of this principle, I don't think of that as an opportunity. When I see creators constrained by their tools, their ideas compromised, I don't say: Oh good, an opportunity to make a product. An opportunity to start a business. Or an opportunity to do research or contribute to a field. I'm not excited by finding a problem to solve. I'm not in this for the joy of making things. Ideas are very precious to me. And when I see ideas dying, it hurts. I see a tragedy. To me it feels like a moral wrong, it feels like an injustice. And if I think there's anything I can do about it, I feel it's my responsibility to do so. Not opportunity, but responsibility.

Now this is just my thing. I'm not asking you to believe in this the way I believe I do. My point here is that these words that I'm using: Injustice, Responsibility, Moral wrong, these aren't the words we normally hear in a technical field. We do hear these words associated with social causes. So things like censorship, gender discrimination, environmental destruction. We all recognize these things as moral wrongs. Most of us wouldn't see a civil rights violation and think "Oh good, an opportunity." I hope not.

Instead, we've been very fortunate to have people throughout history who recognized these social wrongs and saw it as their responsibility to address them. And so there's this activist lifestyle where these persons dedicate themselves to fighting for a cause that they believe in. And the purpose of this talk is to tell you that this activist lifestyle is not just for social activism. As a technologist, you can recognize a wrong in the world. You can have a vision of what a better world could be. And you can dedicate yourself to fighting for a principle. Social activists typically fight by organizing but you can fight by inventing.

So now I'd like to tell you about a few other people who have lived this way, starting with Larry Tesler14. Larry has done a lot of wonderful things in his life, but the work I'm gonna tell you about he did in the mid '70s at Xerox PARC15. And at the time, there really wasn't such a thing as personal computers. The notion of personal computing was very young and Larry and his colleagues at PARC felt that they had transformative potential, that personal computing could change how people thought and lived. And I think all of us in this room would agree that they turned out to be right about that.

But at the time, software interfaces were designed around modes. So, in a text editor for instance, you couldn't just type and have words appear on the screen like on a typewriter. You would be in command mode and if you wanted to insert text you'd have to press I to go into insert mode then Escape back out to command mode or maybe you'd hit A to go into append mode. Or if you wanted to move text around you'd hit M go to the Move mode and then you'd have to select and you'd be in the mode to select and move things around. And Larry would watch people using the computer — they actually pioneered the concept of software user studies, another thing that he did — but he would watch people using the software and he found that many people even after training and weeks of use, many people would not become comfortable with the computer.

And he believed that it was these modes that were to blame. That the complexity of modes was a kind of barrier that many people couldn't cross. And so this kind of represented a threat to this dream of what personal computing could be. So Larry made it his personal mission to eliminate modes from software. And he formed a principle: No person should be trapped in a mode. His slogan that he would go around saying was 'Don't mode me in' and he had it printed on his t-shirt. And this principle informed everything that he did. He thought about it with all the work that he did. And eventually he came up with a text editor called Gypsy16, which essentially introduced text editing as we know today. There was an insertion point. And when you typed, words appeared on the screen. To select text, he invented modeless selection with click and drag. So you just click and drag over the text you want to select like using a highlighter — one of the first uses of drag. To move text around, he invented these commands that he called Cut, Copy, Paste. You select and cut. Later on you paste in whenever you're ready. You're never trapped in a mode, you never have to switch between modes. When you hit W on the keyboard you get W on the screen. Always.

And he would watch people using his software and he found that someone who had never seen a computer before — which was most people back then — could be up and running in like half an hour. So this was clearly a transformative change in enabling people to connect with computers. And his ideas about modelessness spread to the rest of the desktop interface which was then being invented at PARC at the same time. And today they're so ingrained in the computing experience that today we kind of take them for granted.

Now I said that Larry made the elimination of modes his personal mission. That's actually his words, and if you think he's exaggerating, here's Larry's license plate for the last 30 years. Nowadays of course Larry has a website, at nomodes.com and he's on twitter: @nomodes. And so like I said, Larry has done a lot of amazing work in his career but his self identity is clearly associated with this cause.

And so I'd like to ask: What exactly did Larry do? Like how could we best describe what Larry did? A typical biography might say Larry Tesler invented Cut, Copy, Paste. Which is true, but I think that's really misleading, because this invention was very different than say, Thomas Edison17 inventing the phonograph. Edison basically just stumbled over the technology for audio recording and he built it out as a novelty. And he came up with this list of possible applications for his technology but he didn't have any cultural intent. Whereas what Larry did was entirely a reaction to a particular cultural context.

So another thing that you might hear is that Larry Tesler solved the problem of modeless text manipulation. Solved the problem. And obviously that's true, he worked on this problem for a long time, eventually solved it. But I think that's really misleading too, because this problem that he solved only existed in his own head. Nobody else saw this as a problem. For everybody else modes were just how computers worked. There wasn't anything wrong with them any more than we think there's something wrong with having two arms. It just kind of was a fact of life.

So the first thing that Larry did was that he recognized a wrong that had been unacknowledged in the culture. And the thing is, that's how many great social changes began as well. So a 150 years ago, Elizabeth Cady Stanton18 had to stand up and say: women should vote. And everybody else said 'That's crazy, what are you talking about'. Today, we recognize gender discrimination as a wrong. Back then, it was part of society, it was invisible. She had to recognize it, and she had to fight it. And to me, that's a much closer model to what Larry did than the Thomas Edison model of inventing a bunch of random technology so he could patent it.

Now to be clear I'm not making any judgements about the relative importance or the impact of these two people, I'm just talking about their motivations and their approach. Both of them recognized a cultural wrong, they envisioned a world without that wrong and they dedicated themselves to fighting for a principle. She fought by organizing, he fought by inventing.

And many other seminal figures in computing had similar motivations. So certainly Doug Engelbart19. Doug Engelbart basically invented interactive computing. The concept of putting information on a screen. Navigating through it. Looking at information in different ways. Pointing at things and manipulating them. He came up with all this at a time when real-time interaction with a computer was just almost unheard of. Today he is best known as the inventor of the mouse, but what he really invented is this entirely new way of working with knowledge. His explicit goal from the beginning was to enable mankind to solve the world's urgent problems. And his vision, he had this vision of what he called knowledge workers using complex powerful information tools to harness their collective intelligence. And he only got into computers because he had a hunch that these new things called computer things could help him realize that vision. Everything that he did was almost single-mindedly driven by pursuing this vision.

Here's Alan Kay20. Alan Kay ran the lab at Xerox PARC where we got the desktop interface, so things like windows and icons, command menus. He also invented object-oriented programming and lots of other things. His goal, and I quote, was to 'amplify human reach, and bring new ways of thinking to a faltering civilization that desperately needed it.' Isn't that great? His approach was through children. He believed that if children became fluent in thinking in the medium of the computer, meaning if programming was a form of basic literacy like reading and writing, then they'd become adults with new forms of critical thought, and new ways of understanding the world. And we'd have this more enlightened society, similar to the difference that literacy brought to society. And everything that he did, everything he invented, came out of pursuing this vision, this goal with children. And following principles that he adopted from Piaget21 and Montessori22, Jerome Bruner23, these people who would study how children think.

And the figure probably most widely associated with software activism is Richard Stallman24. Stallman started the GNU project25 which today makes up a big chunk of any Linux26 system. He also started the Free Software Foundation27, wrote GCC28, GPL29, many, many other things. His principle is that software must be free, as in freedom, and he has very precise meaning associated with that statement. He's always been very clear that software freedom is a matter of moral right and wrong. And he has taken a particularly uncompromising approach in his own life to that.

All of these tremendously influential people dedicated their lives to fighting for a particular ideal with a very clear sense of right and wrong. Often really fighting against an authority or mainstream that did not recognize their wrong as being wrong. And today, the world is still very far from any of their ideals and so they still see a world in crisis and they keep fighting. They're always fighting.

Now I'm not saying you have to live this way. I'm not saying that you should live this way. What I'm saying is that you can. That this lifestyle is an option that's available to you. And it's not one you're gonna hear about much. Your career counselor is not going to come back to you and say you should start a personal crusade. In a social field, they might, but not in technology. Instead the world will try to make you define yourself by a skill.

That's why you have a major in college. That's why you have a job title. You are a software engineer. And you'll probably specialize to be a database engineer or a front-end engineer, and you'll be given front-ends and asked to engineer them. And that could be worthwhile and valuable, and if you want to spend your life pursuing excellence and practicing a skill, you can do that. That is the path of a craftsman. That is the most common path. The only other path you really hear about much is the path of the problem solver. So I see entrepreneurship and academic research as kind of two sides of that coin. There is the field. There's the set of problems in that field, or needs in the market. You go in, you choose one, you work it, you make your contribution there. Maybe later on, you choose another problem, you work it, you make your contribution there. Again, that could be worthwhile and valuable and if that's what you want to do, then you can take that path.

But I don't see Larry Tesler on either of those paths. I wouldn't say that he was contributing to the field of user experience design because there was no such thing. He didn't choose some open problem to solve, he came up with some problem that only existed in his own head, and no one else even recognized. And certainly he did not define himself by his craft, he defined himself by his cause. By the principle he fought to uphold. And I'm sure if you look at Wikipedia it will say that he's a computer scientist or a user experience something but to me that's like saying Elizabeth Cady Stanton was a community organizer. No, Elizabeth Cady Stanton established the principle of women's suffrage. That's who she was. That was the identity she chose and Larry Tesler established the principle of modelessness. He had this vision, and he brought the world to that vision.

So, you can choose this life. Or maybe it will end up choosing you. It might not happen right away. It can take time to find a principle because finding a principle is essentially a form of self-discovery, that you're trying to figure out what your life is supposed to be about. What you want to stand for as a person. Took me like a decade. Ten years before any real understanding of my principles solidified. That was my twenties. When I was young I felt I had to live this way but I would get little glimmers of what mattered to me but no big picture. It was really unclear. This was very distressing for me. What I had to do was just do a lot of things. Make many things, make many types of things. Study many things, experience many, many things. And use all these experiences as a way of analyzing myself. Taking all these experiences and saying 'Does this resonate with me?'. Does this repel me? Do I not care? Building up this corpus of experiences that I felt very strongly about for some reason and trying to make sense of it. Trying to figure out why. What is this secret ingredient to all these experiences that I'm reacting to so strongly.

Now I think everyone's different. And all the guys I talked about they have their own origin stories which you can read about. I will just say that confining yourself to practicing a single skill can make it difficult to get that broad range of experience which seems to be so valuable for this sort of work.

And finally, if you choose to follow a principle, a principle can't just be any old thing you believe in. You'll hear a lot of people say they that want to make software easier to use. Or they want to delight their users. Or they want to make things simple. That's a really big one right now. Everyone wants to make things simple. And those are nice thoughts and maybe kind of give you a direction to go in but they're too vague to be directly actionable. Larry Tesler likes simplicity. But his principle was this specific nugget of insight: No person should be trapped in a mode. And that is a powerful principle because it gave him a new way of seeing the world. It divided the world in right and wrong in a fairly objective way. So, he could look at somebody selecting text and ask: Is this person in a mode? Yes or no? If yes, he had to do something about that. And likewise, I believe that creators need powerful tools. It's a nice thought, but it doesn't really get me anywhere. My principle is that creators need this immediate connection. So I can watch you changing a line of code and I can ask: Did you immediately see the effect of that change? Yes or no? If no, I got to do something about that.

And again, all those demos that I showed you came out of me doing that, of me following this principle and letting it lead me to exactly what I needed to do. So if you're guiding a principle and bodies of specific insight, it will guide you. And you will always know if what you're doing is right.

There are many ways to live your life. That's maybe the most important thing you can realize in your life, is that every aspect of your life is a choice. But there are default choices. You can choose to sleepwalk through your life and accept the path that's been laid out for you. You can choose to accept the world as it is. But you don't have to. If there is something in the world you feel is a wrong and you have a vision for what a better world could be, you can find your guiding principle. And you can fight for a cause. So after this talk, I'd like you to take a little time and think about what matters to you. What you believe in. And what you might fight for.

Thank you.

Truly great companies, for the most part, have always been great. And the vast majority of good companies never become great.

We don't have more great companies precisely because most companies become quite good. Good is the enemy of great.

But the transition from good to great does happen, but how does it happen?

To find out, Jim and his team identified companies that made the leap from good results to great results for at least fifteen years. They compared these companies to a group of comparison companies that failed to make or sustain the leap, then they looked for the distinguishing factors.

The good-to-great companies averaged cumulative stock returns 6.9 times the general stock market in the fifteen years following their transition points. To put that into perspective, every dollar invested into a fund of the good-to-great companies in 1965 that was held until January 1, 2000 would have multiplied 471 times, compared to a 56 fold increase in the market.

These are remarkable numbers, made more remarkable when you consider that many of these companies had been utterly unremarkable before their transition.

The search

Jim and his team looked for companies with fifteen-year cumulative stock returns at or below the general stock market, punctuated by a transition point, then cumulative returns at least three times the market over the next fifteen years.

Fifteen years was long enough to avoid one-hit wonders and lucky breaks, and is longer than the tenure of most chief executive officers. This helps separate great companies from companies that happened to have a single great leader. And three times the market exceeded the performance of most widely recognized great companies between 1985 and 2000, think companies like Coca-Cola, Intel, Wal-Mart, and Disney.

Good-to-great companies also had to demonstrate the good-to-great pattern independent of their industry, if the industry showed the same pattern, they dropped the company.

The final set of companies is as follows:

Compared to what?

Next they contrasted the good-to-great companies with comparison companies to determine what differentiated them.

There were two sets of comparison companies. The first were direct comparisons. Companies in the same industry with the same opportunities and similar resources at the time of transition that showed no change in performance.

The second were unsustained comparisons. Companies that made a short-term shift from good to great but failed to maintain the trajectory.

This produced a list of twenty-eight companies: eleven good-to-great companies, eleven direct comparisons, and six unsustained comparisons.

Inside the black box

Then they collected all articles published on the twenty-eight companies, dating back fifty years or more, coded all the material into categories, interviewed most of the good-to-great executives who held key positions during the transition, and performed a wide range of qualitative and quantitative analysis.

The core of their method was to contrast the good-to-great companies to the comparisons. Think of their research as akin to looking inside a black box, each step shed more light on the good-to-great process.

What they didn't find turned out to be some of the most important findings.

Larger-than life leaders who came from the outside were negatively correlated with taking a company from good to great. Ten of the 11 good-to-great CEOs came from within. Executive compensation didn't matter.

Strategy did not separate the good-to-great companies from the comparisons. Both sets had well-defined strategies.

The good-to-great companies didn't focus on what to do, rather what not to do and what to stop doing. Technology accelerated but did not cause transformations. Mergers and acquisitions played no role in transforming from good-to-great.

Good-to-great companies paid little attention to managing change, motivating people, or creating alignment.

There was no single event that signified the transformation from good-to-great. Some even reported being unaware of the magnitude of the transformation at the time, and many good-to-great companies were not in great industries and some were in terrible industries.

Chaos to concept

Now let's go over the framework of concepts and preview what is to come. Think of the transformation as a process of buildup followed by breakthrough, broken into three stages: disciplined people, disciplined thought, and disciplined action.

Each stage has two key concepts and wrapping the entire framework is the flywheel which captures the entire process of going from good to great. Good-to-great transformations never happened in one fell swoop. Rather it resembles pushing a giant heavy flywheel in one direction, building momentum, until a point of breakthrough.

• Level 5 Leadership: Good-to-great leaders tend to be self-effacing, quiet, reserved, and even shy. A paradoxical blend of personal humility and professional will.

• First Who...Then What: Get the right people on the bus, the wrong people off the bus, the right people in the right seats then figure out where to go.

• Confront the Brutal Facts (Yet Never Lose Faith): Every good-to-great company embraced the Stockdale Paradox. Unwavering faith that you can and will prevail regardless of difficulty while having the disciple to confront reality, whatever it may be.

• The Hedgehog Concept (Simplicity within the Three Circles): Going from good to great means transcending the curse of competence. If you can't be the best in the world at it, it can't form the basis of a great company.

• A Culture of Discipline: When you have disciplined people, you don't need hierarchy, bureaucracy, or excessive controls.

• Technology Accelerators: Technology alone cannot transform a company from good to great, but good-to-great companies were pioneers in the application of carefully selected technologies.

Each of these concepts showed up as a change variable in 100 percent of the good-to-great companies and in less than 30 percent of the comparison companies.

The timeless "physics" of good to great

Good to great isn't about specific companies or timeframes, it's about identifying principles–the enduring physics of great organizations–that remain true and relevant no matter how the world changes around us. Specific applications will change but the principles will endure.

It's about how to take a good organization and turn it into one that procedures sustained great results, using whatever definition of results best applies to your organization.

Good being the enemy of great is not a business problem. It's a human problem. If we can understand what turns good into great, we have something of value for any organization.

Level 5 leadership

Every good-to-great company had a Level 5 leader. Furthermore, the absence of Level 5 leadership was a consistent pattern in the comparison companies.

Level 5 refers to a five-level hierarchy of executive capabilities:

• Level 1: A highly capable individual who makes productive contributions through talent, knowledge, skills, and good work habits

• Level 2: A contributing team member who contributes individual capabilities to the achievement of group objectives and works effectively with others in a group setting

• Level 3: A competent manager who organizes people and resources toward the effective and efficient pursuit of predetermined objectives

• Level 4: An effective leader catalyzes commitment to and vigorous pursuit of a clear and compelling vision, stimulating higher performance standards

• Level 5: A level 5 executive builds enduring greatness through a paradoxical blend of personal humility and professional will.

They're incredibly ambitious, but the ambition is first and foremost for the cause, not themselves.

Contrary to conventional wisdom, Level 5 leaders are often self-effacing, quiet, reversed, even shy. More likely to motivate with inspired standards than inspiring personality.

One such leader was Darwin E. Smith who became chief executive of Kimberley-Clark, a paper company whose stock had fallen 36 percent behind the general market over the previous 20 years.

Smith, the company's mild-manner lawyer, wasn't sure the board made the right decision which was reinforced when a director reminded him that he lacked qualifications.

But over the next twenty years, Smith transformed Kimberley-Clark, generating cumulative stock returns 4.1 times the general market.

Yet few people know anything about Darwin Smith, and he probably would have liked that.

In retirement, Smith reflected on his exceptional performance, stating "I never stopped trying to become qualified for the job."

Not what is expected

Jim gave his research teams explicit instructions to downplay the role of leadership to avoid the simplistic "credit or blame the leader" thinking that is prevalent today.

To use analogy, the "Leadership is the answer to everything" perspective is equivalent to "God is the answer to everything" that held back scientific understanding of the physical world. If we attribute everything to "Leadership", we prevent ourselves from gaining a deeper understanding of what makes great companies tick.

The comparison companies also had leaders, so how could leadership be a differentiator?

Because a Level 5 leader was at the helm of every good-to-great company during their transition.

Furthermore, the absence of Level 5 leadership was a consistent pattern in the comparison companies.

Given Level 5 leadership goes against the conventional wisdom of needing larger-than-life leaders to transform companies, it's important to note Level 5 is an empirical, not ideological, finding.

Humility + will = level 5

Level 5 leaders are a study of duality: modest and willful, humble and fearless.

Consider Colman Mockler, CEO of Gillette from 1975 to 1991. During Mockler's tenure, Gillette faced three attacks that threatened to destroy their chance at greatness.

Two attacks came as hostile takeover bids from Revlon, led by Ronald Perelman. The third came from Coniston Partners, who bought 5.9 percent of Gillette stock and initiated a proxy battle to seize control of the company in hopes of selling it to the highest bidder.

Had Mockler accepted Perelman's offer, shareholders would have reaped an instant 44 percent gain on their stock.

Most executives would have cashed in, but Mockler and Gillette's executives instead chose to fight reaching out to thousands of individual investors to win the battle.

Mockler and his team were staking the company's future on huge investments into Sensor and Mach3, had the takeover been successful these projects would have almost certainly been eliminated.

These investments promised significant future profits that weren't reflected in the stock price because of their secrecy.

The board and Mockler believed the future value of the shares far exceeded the current price, even with the price premium offered, and they were right. To sell out would have made short-term investors happy but would have been terrible for long-term shareholders.

If Mockler had accepted the 44 percent price premium offered on October 31, 1986 and then people invested the full amount in the general stock market for ten years to the end of 1996, they would have come out three times worse off than a shareholder who stayed with Mockler and Gillette.

Setting up successors for success

Level 5 leaders want to see the company even more successful in the next generation. They're comfortable with the idea that most people won't even know that the roots of the success trace back to their efforts.

In contrast, the comparison leaders were more concerned with their own reputation. Over three quarters of the comparison company leaders failed to set up their successors for success, chose a weak successor, or both.

While some of the comparisons took their companies from good to great for a brief moment, when they left, the company went from great to irrelevant.

After all, what better testament to your own personal greatness than a place that falls over after you leave?

Compelling modesty

In contrast to the I-centric style of the comparison leaders, good-to-great leaders didn't talk about themselves. They talked about the company and the contributions other executives made, avoiding discussions about their own accomplishments.

It wasn't false modesty. Those who worked or wrote about good-to-great leaders frequently used words like quiet, humble, modest, reserved, shy, gracious, mild-mannered, self-effacing, and understated.

The eleven good-to-great CEOs were some of the most remarkable CEOs of the century yet almost no one ever remarked about them. They never wanted to be larger-than-life heroes. They were ordinary people quietly producing extraordinary results.

In contrast, two thirds of the comparison companies had an egocentric leader that contributed to the demise or continued mediocrity of the company.

This pattern was particularly strong in the unsustained comparisons–cases where the company would show a leap in performance under a talented leader, only to decline in later years.

Unwavering resolve to do what must be done

Level 5 leadership is not just about humility and modesty. It's equally about the resolve to do what needs to be done.

They're fanatically driven to produce results, displaying workman like diligence–more plow horse than show horse.

George Cain, who became CEO of Abbott Laboratories after working there for eighteen years, is one example. He couldn't stand mediocrity and was intolerant of anyone who would accept the idea that good was good enough.

He set out to destroy one of the key causes of Abbott's mediocrity: nepotism. Systematically rebuilding the board and executive team with the best people he could find.

If you didn't have the capability to be the best executive in the industry, you were out.

This is something you'd expect from an outsider brought in to turn the company around, but Cain was an insider. In fact, he was a family member, the son of a previous Abbott president.

Even if they got fired, family members had to be pleased with the performance of their stock. Cain set in motion a profitable growth machine that beat the market by 4.5 times between 1974 and 2000.

This reflects a more systematic finding of Good to Great. Evidence did not support the idea that you need an outside leader to go from good to great. In fact, high profile outsiders were negatively correlated with a sustained transformation.

Ten out of the eleven good-to-great CEOs came from inside the company, three of them by family inheritance. The comparison companies turned to outsiders six times often — yet they failed to produce sustained great results.

The window and the mirror

The good-to-great executives talked a lot about luck. Some even flat-out refused to take credit for their company's success, attributing it to the good fortune of having great colleagues, successors, and predecessors.

At first, Jim and his team were puzzled by this emphasis on luck. They found no evidence that the good-to-great companies were blessed with more good luck than the comparison companies.

Then they noticed a contrasting pattern in the comparison executives. They credited substantial blame to bad luck.

The emphasis on luck turns out to be part of a pattern called the window and the mirror.

Level 5 leaders look out the window and attribute success to factors other than themselves. When things go poorly, they look in the mirror and blame themselves, taking full responsibility.

The comparison CEOs did the opposite, looking in the mirror to take credit and out the window to assign blame.

Cultivating level 5 leadership

Jim's hypothesis is there are two types of people: those who have the seed of Level 5, and those who don't. For those who don't, work will always be about what they get, not what they build.

The great irony is personal ambition often drives people toward positions of power but stands at odds with the humility required for Level 5 leadership.

Most people have the seed of Level 5. The problem is not a dearth of potential Level 5 leaders. Rather it's the fact that people operate under the belief that you need to hire larger-than-life leaders to make an organization great, this is why Level 5 leaders rarely appear as CEOs.

To find Level 5 leaders, look for places where extraordinary results exist but no individual steps forward to claim credit.

For your own development, begin practicing the other good-to-great concepts. This is about what Level 5s are, the remaining concepts cover what they do.

First Who...Then What

You'd think the first step in taking a company from good to great would be to set a new direction and get people committed to it.

The opposite is closer to the truth. The transformation begins by getting the right people on the bus, the wrong people off the bus, then you figure out where to go.

The key point is not just the idea of getting the right people on the team. It's about answering the who question before making the what decisions. First who, then what.

Starting with who, rather than what, means you can more easily adapt to change.

If people join because of where you're going, you're going to have problems if you need to change course. But if people are on the bus because of who else is there, it's going to be easier to change direction.

When you have the right people, problems with motivation and management largely go away. The right people have an inner drive to produce the best results and to be part of something great.

If you have the wrong people, it doesn't matter whether you find the right direction; you still won't have a great company.

Great vision, without great people, is irrelevant.

First who...then what is a simple idea to grasp, but hard to do in practice–and most don't do it well.

Not a genius with a thousand helpers

In contrast to the good-to-great companies, the comparison companies followed a genius with a thousand helpers model. They were platforms for an extraordinary individual to set their vision and enlisted a crew of highly capable helpers to make the vision real.

Having a single driving force for a company's success is a great asset–as long as the genius sticks around. The issue is these geniuses seldom build great management teams, for the simple reason that they didn't need, or often, want one.

When the genius leaves, the helpers are often lost. Or, worse, they try to mimic their predecessor with bold, visionary moves–trying to act like a genius, without being one.

It's who you pay, not how you pay them

You'd expect that changes in incentives would be correlated with making the leap from good to great, but compensation played no part.

Not that executive compensation is irrelevant. You need to be reasonable, but once you've got something that makes sense, compensation falls away as a distinguishing factor.

Compensation isn't about motivating the right behaviors from the wrong people, it's how you get and keep the right people in the first place.

It's who you compensate, not how you compensate them that's important.

If you have the right people, they'll do everything they can to build a great company. Not because of what they'll get, but because they can't settle for less.

The right people do the right things and deliver the best results they're capable of, regardless of incentive.

And whether someone is the right person has more to do with character traits and innate capabilities than specific knowledge, background, or skill. Not that these are unimportant, but these traits are teachable while character traits, work ethic, dedication to commitments, and values are more ingrained.

Rigorous, not ruthless

Good-to-great leaders are rigorous, but not ruthless, in people decisions.

Ruthless means using layoffs and restructuring, especially during difficult times, as the primary strategy for improving performance. The evidence suggests such tactics are contrary to producing sustained great results.

Six of the eleven good-to-great companies recorded zero layoffs from ten years before their breakthrough date all the way through 1998, and four others reported only one or two layoffs. In contrast, layoffs were used five times more frequently in the comparison companies.

Some had an almost chronic addiction to layoffs and restructuring.

Rigorous means consistently applying exacting standards at all times and at all levels, especially in upper management. Rigor applies first at the top, focused on those who hold the burden of responsibility.

How to be rigorous

There are three disciplines you need to follow to be rigorous:

1. When in doubt, don't hire–keep looking

2. When you know you need to make a people change, act

3. Put your best people on your biggest opportunities, not your biggest problems

1. When in doubt, don't hire–keep looking

One of the immutable laws of management physics is Packard's Law: No business can grow revenues consistently faster than its ability to get enough of the right people to implement that growth and still become a great company.

If your revenue growth consistently outpaces your ability to get and retain the right people, you can’t build a great company.

Those who build great companies understand the ultimate throttle on growth isn't the market, technology, competition, or products. It's people.

Limit growth based on your ability to attract enough of the right people.

2. When you know you need to make a people change, act

Good-to-great companies showed a bipolar pattern at the top management level: People either stayed for a long time or got off in a hurry.

Letting the wrong people hang around is unfair to the right people as they'll have to compensate for their inadequacies. Worse, it drives away the best people.

Waiting too long before acting is equally unfair to those who need to get off the bus. Every minute you allow a person to continue working when you know that they won't make it, you steal a portion of their life, time they could spend finding a place where they can flourish.

If you're honest with yourself, the reason you wait too long often has less to do with concern for the person and more to do with convenience.

When you know you need to make a people change, act. But be sure you don't have someone in the wrong seat.

Instead of firing people who aren't performing, try to move them once, twice, or even three times to positions where you think they might blossom.

How do you know when you have the wrong person? Two questions can help. First, would you hire them again? Second, if the person told you they were leaving for a new opportunity, are you disappointed or relieved?

3. Put your best people on your biggest opportunities, not your biggest problems

Put your best people on your best opportunities, not your biggest problems. Managing problems can only make you good, building your opportunities is the only way to become great.

If you sell off your problems, don't sell off your best people.

If you create a place where the best people always have a seat, they're more likely to support changes in direction. Always keep your best people, even if they have little or no experience with the new direction.

The only way this happens is if you have a Level 5 atmosphere at the top executive level. Not every executive on the team will become a fully evolved Level 5 leader, but each member must transform their personal ambition into ambition for the company.

It's somewhat paradoxical, but you need executives who can, as Jeff Bezos says, disagree and commit.

First who, great companies, and a great life

Adherence to the idea of first who might be the closest link between a great company and a great life. When you have the right people around you and put the right people in the right seats, you don't have to be at work at all hours of the day and night.

And when you are at work, it'll be with people you like.

Members of the good-to-great teams tended to become and remain friends for life. In many cases, they're still in close contact with each other years, or even decades, after working together.

They enjoyed each other's company and many characterized their years on the good-to-great teams as the high points of their lives.

No matter what you achieve, if you don't spend the majority of your time with people you love and respect, you can't possibly have a great life. If you spend the vast majority of your time with the right people, then you'll almost certainly have a great life, no matter where the bus goes.

Confront The Brutal Facts (Yet Never Lose Faith)

The good-to-great companies faced as much adversity as the comparisons. It was how they responded that was different.

They confronted the facts head-on, focused on the few things that would have the greatest impact, and emerged from adversity stronger.

When you start with an honest effort to find the truth, the right decisions often become self-evident. Not always, but often.

Even if all decisions don’t become self-evident, one thing is certain: It’s impossible to make good decisions without confronting the facts.

This is why every company that went from good to great created a culture where people, and ultimately, the truth could be heard.

They operated on both sides of the Stockdale Paradox: retaining absolute faith that they could and would prevail in the end, regardless of difficulties, while confronting the facts of their reality.

Facts are better than dreams

Good-to-great companies didn’t have a perfect record, but on the whole they made more good decisions than bad ones, and on the really big choices, they were remarkably right.

This begs the question:

“Were Jim and his team merely studying a set of companies that stumbled upon the right decisions? Or was there something distinctive about their process that increased the likelihood of being right?”

It turns out there was. The good-to-great companies confronted the facts of their reality, and the comparison companies generally did not.

This leads to the next finding. Strong, charismatic leaders can all too easily become the de facto reality driving a company. The comparison companies often had leaders who led with such force that people worried more about what they thought than reality and what it could do to the company.

This is why less charismatic leaders often produce better long-term results. If you have a strong, charismatic personality, understand that charisma can be as much of a liability as an asset.

Your strength of personality can cause people to hide the facts from you.

You can overcome this, but it does require conscious effort.

A climate where the truth is heard

You might be wondering, “How do you motivate people to confront the facts? Doesn’t motivation flow from a compelling vision?”

The answer is surprisingly no.

Not because vision is unimportant, but because expending energy trying to motivate people is a waste of time.

If you have the right people, they will be self-motivated. The key is to not de-motivate them.

And one of the primary ways to do this is by ignoring reality.

Yes, leadership involves vision. But it’s equally about creating an environment where the truth can be heard and acted on.

This involves four practices:

1. Leading with questions, not answers

2. Engaging in dialogue, not coercion

3. Conducting autopsies, without blame

4. Building red flag mechanisms

1. Leading with questions, not answers

Resist the urge to provide the answer. Instead, operate in a Socratic style and use questions to gain understanding.

Keep asking questions until you have a clear picture of reality and its implications.

Never use questions as a form of manipulation or a way to blame or put down others.

Spend the bulk of your time just trying to understand.

Start with questions like:

• What’s on your mind?

• Can you tell me about that?

• Can you help me understand?

• What should we be worried about?

Open-ended questions tend to bubble current realities to the surface.

2. Engaging with dialogue, not coercion

The good-to-great leaders often played the role of a Socratic moderator in a series of raging debates. People would scream, argue, and fight, then emerge with a conclusion.

In the words of Jeff Bezos, they would disagree and commit.

Nearly all the good-to-great executives described climates where the company’s strategy evolved through arguments and fights. Phrases like “loud debate,” “heated discussions,” and “healthy conflict” were common.

Discussions weren’t a sham process to let people have their say so they would buy in to a predetermined decision. Discussions were how they found the right answers.

3. Conducting autopsies, without blame

Good-to-great companies spent hundreds, if not thousands, of people hours analyzing mistakes, but no one pointed fingers.

Unless it was the CEO standing in front of the mirror blaming themselves.

This is contrarian in an era where leaders go to great lengths to preserve their image and step forth to claim credit for how visionary they are while blaming others when their decisions go awry.

By conducting autopsies without blame, you create a climate where the truth can be heard. And if you have the right people, you shouldn’t need to assign blame.

Focus on understanding and learning from mistakes.

4. Building red flag mechanisms

More information is supposed to be an advantage. But if you look across the rise and fall of organizations, you rarely find companies stumbling because they lack information.

There was no evidence that the good-to-great companies had more, or better, information than the comparisons. Both sets had virtually identical access to information.

The key is not better information, but making sure that information cannot be ignored. One particularly powerful way to accomplish this is through red flag mechanisms. One example is to tell customers to not pay you if they aren’t satisfied for any reason without the need to return the product.

This works because it creates hard-to-ignore feedback.

The Stockdale paradox

The good-to-great companies faced just as much adversity as the comparison companies, but how they responded was different.

On the one hand, they stoically accept reality. On the other, they maintained an unwavering faith in the endgame, and a commitment to prevail as a great company despite reality.

Jim and his team came to call this duality the Stockdale Paradox named after Admiral Jim Stockdale, the highest-ranking US military officer in the “Hanoi Hilton” prisoner-of-war camp between 1965 and 1973.

Stockdale lived out the war without any rights, no set release date, and no certainty as to whether he would even survive to see his family again.

When Jim asked Stockdale how he dealt with it, he said “I never doubted not only that I would get out, but also that I would prevail in the end and turn the experience into the defining event of my life, which, in retrospect, I would not trade.”

This sounds optimistic, but when Stockdale was asked who didn’t make it out, he said:

“Oh, that’s easy, the optimists. They were the ones who said, ‘We’re going to be out by Christmas.’ And Christmas would come, and Christmas would go. Then they’d say, ‘We’re going to be out by Easter.’ And Easter would come, and Easter would go. And then Thanksgiving, and then it would be Christmas again. And they died of a broken heart.”

What separates people, and companies, is not the presence or absence of difficulty, but how they deal with it.

If you adopt this dual pattern, you dramatically increase the odds of making a series of good decisions, and ultimately discovering a simple, yet insightful, concept for making the really big choices.

The Hedgehog Concept (Simplicity within the Three Circles)

Isaiah Berlin divided the world into hedgehogs and foxes, based on the ancient Greek parable: “The fox knows many things, but the hedgehog knows one big thing.”

Hedgehogs simplify the world into a single organizing idea, basic principal, or concept that unifies or guides everything.

On the other hand, foxes pursue many things at the same time and see the world in all its complexity.

To be clear, hedgehogs aren’t stupid. Quite the opposite. They understand the essence of insight is simplicity. They take a simple idea and take it seriously, and this is what allows them to see through complexity and discern underlying patterns.

Hedgehogs see what’s essential and ignore the rest.

Those who build good-to-great companies are more like hedgehogs. They know one big thing and stick to it. The comparison companies tended to be foxes, crafty, cunning, and knowledgeable, but never gaining the clarifying advantage of a Hedgehog Concept.

The three circles

To go from good to great requires a deep understanding of your Hedgehog Concept, then using that concept to guide all your efforts.

This comes at the intersection of three circles:

1. What you can (and can not) be the best at: This goes beyond core competence. Competence doesn’t necessarily mean you can be the best in the world at it. Inversely, what you can be the best at might not even be something you’re currently doing.

2. What drives your economic engine: You need insight into how to generate sustained and robust cash flow and profitability. In particular, a single denominator–profit per x–that has the greatest impact on your economics.

3. What you are deeply passionate about: Focus on activities that ignite your passion.

Understanding what you can (and can not) be the best at

The most important thing to understand is what you can be the best at, and equally, what you cannot be the best at. Not what you want to be the best at.

This distinction is crucial.

Your Hedgehog Concept isn’t a goal, strategy, or intention. It’s an understanding.

If you cannot be the best in the world at your core business, then your core business cannot be the basis of your Hedgehog Concept.

Best in the world is a much more severe standard of excellence than a core competence. You might be competent, but not necessarily have the capacity to truly be the best in the world.

Inversely, there may be activities you could be the best in the world at but have no current competence.

Understand what you can truly be the best at, then stick to it.

Doing what you are good at will only make you good. Focus solely on what you can do better than anyone else.

Insight into your economic engine – what is your denominator?

You don’t need to be in a great industry to build a fantastic economic engine. The good-to-great companies often produced specular returns in unspectacular industries.

You need a deep understanding of the key drivers of your economic engine and you need to build a system in accordance with this understanding.

You need a single economic denominator.

To find it, ask yourself:

“If I could pick one and only one ratio–profit per x–to systematically increase over time, what x would have the greatest and most sustainable impact on my economic engine?”

This question leads to profound insight into the inner workings of your economics.

Push for a single denominator. It’ll produce better insight than letting yourself off the hook with three or four.

Understanding your passion

It may seem odd to talk about something as fuzzy as passion as an integral part of your strategy. But passion is a key part of the Hedgehog Concept.

It’s not about saying, “Let’s get passionate about what we do.” Rather, you should only do things that you are or can get passionate about.

This doesn’t mean that you have to be passionate about the mechanics of the business. The passion circle can equally focus on what the company stands for.

Remember, you can’t manufacture passion or “motivate” people to feel passionate. You can only discover what ignites your passion and the passions of those around you.

The triumph of understanding over bravado

Set your goals and strategies based on the intersection of the three circles, not bravado.

Setting goals based on bravado is most obvious in the mindless pursuit of growth.

“Growth” is not a Hedgehog Concept.

However, if you have the right Hedgehog Concept and make decisions consistent with it, you create momentum and your main problem becomes not how to grow, but how to not grow too fast.

Finding your Hedgehog Concept is a turning point for going from good to great.

But it’s a terrible mistake to thoughtlessly try to jump right to a Hedgehog Concept. Discovering your Hedgehog Concept is an inherently iterative process, not an event.

On average, it takes about four years to clarify.

The council

A useful device for iterating toward your Hedgehog Concept is the Council.

The Council is a group of five to twelve people who ask questions guided by the three circles:

• Do we know what we can be the best in the world at?

• Do we understand the drivers of our economic engine, including our economic denominator?

• Do we understand what best ignites our passion?

Every member must be able to argue and debate the answers, while maintaining the respect of every other Council member, without exception.

The goal isn’t to seek consensus. The responsibility for the final decision remains with the leading executive.

But once the decision is made, the Council should perform an autopsy and analysis of the results and repeat the process periodically, as much as once a week, or as little as once a quarter.

Members should come from a range of perspectives, but each must have deep knowledge about some aspect of the organization or environment.

This typically includes key members of management, but is ’t limited to management, and being an executive does ’t automatically make you a member.

Your hedgehog concept

Most companies aren’t the best in the world at anything, and show no prospect of becoming so. But every company has a Hedgehog Concept to discover.

There’s something you can become the best at, and you can find it. But only if you confront the facts of what you cannot be the best at, no matter how awful your starting point.

As you search for your own, keep in mind that when you finally grasp your Hedgehog Concept, it’ll be free of bravado and feel more like a recognition of a fact.

Once you successfully apply these ideas, don’t stop. The only way to remain great is to keep applying the fundamental principles that made you great.

Culture of discipline

Few startups become great companies because as scale introduces complexity, it becomes easier and easier to trip over your own success.

Too many people. Too many customers. Too many products. Too many problems.

In response, someone (usually from the board) says, “It’s time to grow up.” In come the MBAs and seasoned executives. And with them come processes, procedures, and checklists.

The self-organizing, egalitarian environment is replaced with strict hierarchy, chains of command, and complex reporting structures. These professionals managers do rein in the mess, but they all too often kill the thing that got them there in the first place – entrepreneurial spirit. And members of the founding team start to grumble:

“This isn’t fun anymore.” “I can’t get anything done.” “I’m always stuck in stupid meetings.”

Once your most innovative people leave, your once exciting startup transforms into any old company, and mediocrity creeps in. The issue is companies build bureaucracy to compensate for incompetence and lack of discipline – two things that largely don’t exist when you have the right people.

And if you lean too heavily on managing the small percentage of people who shouldn’t be on the bus, you’ll drive away the right people, which increases the percentage of the wrong people and the need for hierarchy, and so on.

The best companies avoid bureaucracy and instead create a culture of discipline that enables creativity and entrepreneurship, instead of stifling it.

More precisely, they:

• Give people freedom and responsibility, within a framework.

• Hire self-disciplined people who go to extreme lengths to fulfill their responsibilities.

• Don’t confuse a culture of discipline with a tyrannical disciplinarian.

• Adhere to the Hedgehog Concept and focus on the intersection of the three circles.

• Create a “stop doing list” and unplug from the extraneous.

When you combine a culture of discipline with an ethic of entrepreneurship, you get superior performance and sustained results.

Freedom (and responsibility) with in a framework

Much of the answer to what it takes to go from good to great lies in the discipline to do whatever it takes to become the best in the world in a carefully selected arena, and then to seek continual improvement from there.

It’s simple, but not easy.

Everyone wants to be the best, but most people lack the discipline to figure out with egoless clarity what they can be the best at and the will to turn that potential into reality.

The good-to-great companies built consistent systems with clear constraints, but also provided people with freedom and responsibility within those systems. The only way this works if you have people who believe in the system and will do whatever it takes to make it work.

But how do you get to a place of such conviction?

It starts with disciplined people, but don’t conflate this with disciplining people into the right behavior. You need the right people on the bus.

Then you need those people to engage in disciplined thought. They need to confront the brutal facts and persist until you find your Hedgehog Concept.

Only once you’ve found your Hedgehog Concept can you focus on disciplined action.

The order here is essential. You can’t jump to the end.

Disciplined action without disciplined people is impossible to sustain, and discipline alone doesn’t guarantee results. There are plenty of companies with tremendous discipline who march right into disaster.

Disciplined people must engage in disciplined thought, and only take action within a consistent system designed around their Hedgehog Concept.

A culture, not a tyrant

Every unsustained comparison followed the same pattern: a spectacular rise under a tyrannical disciplinarian, followed by an equally spectacular fall when the disciplinarian stepped away.

The leaders of the unsustained comparisons personally disciplined their organization through sheer force of personality, acting as the company’s primary quality control mechanism.

In contrast, the Level 5 leaders at the good-to-great companies built enduring cultures of discipline that outlasted them.

Fanatical adherence to the Hedgehog Concept

The good-to-great companies at their best followed one rule: “Anything that does not fit with our Hedgehog Concept, we will not do.” They did not launch unrelated businesses, they did not make unrelated acquisitions, nor did they go after unrelated “once-in-a-lifetime opportunities”.

If it didn’t fit, they didn’t do it.

This is because great companies are more likely to die of indigestion than starvation. The challenge is opportunity selection, not creation.

In contrast, the lack of discipline to stay within the three circles was a key factor in the demise of nearly all the comparison companies. Every comparison either lacked the discipline to understand its three circles or to stay within them.

Few companies have the discipline to discover their Hedgehog Concept, and fewer still have the discipline to build consistently within it.

Start a “stop doing” list

Most of us lead busy lives with expanding to-do lists, trying to build momentum by doing more, but it rarely works. Those who built the good-to-great companies made as much, if not more, use of stop doing lists.

They had a remarkable knack for removing the extraneous and they institutionalized the practice through budgeting. Most people think the purpose of budgeting is to decide how much to apportion to each activity or to manage costs.

This is wrong. The budget process is not for figuring out how much each activity gets, rather it’s about determining which activities best support the Hedgehog Concept and which should be eliminated entirely.

Budgeting should be used to decide which areas should be fully funded, and which should not be funded at all. The good-to-great CEOs didn’t reallocate resources, they eliminated everything that didn’t align with their Hedgehog Concept.

It takes remarkable courage to channel all your resources into a few areas, but they rarely hedged their bets. After all, the most effective investment strategy is a highly concentrated strategy, if you are right.

This sounds facetious, but it’s the approach the good-to-great companies took. Once they found their Hedgehog Concept, they invested fully in things that fit squarely in their three circles, and got rid of everything else.

Technology accelerators

In every good-to-great transition, there was technological sophistication. However, their transitions never began with the adoption of new technology. At the same time, they all became pioneers in the application of the technologies that they adopted.

The most important thing to understand is technology is only ever an accelerant, not a creator, of momentum. You could give the exact same technology to the comparison companies and they would have failed to produce the same results as the good-to-great companies.

Technology and the hedgehog concept

You can’t make good use of technology without knowing which technologies are relevant to you. The question you need to answer when adopting any technology is: “Does this technology fit with my Hedgehog Concept?”

If it does, you need to become a pioneer in the application of that technology. If it doesn’t, you need to ask yourself whether you need it at all. If you do, then aim for parity. If you don’t, ignore it completely.

Deciding on which technology to adopt is another example of good-to-great companies remaining disciplined. Their relationship to technology is no different than their relationship to any other decision: Disciplined people engage in disciplined thought, and take disciplined action.

If something new doesn’t fit within your three circles, ignore the hype. If it does, become fanatical about its adoption. Technology without a clear Hedgehog Concept, and without the discipline to stay within the three circles, cannot make a company great.

The technology trap

The idea that technology is the principal cause of decline of companies is not supported by research. Companies cannot remain laggards and hope to be great, but technology alone isn’t the primary cause of greatness or decline.

In fact, 80 percent of the good-to-great executives who were interviewed didn’t mention technology as one of the top five factors in their transition. And in the cases where they did mention it, it had a median ranking of fourth. Only two executives of eighty-four interviewed ranked technology as their primary factor.

If technology is so important, why was it so deemphasized? Not because they ignored technology. They all leveraged technology and many received extensive media coverage about their pioneering use of it.

It’s because mediocre results stem from management failure, not technological failure. Remember, the adoption, or lack thereof, technology is an accelerator, not a cause. A failure to adopt technology is simply that principle operating in reverse.

Technology and the fear of being left behind

Those who built the good-to-great companies weren’t motivated by FOMO or their competitors. They were motivated by creative urge and an inner compulsion for excellence for its own sake.

“No technology, no matter how amazing—not computers, not telecommunications, not robotics, not the Internet—can by itself ignite a shift from good to great. No technology can make you Level 5. No technology can turn the wrong people into the right people. No technology can instill the discipline to confront brutal facts of reality, nor can it instill unwavering faith. No technology can supplant the need for deep understanding of the three circles and the translation of that understanding into a simple Hedgehog Concept. No technology can create a culture of discipline. No technology can instill the simple inner belief that leaving unrealized potential on the table—letting something remain good when it can become great—is a secular sin.”

The flywheel and the doom loop

Picture a heavy flywheel—a metal disk mounted horizontally on an axle, 30 feet in diameter, 2 feet thick, and weighing about 5,000 pounds. Your task is to get the flywheel spinning as fast and long as possible.

Pushing hard, you get the flywheel to inch forward, almost imperceptibly. You keep pushing and pushing, after hours of effort, you get one complete turn. You keep pushing, and the flywheel moves faster and with continued effort, you get a second rotation. Three turns, four, five, six, the flywheel speeds up, seven, eight, nine, ten.

Momentum is building. Eleven, twelve, each turn moving faster than the last.

Then, at some point—breakthrough! Momentum turns in your favor, hurling the flywheel forward, turn after turn. Its own weight now works for you. You’re pushing no harder than the first rotation, but the flywheel goes faster and faster.

Each turn of the flywheel builds upon work done earlier, compounding your effort.

Now suppose someone asks you, “What was the one thing that caused your flywheel to spin so fast?”

You can’t answer. It’s a nonsensical question. It wasn’t the first, second, or fifth push. It was all of them added together in an accumulation of effort in a consistent direction. No single effort, no matter how large, outweighs the cumulative effect.

Build up and breakthrough

The flywheel captures what it feels like to be inside a company going from good to great. No matter how dramatic the end result, good-to-great transitions never happen in one fell swoop. There is no single action, no grand program, no killer innovation, no lucky break, no revolution.

Good to great is a cumulative process—step by step, action by action, choice by choice, turn by turn of the flywheel—that adds up to sustained and spectacular results.

But if you read media accounts of each company, you may draw a different conclusion. Often, the media doesn’t cover a company until its flywheel is already spinning. This skews our perception of how transformation occurs, making it seems like they jumped from nothing to breakthrough overnight.

It’s often as if the company hadn’t even existed prior to its discovery, despite the remarkable process made by the team in the decade leading up to the breakthrough point. You may be thinking, “Of course we should expect that. Companies get more coverage as they become more successful. What’s so insightful about that?”

Don’t allow the way transitions look from the outside to drive your perception of what they feel like for those on the inside. From the outside, they look dramatic. From the inside, they feel like a cumulative process.

The good-to-great executives couldn’t pinpoint a single key event that exemplified the transition. Frequently, they chafed against the idea of allocating points and prioritizing factors. It was a whole bunch of interlocking pieces built upon themselves.

The transition wasn’t like night and day. It was gradual and wasn’t entirely clear until years into it. The good-to-great companies didn’t even name their transformations. There was no launch event, no tag line, no programmatic feel whatsoever.

Some executives said that they weren’t even aware that a major transformation was under way until they were well into it. It was more obvious to them after the fact. There was no miracle moment. It was a quiet, deliberate process of figuring out what needed to be done to create the best results and then taking those steps, one after the other.

After pushing on their flywheel in a consistent direction over an extended period of time, they hit a point of breakthrough. Lasting transformations from good ot great follow a general pattern of buildup followed by breakthrough.

Not just a luxury of circumstance

Understand that following the buildup-breakthrough flywheel model isn’t a luxury of circumstance. People like to say,. “we’ve got constraints that prevent us from taking this longer-term approach.”

The good-to-great companies followed this model no matter how dire their short-term circumstances. This also applies to managing the short-term pressure of investors. You communicate with investors, educate them on what you are doing and where you are going. If people don’t buy into that in the short term—you have to wait and let the results of your work do the heavy lifting.

All the good-to-great companies effectively managed investors during their buildup-breakthrough years, and saw no contradiction between the two. They simply focused on accumulating results, under-promising and overdelivering. And as their flywheel built momentum—the investing community came along.

The flywheel effect

There is power in continued improvement and delivery of results. Point to tangible accomplishments—however small—and show how these steps fit into the context of an overall concept that will work.

When people see and feel the buildup of momentum, they line up with enthusiasm. This is known as the flywheel effect, and it applies not only to investors but also to employees.

What’s interesting is that none of the good-to-great leaders found that creating alignment was difficult. Clearly, all the good-to-great companies had incredible alignment, they just never spent much time thinking about it. Under the right conditions, issues with commitment, alignment, motivation, and change take care of themselves.

The good-to-great companies tended not to publicly proclaim big goals at the outset. Instead focusing their efforts on turning the flywheel, creating tangible evidence that their plans made sense. Once the flywheel was spinning, people wanted to keep it going.

When you let the flywheel do the talking, you don’t need to constantly communicate your goals. When people feel the magic of momentum and begin to see tangible results, the bulk of people line up to start pushing the flywheel harder.

The doom loop

There was a very different pattern at the comparison companies. Instead of a quiet, deliberate process of figuring out what needed to be done and simply doing it. The comparison companies frequently launched new programs, only to see the programs fail to produce sustained results.

They would push the flywheel in one direction, stop, change course, and throw it in a new direction, and then they would stop and change course again. After years of lurching back and forth, the comparison companies failed to build momentum and fell into the doom loop.

While the specific permutations of the doom loop varied from company to company, there were two prevalent patterns:

1. The misguided use of acquisitions: The drive for M&A often comes from the fact that doing deals is more exciting than doing actual work. Good-to-great companies used acquisitions after the development of their Hedgehog Concept and after the their flywheel had significant momentum. Acquisitions were accelerants, not creators of momentum. In contract, the comparison companies consistently reached for an acquisition or merger to increase growth, diversify away from troubles, or make a CEO look good.

2. The selection of leaders who undid the work of previous generations: The other common doom loop pattern is that of new leaders who stepped in, stopped a spinning flywheel, and threw it in a new direction.

The flywheel as a wraparound idea

Consistency and coherence came up over and over in the good-to-great transformations. Coherence is an idea from physics, the magnifying effect of one factor upon another. Each piece of the system reinforces another piece of the system to form an whole that is more powerful than the sum of its parts. Consistency over time, through multiple generations, gets maximum results.

It starts with Level 5 leaders. They’re less interested in flashy programs that make it look like they’re leading, and more interested in the quiet deliberate process of pushing the flywheel.

Getting the right people on the bus, the wrong people off the bus, and the right people in the right seats are all crucial steps in the early stages of buildup. Equally important is to remember the Stockdale Paradox. You might not hit breakthrough this year, but if you keep pushing, you will eventually.

This process of confronting the brutal facts helps you see the obvious, albeit difficult, steps that must be taken to turn the flywheel. Next, you must attain deep understanding of the three circles of your Hedgehog Concept and begin to push in a consistent direction, accelerating by pioneering the application of technology tied back to your three circles.

Breakthrough comes from having the discipline to make a series of good decisions consistent with your Hedgehog Concept—disciplined action, following from discipline people who exercise disciplined thought. That’s it.

Once you get there, the challenge is no longer about how to go from good to great, but how to stay consistently great.

From good to great to built to last

Built to Last, one of Jim’s books published prior to Good to Great, is based on a six-year research project conducted at Stanford Business School to answer the question, “What does it take to start and build an enduring great company from the ground up?”

Much like Good to Great, Jim and his coauthor Jerry I. Porras studied 18 enduring great companies and 18 paired comparisons. In short, they short to identify the differences between good and great companies as they endured over decades and centuries.

Early in the research, they made the decision to conduct the Good to Great research as if Built to Last didn’t exist to ensure they could identify the key factors in transforming from a good company to a great one with minimal bias from prior work.

With both studies complete, Jim had four conclusions:

1. The early leaders of the enduring great companies in Built to Last followed the good-to-great framework. The only difference was that they did so as entrepreneurs in early-stage companies rather than CEOs trying to transform established companies.

2. Good to Great is really a prequel to Built to Last. Applying its findings creates sustained great results, then applying the findings in Built to Last can help you go from great results to enduring great company.

3. Shifting from a company with sustained great results to an enduring company means discovering core values and purpose beyond making money and combining this with the dynamic preserve the core/stimulate the process.

4. There is tremendous resonance between the two studies: the ideas in each enrich and inform the other. In particular, Good to Great answers the fundamental question raised in Built to Last: “What is the difference between a good and a bad big hairy audacious goal (BHAG)?”

Core ideology: The extra dimension of enduring greatness

The “extra dimension” that helped elevate great companies to enduring great companies is a core ideology, which consists of core values and core purpose beyond making money.

Enduring great companies don’t exist merely to deliver returns to shareholders. Profits and cash flow are like blood and water to a healthy body: Absolutely essential, but not the point of life.

An important caveat is that there are no “right” core values for becoming an enduring great company. Core values are essential to enduring greatness, but it doesn’t seem to matter what the core values are. Enduring great companies preserve their core values and purpose while their business strategies and operating principles adapt. This is the magical combination of preserve the core and simulate progress.

Good BHAGs, bad BHAGs, and other conceptual links

Good to Great provides the core ideas for getting a flywheel turning from buildup through breakthrough, while Built to Last outlines the core ideas for keeping a flywheel accelerating over the long term. The four key ideas from Built to Last are:

1. Clock Building, Not Time Telling. Build an organization that can endure and adapt through generations of leaders and product life cycles; the exact opposite of being built around a single great leader or idea.

2. Genius of AND. Embrace both extremes on a number of dimensions at the same time. Instead of choosing A OR B, figure out how to have A AND B.

3. Core Ideology. Instill core values (essential and enduring tenets) and core purpose (fundamental reason for being beyond making money) as principles to guide decision making and inspire people.

4. Preserve the Core/Stimulate Progress: Preserve the core ideology as an anchor point while stimulating change, improvement, and renewal in everything else. Change practices and strategies while holding core values and purpose fixed. Set and achieve BHAGs consistent with the core ideology.

There is a clear connection between BHAGs and the three circles of the Hedgehog Concept. BHAGs are a key way to stimulate progress while preserving the core. A BHAG, Big Hairy Audacious Goal, serves as a unifying focal point of effort.

Good BHAGs are set with understanding. Bad BHAGs are set with bravado.

A good BHAG, while huge and daunting, is not a random goal. It’s a goal that makes sense within the context of the three circles. To remain great over time requires staying squarely within the three circles while being willing to change the specific manifestation of what’s inside the three circles at any given moment.

To create an enduring great company requires all the key concepts from both studies, tied together and applied consistently over time. If you ever stop doing any one of the key ideas, you will inevitably slide toward mediocrity. And remember, it’s easier to become great than remain great.

Why greatness?

First, it is no harder to build something great than to build something good. It might be rarer, but it does not require more suffering than mediocrity. If some of the comparison companies are any indication, it may involve less suffering, and perhaps less work.

The key thing to realize is that much of what you do at work is at best a waste of energy. If you organize the majority of your work time around applying these principles, and ignore or stop doing everything else, your life will be simpler with better results. If it’s no harder, the results better, and the process more fun—why wouldn’t you go for greatness?

There is a second answer to the question of why greatness, meaningful work.

If you’re doing something you care about and believe in its purpose deeply, then it’s impossible to imagine not trying to make something great, and then the question becomes not why, but how.

High Output Management by Andy Grove is widely considered to be the best book on management and by the end of this post, you’ll understand why.

I think Tobi Lutke, the co-founder and CEO of Shopify, put it best in an interview with Tim Ferriss:

Andy’s book is unapologetically almost a how-to manual, that kind of deconstructs the world of business into first principles. It’s like, “Here’s what matters. Here’s how to think about it. No one needs a degree.”

I find books easier to remember when I have context about the person who wrote them, so I think it’s fitting to start with Andy Grove himself.

Andy Grove escaped the ruins of postwar Europe and became one of the architects of Silicon Valley and the modern world as we know it.

He was born to a middle-class Hungarian Jewish Family on Sept. 2, 1936 and his early life is summarized in his memoir Swimming Across:

I had lived through a Hungarian Fascist dictatorship, German military occupation, the Nazis' “Final Solution”, the siege of Budapest by the Soviet Red Army, a period of chaotic democracy in the years immediately after the war, a variety of repressive Communist regimes, and a popular uprising that was put down at gunpoint.

Heeding the word of his aunt who had survived Auschwitz, Andy joined the flood of refugees who escaped Hungary via Austria in hopes of a better life in the West.

He arrived in the United States at 21, barely able to speak English with less than $20 to his name.

By 31, Andy had earned his Bachelor's and Ph.D. in chemical engineering, worked at Fairchild Semiconductor for four years, and written a college-level textbook on semiconductors.

Then at 32, Andy made the jump from Fairchild to Intel on the day of its incorporation. Joining its founders Robert Noyce and Gordon Moore as the company's director of engineering and employee number three.

He would go on to establish Intel's early manufacturing operations before becoming President at 43, CEO at 51, and chairman and CEO of Intel at 61.

During his tenure, Intel's revenue grew from $2,672 to $20.8 billion.

High Output Management was born out of Andy's experiences leading Intel. Nearly four decades after its initial publication, High Output Management has reached cult-like status and become a Silicon Valley staple.

CEOs of highly productive teams re-read it, top venture capitalists give copies to startup founders, and every manager worth their salt devours its content. 

This is because the book's three core concepts remain true:

1. You can apply the principles and discipline of manufacturing to management

2. Work is done by teams not individuals

3. Teams only perform well when each member is working at their best

If you remember one thing, remember this:

A manager’s output = the output of their organization + the output of the neighboring organizations under their influence

Besides Tobi and Tim, fans of High Output Management include:

• Bill Campbell: Former executive coach to Larry Page, Sergey Brin, Eric Schmidt, and Sundar Pichai at Google, Steve Jobs at Apple, Jeff Bezos at Amazon, Jack Dorsey and Dick Costolo at Twitter, and Sheryl Sandberg at Facebook;

• Drew Houston: Co-founder and CEO of Dropbox;

• Keith Rabois: Partner at Founders Fund, co-founder of Opendoor, former COO at Square, and early employee at LinkedIn and PayPal; 

• Mark Zuckerberg: Co-founder and CEO of Facebook;

• Marc Andreessen: Co-founder of a16z, co-founder of Netscape and Opsware, and co-author of Mosaic;

• Ben Horowitz: Co-founder of 16z and co-founder of Opsware;

• Brian Chesky: Co-founder and CEO of Airbnb; 

• Brian Armstrong: Co-founder and CEO of Coinbase; and

• Ev Williams: CEO and founder of Medium and co-founder of Twitter.

Learning the basics of production

Before we can apply the principles and discipline of manufacturing to management, we first need to understand the principles of production.

Imagine you're tasked with creating a cafe that sells one dish: a three-minute soft-boiled egg, buttered toast, and coffee.

You must make a profit and each item must be served simultaneously.

While customers would prefer to have meal when they sit down, that would require infinite production capacity or ready-to-serve inventory. Either option eats into your margins, making it harder to sell at a competitive price while making a profit.

What you need a system to deliver breakfasts at a scheduled time, acceptable quality level, and low cost.

In reality, customers don't mind waiting five to ten minutes for breakfast. But how do you ensure that?

Start with the limiting step and plan production around it. The limiting step is the most difficult, sensitive, expensive, or time-consuming step.

When it comes to cooking breakfast, your limiting step is the egg.

It takes the longest, costs the most, and is the crucial part. Using the egg’s cooking time as your base, give yourself enough time to toast the bread.

Use the toasting time to determine when to pour the coffee.

Finally, add the time it takes to plate up and offset each step. 

The time it takes to prepare the dish is known as the total throughput time.

Production operations

We can learn three fundamental production operations from cooking breakfast:

1. Process manufacturing: These are physical or chemical changes to the material like boiling the egg and toasting the bread;

2. Assembly: When you arrange the egg, toast, and coffee to make a meal; and

3. Testing: This is when you examine components or the final product for faults like when you check the toast is browned and the coffee is steaming.

Complications

So far we've assumed you won't have to wait for the toaster, can always boil an egg, and will never run out of coffee. In practice, any of these things could happen which would in turn makes them your limiting step. So what can you do?

• You can hire specialists: Like a chef or barista but this creates overhead which may be too expensive;

• You can invest in capital equipment: Like a new pot, toaster, or coffee maker; or

• Create an inventory: You can pre-boil eggs and pre-toast bread but this risks waste.

As the manager, your job is to find the most cost-effective way to deliver the best breakfast in the least time. 

You decide to invest in a continuous egg-boiler that provides a constant supply of perfectly boiled three-minute eggs.

Relying on continuous operation means lower flexibility, you can't adjust eggs if requested but customers benefit from lower costs and predictable quality.

Another issue is malfunction. If the egg-boiler malfunctions, the eggs in the machine can’t be sold, and you'll have to throw out the toast because there is nothing to serve with it. 

Use a functional test to prevent this. Open up the occasional egg to ensure quality.

Better yet, use in-process inspection and monitor the temperature of the water. Always opt for in-process tests over those that destroy the product.

Eggs could also be cracked, rotten, or take different times to cook, which is why incoming or receiving inspections are important.

If all your eggs are rotten, you'll need to close the cafe for the day. Consider having a raw material inventory capable of covering your consumption rate for the time it takes to replace it.

But inventory costs money, so weigh the pros against cons.

Just don’t forget about the opportunity at risk: how many customers would you lose, and what would it cost to lure them back?

The final thing to remember is that material becomes more valuable as it moves through the production process.

So detect and fix problems at the lowest value possible. Find and reject rotten eggs when they’re delivered, not on when they're on the customer’s plate.

A boiled egg is more valuable than raw, and breakfast in front of the customer is the most valuable.

As it turns out, even managing a cafe with a single dish and no employees is complicated. Just wait until we add more people…

Introduction to Management

Customers like what you're selling. You've made enough to invest in additional staff and automated equipment for the eggs, toast, and coffee.

Your output is no longer the breakfasts you deliver personally. Rather it's the breakfasts your cafe delivers, profits generated, and customer satisfaction.

But how do you track that? You need to come up with indicators focused on specific goals and track progress against them.

Here are the five that I'd pick:

1. Sales forecast: How many breakfasts should you plan to deliver? To assess how confident you should be in your prediction, you need to measure the variance between breakfasts delivered versus forecasted.

2. Raw material inventory: Do you have enough eggs, bread, and coffee to get through the day? Have too much? Cancel today’s order. Too little? Order more.

3. Equipment: Does anything need repair or replacement? If so, rearrange the production flow and/or lower your forecast.

4. Workforce: Are all staff there? Do you need to move someone from eggs to toast?

5. Quality: You need to know what customers think. Otherwise, quality may dip as volume increases.

Indicators are essential for guiding your attention and driving decision-making. You need to look at them early each day so you can correct problems before they get out of hand.

At the same time, you need guard against overreacting. Do this by pairing indicators, so effect and counter-effect are measured. For example, increased sales and decreased quality is a problem.

Any measurement is better than none but truly effective indicators focus on output not activity. Indicators that are countable should be paired with indicators that stress quality. Measure breakfasts sold, not the number of eggs boiled.

Indicators:

• Outline objectives;

• Provide objectivity and measurability; and

• Allow comparison of similar teams.

The black box

Think of your cafe as a black box: raw materials and labor go in and breakfast comes out.

Indicators are windows into the black box which allow you to better understand the internal workings of the process without actually doing the work yourself.

Leading indicators show what the future may look like, providing time to take corrective action to avoid problems if needed.

To be effective, you must believe in their validity.

Choose credible indicators, so you act when warning signs appear. Otherwise all you'll get from monitoring is anxiety.

Quality is an excellent leading indicator as a reduction in quality often leads in fewer sales in the future.

Sales are what is known as a trend indicator where output is measured against time (this month vs. last month) and a standard or expected level.

By extrapolating from the past and comparing real results to the forecast, you are forced to think through why. These types of indicators are the best to get a feel for future trends.

You can also use linearity indicators to plot a goal against the month of the year. If everything goes to plan, your numbers should follow a straight line that would hit your target by the end of the period.

If you find yourself below the ideal straight line, you know you can only hit your target if you do better than you've done in the previous months for the remaining time. Linearity indicators give you early warning and time to take corrective action.

As you can tell, indicators are a big help for solving all kinds of issues. If something goes wrong, you have a bank of information ready to analysis. Without them, you're flying blind.

By the time you find a problem and gather the information needed to make a decision, the situation will have gotten worse. 

Controlling future output

There are two ways to control output:

1. Build to order: Making breakfast as requested; and

2. Build to forecast: Making breakfast in expectation of orders.

Build to order reduces inventory risk but slows production times. 

In contrast, build to forecast suffers when orders are higher or lower than the forecast. 

When building to forecast, you must run two simultaneous processes:

• Manufacturing: Raw material moves through production and becomes a finished good.

• Selling: Sales finds prospects and sells the product.

Ideally, these two processes finish at the same time with the order from the prospect arriving just as the product is finished.

In practice, this is rare. Orders might not come in time, customers can change their mind and manufacturing can miss deadlines or hit unforeseen issues.

This is why manufacturing and sales must prepare their own forecasts and work together to determine how much to produce ahead of time.

The more inventory you have, the quicker you can adapt to changes in market conditions.

But inventory costs money. So keep stock at the lowest-value stage possible, which has the highest production flexibility for a given inventory cost: raw eggs, not boiled eggs.

You can apply these same principles to the management of people rather than manufacturing flows. Forecast the number of people you need to accomplish a task and keep slack in the system to account for problems or increased demand.

Assuring quality

Assuring quality in production requires inspection.

The first type of inspection is known as an incoming material inspection or receiving inspection and happens before material enters your black box.

Inspections that occur within your black box are called in-process inspections and inspections that occur before the customer receives the product are called the final or outgoing quality inspection.

Assure quality and reduce costs by rejecting defective material at its lowest-value stage.

When material is rejected at incoming inspection, you have the choice to send it back to the vendor as unacceptable or you can waive your specifications and use it anyway. This latter would likely result in a higher reject rate during in-process and final inspections but may be less expensive than stopping production all together.

The choice you make will generally come down to economics. But never let substandard product reach the customer if it could cause complete failure.

Inspections, like inventory, costs money and interferes with production. So you'll need to balance improving quality while minimizing disturbances.

There are two main techniques to balance these needs:

1. Gate-like inspection: Holds material until tests are complete then accepts or rejects it. This slows down the manufacturing process.

2. Monitoring: The bulk of the material flows but you take samples to determine a failure rate.

Monitoring is cheaper and has no comparable slowdown but bad material may escape if this happens you'll need to reject material at a higher-value stage. If the failure rate is too high, you'll need to stop production.

As a rule of thumb, prefer monitoring when experience shows you aren't likely to encounter issues.

Another way to reduce costs is to use variable inspection.

No problem for weeks? Check less often. If problems develop, test frequently until quality returns. Variable review means lower costs and interference.

Productivity

Productivity is output divided by the labor required.

To increase productivity, you can increase the speed you work at or you can change what you do. The latter is the better choice. You want to increase the ratio of output to activity rather than increasing activity itself.

You want leverage. Leverage is output generated by specific work. The higher the leverage, the higher the output you generate for the same amount of activity.

Arrange work in your black box so every activity is high leverage. You can do this through automation and work simplification. If you can automate a task or make it simpler, invest the time to do so.

Remember, increased productivity comes from stressing output and increasing leverage. Increased activity can result in the opposite.

Managerial leverage

Let's first understand what a manager's output is. As a manager you must form opinions, make judgments, provide direction, allocate resources, and detect mistakes. But these are not your output, these are activities that can positively or negatively impact your team's output.

As a manager, your output is equal to the output of your organization plus the neighboring organizations under your influence.

This is because work is done by teams. You can do your individual work well but that is not enough. If you have people reporting to you, your output is the output created by your team and associates.

Even if you don't have any direct reports, if you gather and share information you're a manager. More specifically, a know-how manager and your influence on neighboring teams can be huge.

The most important thing to understand is that the individual work you do as a manager is not output. You need to focus on improving the output of your team and the teams you influence.

Output can be positively or negatively impacted through one or more of the five managerial activities:

1. Information-gathering: Reading reports, customer complaints, or memos and talking to people inside and outside of the organization. This is the basis of all other managerial work and where you should spend the majority of your time;

2. Information-giving: Conveying knowledge to members of your team and groups you influence. Beyond facts, you must communicate objectives, priorities, and preferences. This is extremely important because subordinates need the context to know how to make decisions themselves;

3. Decision-making: Occasionally, you'll make a decision. Prefer to participate in decision-making by offering input, forcing better choices to emerge, reviewing decisions made, and providing feedback. Decisions come in two forms: forward-looking decisions (deciding on an activity set) and responses to developing problems, which can be either technical (e.g. quality control) or involve people (e.g. talking somebody out of quitting);

4. Nudging: You shouldn't always provide instruction, instead nudge people toward a preferred course of action; and

5. Role modeling: You are a role model for your organization, subordinates, peers, and even supervisors.

Each of these activities can improve output but you need to spend time where leverage is the greatest. For some, this is in large groups. For others, one-on-one in a quieter, more intellectual environment is best.

Increasing managerial output

For every activity you perform, the output of your organization should increase by some degree. The extent to which it increases is determined by leverage.

Leverage is the measure of output generated by a given activity. The higher the leverage, the more output a given activity produces.

But not every activity increases output and doing more can often reduce output.

That's why the key to high output is focusing on increasing leverage, not activity.

In summary, you can increase managerial productivity in three ways:

1. By doing things faster

2. By improving the leverage of existing activities; and

3. By shifting the mix of activities from low to high leverage

High-leverage activities

High-leverage activities:

• Affect many people;

• Change a person’s activity or behavior for a long time based on a brief interaction; or

• Impact a large group’s work by providing a critical piece of information. 

It's important to flag that leverage isn't always a good thing. Just as high-leverage activities can dramatically increase output, high-leverage activities done poorly can dramatically reduce it.

A great example of this is managerial meddling. If you constantly assume control of a subordinate's work, they'll show less initiative in solving their own problems and instead refer the work to you.

The art of management is selecting and concentrating on one, two, or three high-leverage activities and ignoring the rest.

Delegating as leverage

Delegation is an essential part of management.

Given a choice, do you delegate activities you are familiar with or those you that aren’t? Before answering, consider that delegation without monitoring is abdication. 

When you delegate, you are still responsible for the task’s completion. Monitoring the delegated task is the only practical way of ensuring that. And it's easier to monitor what you know, so given the choice, delegate activities you know best.

But monitoring should not be confused with meddling. Your goal is to check in to ensure the activity is proceeding along as expected, not to do the work yourself.

Like any production flow, monitor at the lowest-value stage. Review rough drafts, not final reports.

A second principle we borrow from production is variable inspection. Employ different sampling rates for different subordinates. How often you check in should be based on the subordinate's task-relevant maturity, not what you believe they can do in general.

As task-relevant maturity goes up, you monitor less often.

Finally, go into details randomly. Checking everything is equivalent to stopping the product line to check for faults. Remember, always opt for in-process tests over those that stop production.

Production principles for time management

A great deal of your time as a manager is focused on allocating resources: manpower, money, and capital. But the single most important resource you can allocate is your own time. Your time is the only truly finite resource you have.

Improving how you handle your time is the single most important thing you can do to increase output.

You can manage your time better by applying production principles:

1. Identify the limiting step: Determine what is immovable and manipulate more flexible activities around it.

2. Batch similar tasks: Everything requires a certain amount of mental set-up. Efficient work relies on grouping related activities. 

3. Build to forecast: The majority of your work should be by forecast, and the medium is your calendar. Most people treat calendars as a place for orders to come in. You should use it as a production planning tool. Schedule work that is not time-critical between the limiting steps of your day. And just as a factory manager says no to additional jobs if the factory is at capacity, you should say no to tasks that would overload your system. 

4. Say no earlier: Stop work before things reach a higher-value stage. 

5. Allow slack in your schedule: One interruption shouldn’t kill your entire day.

6. Carry a raw inventory: In the form of projects that don’t need to be finished now, but would increase your team’s productivity over the long term. This also prevents you from meddling in your subordinates’ work. 

7. Standardize: While continuing to think critically about what you do and the approaches you use. 

How many direct reports should you have?

Too few or too many direct reports reduces your leverage. If your work is supervisory, aim for six to eight direct reports—this ensures half a day per week for each subordinate.

As a know-how manager, aim for six to eight subordinates or equivalent internal customers. Anything less than six to eight direct reports will result in on-the-job retirement or managerial meddling.

Interruptions: the plague of managerial work

Do everything you can to prevent starts and stops in your day. Even if there is an emergency, think about creating indicators that would have provided insight into the problem before it became time-sensitive.

Your work relies on working with other managers. You can only move toward regularity as others do. Do your part by scheduling recurring meetings at the same time each week.

Uncontrolled interruptions are inevitable and most frequently come from subordinates and people outside your team whose work you influence. 

Force frequent interrupters to make an active decision about whether the issue can wait. Block out your calendar with a note that says, “I am doing individual work. Please don’t interrupt me unless it can’t wait”.

But understand that interrupters have legitimate problems they need solved. That’s why they're asking you.

To reduce interruptions, batch them into an organized, scheduled form such as scheduled meetings, one-on-ones, or office hours. If held regularly, people will batch questions instead of interrupting you whenever they want.

And just as a manufacturer produces standard products, you should pin down the common types of interruptions you get and prepare standard responses.

Finally, use indicators to reduce the time you spend dealing with interruptions. A good set of indicators means you can answer questions quickly and without the need for ad-hoc research.

The key is to impose a pattern on the way you handle interruptions.

Meetings: The medium of managerial work

Meetings often feel like a waste of time. Just something you need to endure before you can get back to your real job. But remember, as a manager your output is equal to the output of your team and the teams you influence. You can increase or decrease output through one or more of the five managerial activities:

1. Information-gathering

2. Information-giving

3. Decision-making

4. Nudging; and

5. Role modeling.

Meetings are nothing more than a medium to do managerial work. You can do any these activities inside or outside of meetings. Just choose the most high leverage medium for what you want to accomplish. The one that produces the most output for the least amount of work.

Meetings fall into two categories:

1. Process-oriented: These meetings take place on a regular cadence and are designed to promote knowledge sharing and information exchange; and

2. Mission-oriented: These are ad-hoc meetings designed to solve a specific problem by making a decision.

Process-oriented meetings

To maximize efficiency, infuse process-oriented meetings with regularity. Attendees should know how the session will run, what matters will be discussed, and what the goal is.

By doing this, you ensure that the meeting has minimal impact on output. Process-oriented meetings come in three types:

1. One-on-ones: Meetings with you and a direct report;

2. Staff meetings: Meetings with you and all your direct reports;

3. Operation reviews: Meetings that allow people who don’t frequently meet to meet. Operation reviews should include formal presentations where managers describe their work to other managers who aren’t immediate supervisors, and to peers in other parts of the company.

One-on-ones

One-on-ones are meetings between you and a direct report and are the primary way of maintaining and deepening your relationship. They can be incredible high-leverage because the hour you spend together can impact the report's work for weeks.

The primary purpose of a one-on-one is mutual teaching and information exchange.

You teach your team members your skills, know-how, and suggest ways to approach things while they provide information about what they’re doing and their concerns.

Schedule one-on-ones based on the job- or task-relevant maturity of each report. Have frequent one-on-ones (once a week) with those inexperienced in the specific situation, and less frequently like once a month for experienced veterans. 

Carve out at least an hour. Anything less tends to confine people to simple problems that can be solved quickly.

A key thing to understand is that one-on-ones are the report's meeting. They should set the tone and agenda. This increases your leverage because you don't have to create a plan for each of your direct reports.

It's also important because it forces them to think through what they want to talk about in advance.

The meeting typically starts with indicators managed by the report, such as order rates, production output, or project status. Focus on indicators that signal trouble. 

You should also cover anything important that has happened since the last meeting: hiring problems, people problems, organizational issues, and new plans. The criteria of inclusion is whether the issue bothers your subordinate. 

Your role is to learn and coach. When they have stopped talking about a topic ask another question about it until you feel like you've gotten to the bottom of the problem. 

You should both have a copy of the agenda and take notes against it. Taking notes promotes active listening and ensures that actions required by either party aren’t lost. Consider using a “hold” file where you can both accumulate essential but not urgent issues to discuss in the future. 

And where possible encourage heart-to-heart conversations. One-on-ones are the perfect forum for getting at subtle work-related problems. 

Schedule one-on-ones on a rolling basis. Setting up the next one as the current one ends makes it easy to account for other commitments and avoid cancellations. 

Staff meetings

Staff meetings involve you and your team members and allow peers to interact because of this they're an ideal place for decision-making.

Anything impacting more than two people present is fair game. If something degenerates into a conversation between two people, break it off, and move onto something that affects more people. Suggest the two book a separate meeting to discuss. 

Staff meetings should be relatively structured with an agenda prepared ahead of time so everyone has a chance to prepare. You should also include unstructured time so people can bring up anything they want. If required, the things brought up in the unstructured time can become part of the agenda of a future meeting.

Your role is to be the leader, observer, expediter, questioner, and decision-maker when required.

But ideally, you just keep things on track while your team works out issues.

Operation reviews

Operation reviews are the medium of interaction for those who don’t otherwise have the opportunity to meet. They allow employees several organizational levels apart to teach and learn from each other.

Operation reviews involve four groups:

1. The organizing manager: They help presenters decide what issues and details to present, book the meeting, and are timekeeper;

2. The reviewing manager: A senior supervisor whom the review is aimed at. They ask questions, make comments, and are a role model for the junior managers who present;

3. The presenters: Presenters use visual aids where possible, dedicating four minutes of presentation and discussion to each slide; and

4. The audience: Audience participation is key so they should questions and make comments.

Mission-oriented meetings

Mission-oriented meetings are ad-hoc and designed to produce a specific output like a decision. 

The key to their success is the chairperson, who has the most at stake in the outcome. Usually, this is who calls the meeting. 

All too often chairpeople turn up like any other attendee and hope things go as planned. This is why meetings often feel like a waste of time. You need to have a clear understanding of the meeting's objective before you call for it.

So before calling a meeting, ask yourself:

• What am I trying to accomplish?

• And is a meeting necessary? 

If both answers aren’t yes then the meeting will be a waste of time. Likewise, if you are invited to a meeting and the requester can't answer these questions get it cancelled. Meetings are expensive and time is finite.

Assuming a meeting does need to happen, your role is to identify who should attend. If someone can't attend, see if they can send someone in their place.

Remember, mission-oriented meetings are called for a specific purpose. So keep them to no more than six or seven people, decision-making is not a spectator sport.

If you call the meeting you are responsible for the logistics, setting the agenda, and writing the minutes. Once the meeting is over you must share the minutes quickly before attendees forget what happened. They should be clear and as specific as possible, telling the reader what was decided, who is to do it and when.

You need to get attendees the minutes fast before they forget what happened. If writing minutes is too much of a problem, the meeting wasn’t worth calling in the first place. 

Ideally, you would never call mission-oriented meetings.

In practice, routine meetings should take care of 80% of problems. Ad-hoc meetings are needed to deal with the rest. If you’re spending more than 25% of your time in ad-hoc meetings, that’s a sign of poor organization.

Decision-making

Making decisions–or more accurately, participating in the decision-making process is an essential part of managerial work. Decision can range from trivial like where to go to lunch to complicated like should we disrupt our own product?

Traditionally, the chain of command was precisely defined and a person's ability to make a decision was dictated by their position in the org chart.

Today, businesses mostly deal with information and know-how. This can cause a large gap between a person's power based on position and another person's power based on knowledge. It's increasingly common that the person who is higher up in the org chart is wrong person to make a decision.

Subordinates often have a better idea of what the right solution is. They're closer to the problem and are immersed in it day-to-day.

The faster the business is changing, the greater the divergence between knowledge and position power will be. Your job as a manager is to ensure that the best decision is made, not that you make the decision yourself.

The ideal decision-making process

The ideal decision-making process begins with free discussion where all points of view and aspects of the issue are openly debated. The greater the disagreement, the more important free discussion is.

When things get heated, people tend to hang back and wait to see what the likely outcome will be. Once they know, they throw their support behind the idea to avoid being associated with the losing position.

All this produces is bad decisions, because knowledgeable people withhold opinions and whatever decision is made is based on information and insight that is less complete than it could have been. A good way to avoid this is to have the most junior person to state their opinion first.

The next step is to reach a clear decision. Again, the greater the disagreement, the more important this step is. You need to articulate the decision and make sure everyone understands. Our natural tendency is to do the opposite.

Once a decision is made, everyone must give their full support.

Not everyone has to agree that it is the right decision, as long as they commit to backing it. As Jeff Bezos says, disagree and commit.

This process may seem simple but it's easier to accept in theory than in practice. It's hard for express your views forcefully, it's hard to make unpleasant decisions, and it's even harder to go against the group. But it's worth doing to reach the best decision.

To recap, the ideal model for decision-making is:

1. Free discussion: All points of view and aspects of an issue are openly debated regardless of a person's status;

2. Clear decision: The decision is clearly articulated; and

3. Full support: Everyone commits to supporting the decision, even if they disagreed prior.

If the decision turns out to be the wrong, go back to step 1.

The final thing you should aim for is to have decisions made at the lowest competent level. Decisions should be made by those who are closest to the situation and know the most about it.

The peer-group syndrome

The most common problem in decision-making is the peer-group syndrome.

People are reluctant share their opinion in the presence of their peers because they fear that their opinion is different to the group's. This leads to a feeling out process where people wait for consensus before taking a position.

If and when a decision emerges it will be stated as the group's opinion, I think our position is...not as a personal opinion. Once this happens the rest of the group buys in and the position becomes solidified.

This is obviously not free discussion and therefore, not the ideal decision-making process.

To prevent this, aim to have peers-plus-one more senior manager in meetings.

Peers tend to look to the more senior manager to take over and shape the meeting. Even if they aren’t the most competent or knowledgeable person involved.

Don't be afraid to sound dumb. Sounding dumb is one of the best ways to surface insights. Remember, every time a fact or insight is withheld the decision-making process is worse off than it could have been.

A related phenomenon that influences lower-level people is the fear of being overruled. 

Suppose the group (or senior-level manager) opposes a junior person’s position. If this happens, they may lose face in front of their peers, which is why junior people hang back and let more senior people set the direction. 

This is why it's a good idea to have the most junior person in the room speak first.

Six questions to answer before making a decision

While you shouldn't end the free discussion phase too quickly, you can't always reach consensus.

When the time for a decision has arrived, the senior person who has guided, coached, and prodded the group along has to make the decision. 

If the ideal-decision making process has been followed, the decision-maker has heard all points of view, facts, opinions, and judgments and has the information needed to make a decision.

Like all other managerial activities, focus on output which in this case is the decision itself.

To avoid blindsiding people, answer these six questions in advance: 

1. What decision needs to be made?

2. When does it have to be made?

3. Who will decide?

4. Who will need to be consulted before making the decision?

5. Who will ratify or veto the decision?

6. Who will need to be informed of the decision?

Planning: Today’s actions, tomorrow’s output

Planning is best understood through the lens of our production principles. Recall that a key method of controlling future output is to forecast demand and then build to forecast. If projections don't match reality, you need to increase or decrease production.

These principles apply not only in production but in any planning process. We can break planning into three steps:

1. Establish environmental demand: What will the environment demand of you, your business, or your organization?

2. Understand your present status: Where will you be if you change nothing?

3. Close the gap: Compare and reconcile steps 1 and 2. What more or less do you need to produce to meet projected demand?

Let's dive into each step in more detail.

Step 1: Establish environmental demand

Your environment is defined by the groups that directly influence what you do. This includes your team and any neighboring teams you rely on, your customers, the vendors you use, and the competitors your customers judge you against. You also need to factor in any technological changes that could augment or replace you.

Once you have established your environment, you need to look at it in two time frames: now and in the future.

Ask yourself: what do my customers want now? Am I satisfying them? And what will they expect from me in the future?

The most important thing is to focus on is the difference between your environmental demands now and what you expect them to be.

Don't worry about how you're going to meet demand just yet. Your goal is only to determine whether your current activities are meeting current demands and whether you need to do anything to meet future demand.

Step 2: Understand your present status

Now it's time to determine your present status. List out your current capabilities and projects. As you do this, try to use the same terms you use to state demand. If you define demand in terms of new features or products, your present status should be too.

You also need to factor in that some of the projects you're working on will be scrapped, some of your time will be eaten up by busy work, and that you need slack in your system to account for problems or changes in demand.

Step 3: Close the gap

The final step is to start new tasks or modify existing ones to close the gap between your environmental demand and what your present status will produce.

What do you need to do to close the gap? And what can you do?

Consider each question separately and then decide what you will actually do. It's likely that you won't be able to do everything you want to do. Your strategy should be to choose the actions that produce the most leverage. These are the actions that produce the most output for the least amount of work.

Learn more about strategy by reading my post on 7 Powers.

The output of the planning process

The output of the planning process are the decisions that are made and the actions that are taken. The goal of planning is to produce a set of tasks that are performed now to affect future events.

Ask yourself: what do I have to do today to solve–or better– avoid tomorrow’s problem?

Today’s gap is yesterday’s planning failure. 

You should focus and implement only the portion of a plan that lies between now and your next planning process. Everything else you can and will look at again. You should also be careful not to plan too frequently, you need feedback to determine whether the decisions you made were correct.

When planning you need to involve the operating management of the organization. The people planning cannot be separate from the people who are implementing the plan. The leverage you can gain from planning can only be realized by marrying planning and implementation.

Finally, remember that saying yes to one thing is saying no to something else.

Management by objectives and key results (OKRs)

Once demand is well-defined and planning is complete, you can use management by objectives and key results (OKRs). The idea behind OKRs is simple, if you don't know where you're going, you won't get there.

To be successful with OKRs, you need to answer two questions:

1. Where do I want to go? The answer provides your objective.

2. How will I know if I am getting there? This gives you your key results.

OKRs are designed to provide feedback relevant to the task at hand and should tell you how you are doing and whether you need to make adjustments. For feedback to be effective, you need to receive it as soon as possible. So OKRs should be set for a relatively short period of time. If you plan for a year, you should set OKRs on a quarterly or even monthly basis.

Keep your number of objectives small. If you focus on everything, you focus on nothing. A few well-chosen OKRs are what make the system work. And if you want to dive deeper into OKRs, I highly recommend Measure What Matters by John Doerr.

Organizational structures

As your organization scales, you'll find that the set of tasks and skills needed to run a small team is very different from those needed to run a large one. The central decision you'll need to make is when to centralize and decentralize decision-making.

If you centralize certain activities, you can do many things at a lower cost and higher quality level. Inversely, centralized decision-making often means you're further from the problem and may not come up with the best solution.

This centralization-decentralization dichotomy is so pervasive that it is one of the most important themes in management.

Recall that management is a team game.

A manager’s output = the output of their organization + the output of the neighboring organizations under their influence

As you scale, management becomes not just a team game but a team of teams game where various teams exist in mutually supportive relationships.

Now let's dive into how organizations can be composed.

Mission-oriented organizations

Mission-oriented organizations are completely decentralized and each business unit pursues its goals with little tie-in to other units. In this model, individual teams are responsible for all elements of their operation from marketing to hiring to developing product.

The only benefit a mission-oriented model provides is that it empowers managers who are closest to the problem to respond to changes in environmental demand fast. All other considerations favor a different model, namely a functional organization.

So why have mission-oriented organizations at all? Because the business of any business is to respond to environmental demand and the need to be responsive is so important that most organizations ends up with mission-oriented groups.

Functional organizations

Functional organizations are at the other extreme and are completely centralized. Decisions are made at headquarters and cascade down with each unit being responsible for a particular discipline across the organization, such as product, growth, engineering, sales, or finance.

The functional model provides obvious economies of scale and allows the expertise of know-how managers in each operational area to be leveraged across the entire organization. It also means you can reallocate resources in response to changes in priorities and business units can focus on mastery.

But there are disadvantages to the functional model. Functional groups suffer from information overload and must respond to demands from different business units that often struggle to adequately communicate their needs. This can lead to internal competition between business units who compete for priority and control of functional groups.

This infighting is clearly a waste of time and resources as it doesn't contribute to the output of the company.

Hybrid organizations

In practice, most companies generally exist between the two extremes of mission-oriented and functional organizations. To quote Grove's Law: All large organizations with a common business purpose end up in a hybrid organizational form.

This mix provides a balance between the responsiveness of mission-oriented units and the leverage provided by functional groups.

The only except to Grove's Law are conglomerates which are typically organized in a totally mission-oriented form. This works when subsidiaries are independent and bear no relationship to one another. But even in this situation, each subsidiary tends to be structured as a hybrid organization.

At the end of the day, each hybrid organization is unique. There are an infinite number of permutations that lie between a totally mission-oriented and a totally functional organization.

In fact, a single organization can and should fluctuation between the two extremes to match the operational styles and aptitudes of its managers.

The most important problem to solve is the optimum and timely allocation of resources and the efficient resolution of conflicts arising over that allocation.

While centralized planning is attractive in theory, it falls over in practice. Centralized planners just can't respond nor predict changes in demand fast enough.

In practice, the answer lies with middle managers who are closest to the problem of generating and consuming internal resources. For middle managers to succeed, two things are necessary:

1. They must accept the inevitability of the hybrid organization

2. They must develop and master the practice through which a hybrid organization is managed. Namely, dual reporting.

Dual reporting

Dual reporting, as the name suggests, is when an employee reports to two managers who are jointly responsible for the employee's performance.

The need for dual reporting is fundamental. Think about how a person becomes a manager. The first step in your career was likely as an individual contributor. As you leveled up and improved your skills, you were promoted to more senior positions and then eventually to a management position where you supervised people in your functional speciality.

But at some point, if you keep getting promoted, you'll find yourself as a mission-oriented manager, managing people in disciplines that you have no real experience in. So while you're perfectly capable of managing the more general aspects of the job, you have no choice but to leave the technical aspects to your report.

One solution is to have all employees of a particular discipline report into a functional manager instead of you. But the more you do this, the more you move toward a totally functional organization which as we learned in the last episode reduces your ability to respond to changes in environmental demand.

The desired state is to have the immediacy and operating principles come from the mission-oriented manager while also having technical supervisory come from a functional manager. The solution is dual reporting.

Peer groups as technical supervisors

The role of the technical supervisor doesn't need to be filled by a single person. In fact, it's commonly solved through a peer group who come together to discuss common problems that their mission-oriented bosses can't help them with. In effect, they're now supervised by their boss and a group of peers.

To make this work you need the voluntary surrender of individual decision-making to the group. Being a member means you no longer have complete freedom and must go along with the decisions of your peers most of the time.

This only works if you trust your peers and trust is built on the back of culture. Culture is the set of values and beliefs, as well as the way things are done within a company. A strong and positive culture is essential if dual reporting and decision-making by peers are to work.

This setup makes a manager's life more ambiguous and most people don't like ambiguity. This leads to people looking for something simpler but the reality is that it doesn't exist. Dual reporting in some form is needed to make hybrid organizations work.

Making hybrid organizations work

To make hybrid organizations work, you need a way to ensure that the resources of functional groups are allocated and delivered to meet the needs of mission-oriented units. The professional methods, practices, and standards are set by the functional manager while general managers give mission-oriented priorities by focusing people on specific problems.

Dual reporting is certainly a tax on productivity as it requires everyone to understand the needs and thought processes of their peers but there is no real alternative when you need to manage hybrid organizations.

Dual reporting and the hybrid organizational form it supports is the inevitable consequence of being a large organization, but should not be an excuse for needless busywork. You still need to remove unnecessary bureaucratic hindrance, apply work simplification, and most of all focus on increasing leverage.

Just don't expect to escape from complexity by playing with reporting arrangements. The hybrid structure is fundamental to any large organization.

The two-plane organization

A subtle variation of dual reporting happens whenever you become involved in coordinating something that is not part of your regular work.

For example, if you work as a sales manager and are part of a committee of sales managers who are responsible for improving the sales processes at your company, you are “reporting” to multiple people.

Your responsibilities can't fit on a single org chart. Instead, the two hierarchies operate on different planes.

This arrangement uses your skills and how-how in two different capacities which allows you to exert much larger leverage on the organization. In your main job, your knowledge improves your team’s performance; in the second, you influence the work of all salespeople at your company.

It also turns out that people who are a subordinate in one plane may be the leader in another. For example, you might be the leader of the sales committee but a subordinate in a strategic planning group.

And if you can operate on two planes, you can likely operate on three or more. The point is that the two- (or multi-) plane organization is very useful. It allows you to be the leader when appropriate and the follower when you are not the most qualified person to lead which gives the organization more flexibility.

The key factors common to all of these arrangements is the use of cultural values as a mode of control.

Modes of control

Our behavior in a work environment is controlled by three invisible and pervasive means:

1. Free-market forces: Free-market forces apply when actions are based on price like when goods and services are exchanged between two entities seeking to enrich themselves. No one needs to oversee the transaction because everyone is openly serving their own self-interest. Free-market forces fail when value isn't easily defined.

2. Contractual obligations: When value is hard to define and free-market forces fail, use contracts as the mode of control. A common example of this control are employment contracts which define the kind of work you do and the standards that will govern it. Because it’s hard to specify what people will do day-to-day, you’ll need to have generalized authority to monitor, evaluate, and correct work where necessary; and

3. Cultural values: When the environment changes faster than rules can or when circumstances are too ambiguous for contracts, use cultural values as the mode of control. The most important characteristic of this control is that the interest of the group must take precedence over individuals. For this to work, you must trust the people you work with and have a common set of values, objectives, and methods which can only be developed through shared experience.

The role of management

Modes of control govern what we do. From one day to the next, you’ll find yourself influenced by all three. When you buy something, market forces are affecting you. When you go to work, you’re adhering to contractual obligations. And when you do work outside your job description to help a colleague, that’s cultural values. 

The role of management is to choose the most appropriate mode of control for a given situation.

When free-market forces work, you don't need management. In a contractual obligation, management's goal is to set and modify the rules, monitor adherence, and evaluate and improve performance.

As for cultural values, management needs to develop and nurture a set of common values, objectives, and methods that are essential for the existence of trust. One way to do this is by articulating the values, objectives, and methods. The other, more important way, is by example.

If your behavior at work is in line with the values you profess, you foster the development of culture.

Choosing the most appropriate mode of control

It's tempting to idolize cultural values as the best mode of control. But it's not the most efficient mode of control under all conditions.

What mode of control you use will depend on a person's motivation and the complexity, uncertainty, and ambiguity of their environment. For short, we'll call this the CUA factor.

Let's construct a 2x2 matrix with individual motivation on the Y-axis and the CUA factor on the X-axis. Individual motivation can range from self-interest to group-interest and the CUA factor can range from low to high.

When self-interest is high and the CUA factor is low, free-market forces are the most appropriate mode of control. As individual motivation moves toward group-interest, contractual obligations become most appropriate. If group interest and the CUA factor are high, then cultural values become the best choice. Finally, when self-interest and the CUA factor are high, no mode of control will work well.

Let's apply this model to the work of a new employee. Their motivation is likely based on self-interest so you need to give them a structured job with a low CUA factor. As they do well, they'll start to feel more at home and worry less about themselves and more about their team. They can then be promoted to a more complex, uncertain, and ambiguous job which also tends to pay more.

As time passes and they gain an increasing amount of shared experience with other members of the team, they can tackle more and more complex, ambiguous, and uncertain tasks. This is why promotion from within tends to be preferred by companies with strong cultures.

This is also why it's hard to bring in senior managers from outside the organization. Like any new hire, they come in with high self-interest but likely inherit a job with a high CUA factor. All you can do is hope they quickly move from self-interest to group-interest and just as quickly gets on top of their new role to lower the CUA factor.

The sports analogy

Recall that a manager's output is equal to the output of their team and the teams they influence.

Put another way, management is a team activity. No matter how well a team is structured, no matter how well it is managed, it will only perform as well as the individuals that make it up.

Everything outlined in the previous episodes is useless unless your team continually offers their best. The single most important task of a manager is to elicit peak performance from their team.

You can improve performance by increasing a person's capabilities through training or by increasing motivation.

When a person isn't doing their job, it's because they are incapable or unmotivated.

To determine which, ask yourself: if their life depended on it, could they do it? If yes, they're not motivated. No? they're not capable.

Motivation through Maslow’s hierarchy

For most of history, motivation was based on fear of punishment. If you didn't work hard, you didn't get paid.

This worked because manual labor was readily measurable, departures from the norm were easy to spot, and most people were operating in the lower levels of Abraham Maslow's hierarchy of needs.

But as the relative importance of manual labor decreases, the corresponding importance of knowledge work increases, and people move into higher levels of Maslow's hierarchy, motivation by fear stops working.

For context, Maslow defined a set of needs that tend to lie in a hierarchy. When a lower need is satisfied, the higher need tends to take over.

Under this model, creating and maintaining motivation relies on some form of dissatisfaction.

Let's take a look at the motivators at each level in more detail.

At the physiological level, one fears the deprivation of food, clothing, and other basic necessities.

At the safety level, the desire to protect yourself from slipping back to the physiological level is the main motivator.

At the social level, you are driven by your inherent desire to belong to a group whose members possess something in common with you.

Our physiological, safety, and social needs are what motivate us to show up to work, but the final two levels of Maslow's hierarchy are what make us perform when we get there.

Our need to keep up with or emulate others is what drives us at the esteem level. It's why we feel the pull to keep up with the Joneses.

But like the other levels, once goals are met, keeping up with people stops being a good motivator.

The final level of Maslow's hierarchy is self-actualization. It stems from the personal realization that "what I can be, I must be". Once motivation is based on self-actualization, the drive to perform is limitless.

Unlike other forms of motivation that extinguish once fulfilled, self-actualization can drive people to ever-higher performance levels. 

Some people–not the majority–need to achieve in everything they do and will self-actualize without external input.

For everyone else, management needs to foster an environment that promotes self-actualization by setting objectives high enough that they stretched the outer limits of each person's abilities.

Objectives should be set high enough, so even if individuals push hard, there is only a 50% chance of reaching them.

Output increases when everybody strives for a level of performance beyond their immediate grasp, even if they fail half the time. 

The most important thing to do is to create an environment that values and emphasizes output.

Money as a motivator

At the lower levels of Maslow’s hierarchy, money is essential. After a certain point, more money doesn’t equal more motivation. 

Unless money is a measure of achievement, money at the physiological and safety stages only motivates until needs are met. Money as a measure of success can motivate without limit.

A simple test can be used to determine where someone is in Maslow's hierarchy. If the absolute sum of a raise is significant, you are working within the physiological or safety levels. If what matters is how your raise stacks up against others, you are motivated by esteem or self-actualization, money is a measure of achievement. 

Once self-actualizing, money is a proxy for progress and an important feedback mechanism for performance. Outside of money, self-actualized people want to improve their performance but feedback will need to be individualized.

Work as a game

You now know how motivation can be used to elicit peak performance. Your role as a manager is to train individuals to improve their capabilities and to bring them to the point where they are self-actualizing. Once there, motivation is self-sustaining and limitless.

Money alone will not get people there. You need to endow work with sense of play and competition.

Work should be a game. Your role as a manager is that of a coach. You establish the rules of the game and give employees ways to measure their performance.

As coach, you take no personal credit for your team's success and because of that, your team trusts you. You're tough, but fair, eliciting peak performance from each person. And finally, you were probably a good player in the past. Having played well, you understand what it takes to perform.

Seeing work as a game also teaches us how to cope with failure. In any competitive sport, at least 50 percent of all games are lost. When everyone knows that from the outset, failure is seen as par for the course and not something to dwell on.

Task-relevant maturity (TRM)

As you now know, the most important task of a manager is to elicit peak performance from their team.

Assuming you now understand what motivates people, the question becomes: what is the optimum management style?

Historically, what is seen as the best management style changes based on the most popular motivational theories at the time.

In the early nineteen-hundreds, ideas about work were simple. Managers told people what to do, and if they did it, they got paid. If they didn't, they were fired.

Leadership was crisp and hierarchical with order-givers and order-takers.

In the 1950's management theory shifted toward humanism as people realized there were nicer ways to get people to work.

Finally, as behavioral science developed, theories of motivation and leadership became subjects of carefully controlled experiments.

The findings were surprising. No one approach was effective under all conditions. There is no single best management style

The optimum approach depends on task-relevant maturity.

What is task-relevant maturity?

Your management style should change depending on your subordinate's task-relevant maturity, which is a combination of their motivation, readiness to take on responsibility, education, training, and experience.

This is all specific to the task at hand. It's very likely that a person will have high maturity in one job but low in another. Likewise, people can have high maturity at a certain level of complexity, uncertainty, and ambiguity, but if the pace accelerates or the role changes, their maturity will drop.

When maturity is low, the most effective approach is structured, offering detailed instruction about what needs to be done, when, and how.

As maturity increases, the most effective style moves from structured to one focused on two-way communication, emotional support, and encouragement. The individual becomes more important than the task at hand.

When maturity peaks, your involvement should be minimal. Focus on ensuring the objectives your subordinate is working towards are agreed upon and correct.

But remember, you always need to monitor your team's work closely regardless of their task-relevant maturity. The presence or absence of monitoring is the difference between delegating and abdicating a task.

Avoid making a value judgment by thinking that a structured management style is less worthy than a communication-oriented one. Remember, a manager's output is the output of their team and the team's they influence. You must focus on the style that is most effective for the subordinate's maturity.

When a subordinate has low maturity, you need to teach them how you think and do things, so they can later make decisions the way you would. If you can't teach your subordinate and transmit a shared set of values, you cannot effectively delegate.

Increased maturity = increased leverage

You need to raise the task-relevant maturity of your team as fast as possible. The higher a person's task-relevant maturity, the less time it takes to manage them and once they learn your operational values, you can delegate tasks to increase your leverage.

At the highest levels, your subordinate's training is complete and motivation will likely come from self-actualization, which we know is the most potent form of motivation a manager can harness.

But remember, a person's maturity depends on their environment. When things change, maturity changes, and you must adapt your management style.

On paper, management by monitoring alone is most productive but you need to work your way up to it. Even when you get there, things can change, and you may need to revert to a what-when-how mode of management.

Why it’s hard to be a good manager

Deciding on the task-relevant maturity of each team member is hard. Even when you know, your personal preference tends to override logic and dictate your management style. We tend to opt for either a structured or a minimal approach, either fully immersed in the work of a subordinate or leaving them completely alone.

Another problem is that we often to see ourselves as more effective communicators and delegators than we really are. We also throw out suggestions that subordinates can perceive as marching orders which compounds this problem further.

A common mistake is to conflate a social relationship with an effective management style. Close relationships off the job can help create an equivalent relationship on the job, but the two shouldn't be confused.

This raises the question: Is friendship between you and a subordinate a good thing?

If a subordinate is a friend, communication-based management tends to be easy but what-when-how is harder to revert to when necessary.

You must decide what is appropriate for you. Imagine delivering a harsh performance review to a friend. If you cringe at the idea, don't make friends at work. If you're unaffected, personal relationships will strengthen work relationships.

Performance appraisal: Manager as judge and jury

Performance reviews are the single most important form of task-relevant feedback you can provide, and the truth is most managers don't do an especially good job at it.

How you assess performance, deliver the assessment, and allocate rewards like promotions, dollars, and stock options, will positively or negatively influence a report's performance for a long time. In short, performance reviews are incredibly high-leverage.

The fundamental purpose of a performance review is to improve a report's performance by focusing on their:

• Skill level: To determine what skills are missing and how to remedy them; and

• Motivation: To increase performance at the same skill level. 

Reviews represent the most formal type of leadership. It's the only time you are mandated to act as judge and jury, you are required to make a judgement and deliver it to a fellow worker.

These two aspects of the review–assessing performance and delivering the assessment–are equally difficult, so let's look at each in more detail.

Assessing performance

The most common mistake when assessing performance is to not clearly define what you want from a direct report. If you don't know where you want to go, you can't know if you've got there.

But assessing the performance of knowledge workers is difficult. Knowledge work is fuzzy and most jobs involve activities that don't produce output in the period under review.

You must give these activities appropriate weight when assessing performance, even if you can't be 100% objective.

Using the managerial black box, we can characterize performance by output and internal measures. Output are things you can and should plot on a chart. Think completed projects, quotas, or increased yield.

Internal measures are what happen inside the black box: activities to create output for the period under review and those that set the stage for future output.

There's no formula to determine how to weigh output versus internal measures. The proper weighting could be 50/50, 90/10, or 10/90 and will shift over time.

A similar trade-off is how you weight long-term performance against short-term performance. Think about the present value: how much will the future-oriented activity pay back over time? and how much is that worth today?

You also need to consider the offset between activity and output. During the review period, output may have all, some, or nothing to do with the activities done during the same period.

You need to judge performance not just see and record what is in plain sight.

If you are reviewing a manager's performance do you judge individual performance or performance of the group under supervision? Ultimately the performance of the group, but you must determine whether they have added value and whether they are likely to increase the output of their team in the future.

Avoid the potential trap. Force yourself to assess performance, not possibility.

When deciding who to promote, understand no action communicates your values more clearly than who you promote. Every promotion creates a role model for others in your organization. 

Finally, no matter how well a person has performed, find a way to suggest an improvement. Remember, the point of the review is to improve performance. 

Delivering the assessment

When you are delivering a review, you want to level, listen, and leave yourself out.

Level with your report. The integrity of the system relies on honesty. And don't be surprised if praising someone is as hard as critiquing them.

The goal of the review is to transmit your thoughts to your report. The more complex the issue, the higher the probability of miscommunication. You need to listen and employ all your sensory capabilities to ensure that you are being heard. It's your responsibility to continue until you are satisfied that the review has been understood.

Leave yourself out. The review is about and for your subordinate, not you. Keep your insecurities, anxieties, guilt, and emotion out of it.

Delivering mixed reviews

Most reviews contain positive and negative assessments. Avoid delivering a list of superficial, cliche, or unrelated observations. Long lists leave your report confused and won't improve performance.

We all have a finite capacity to deal with facts, issues, and suggestions. You may want to raise seven things, but if they can only take on four, at best you'll waste your breath. At worst, you'll overwhelm them and they won't get anything from the review.

Remember, the purpose of the review is to increase performance not to share everything.

Less is often more.

Look for relationships between related issues and group them into a single item in the review.

Once you have your list, ask yourself whether your report will remember everything you have chosen. If not, delete the less important ones. What you don't include can be included in the next review.

It's preferable if reviews don't contain surprises but if you find one bring it up.

Assessing poor performers

If you are assessing a poor performer, you'll need to work through the five stages of problem-solving.

Poor performers tend to ignore their problems. You’ve made progress when they actively deny the problem rather than ignoring it. 

Find facts and examples to demonstrate reality. You know you've made progress when they admit it’s a problem but maintain it's not their problem. 

If they have a problem, there’s no way of resolving it if they blame others. The most significant step is when they assume responsibility, then finding the solution is relatively easy. 

Your job is to get them to move through each stage, but finding the solution should be a shared task.

There are three possible outcomes, the report:

1. Accepts your assessment and recommended cure and commits to taking it;

2. Disagrees with your assessment but accepts your cure; or

3. Disagrees with your assessment and doesn’t commit to doing what you’ve recommended.

The first outcome is ideal, but any commitment to action is acceptable. Complicated issues don’t lend themselves to universal agreement. If your subordinate commits to change, assume they’re sincere. 

If you can't get them past the blame others stage, you need to go back to the what-when-how mode of management and get their commitment to the course of action and then monitor their performance against the commitment.

Assessing top performers

While we provide poor performers with detailed feedback, the majority of reviews for high performers make little or no attempt to define what they need to improve.

If you're assessing a top performer, you must make an effort to suggest ways to improve their performance. Remember, the purpose of a review is to improve future performance.

This is backwards. You should spend more time trying to improve the performance of your best people. After all, these people account for a disproportionate share of the organization's performance. If they get better, the impact on group output is enormous.

Not only is it incredibly high leverage, it's what the best people want. There is always room to improve and top performers crave feedback.

Other thoughts and practices

Is it a good idea to ask a report to prepare some kind of a self-review before being reviewed by you? Probably not. If you need a report to give you a list of accomplishments, you haven't done your job as a manager. The act of evaluating an employee is a formal act of leadership. Don’t get nudged by self-reviews and commit to non-bias performance reviews.

Should your report evaluate your performance? This can be a good idea, if you make it clear that it’s your job to assess their performance, and their assessment of you is only advisory. Don’t pretend you and your reports are equal during performance reviews.

Should you deliver the written review before, during, or after the face-to-face discussion? Give your report the written report before the face-to-face conversation. They can read it and will be more emotionally prepared.

Preparing and delivering performance reviews is one of the hardest tasks you'll have to perform as a manager. The best way to learn to do them well is to think critically about the reviews you have received and what made them good or bad.

Interviewing potential employees, reference checks, and talking existing employees out of quitting

Interviewing potential and talking existing employees out of quitting are two of the most difficult tasks you must perform as a manager.

The purpose of interviewing is to:

• Select a good performer;

• Educate them about you and the company;

• Determine if there is a mutual fit; and

• Sell them on the job.

As we’ve learned, assessing the performance of your team is hard. Sitting down with someone for an hour to try and find out how they'll perform in a new environment is near impossible.

But you have no real choice but to perform the interview. Just realize the chance of failure is high.

You should also research past performance by checking references. The problem is you'll often be talking to a stranger. Even if they're 100% honest, what they say won't have much meaning unless you know about how the company operates, and what values it works by.

Conducting the interview

During the interview, the applicant should do 80% of the talking, but what they talk about should be your main concern.

The good news is you have a great deal of control if you are actively listening.

When you ask a question, they may go on and on until long after you've lost interest. Most people sit and listen until the end out of courtesy.

You should interrupt. Remember, you only have an hour or so. Don't waste the only asset you have to assess potential performance.

The interview is yours to control, and if you don't, you only have yourself to blame.

You'll get the most insight if you steer the discussion toward subjects you're both familiar with.

Applicants should talk about themself, their experience, what they've done, and why, as well as past decisions they would change. Just make sure the words they use mean the same thing to both of you.

Don't use tricks. The interview should be completely straightforward. Remember, candidates are potential employees. The first impression you make will stay long after you hire them, so show yourself and the environment as it really is.

The best interview questions

When writing High Output Management, Andy asked Intel's managers for the best interview questions. These are what they provided:

• Describe some projects that were highly regarded by your management, especially by the levels above your immediate supervisor.

• What are your weaknesses? How are you working to eliminate them?

• Convince me why my company should hire you.

• What are some of the problems you are encountering in your current position? How are you going about solving them? What could you have done to prevent them from cropping up?

• Why do you think you're ready for this new job?

• What do you consider your most significant failures? What did you learn from them?

• Why do you think an engineer should be chosen for a marketing position? (Vary this one according to the situation)

• What was the most important course or project you completed in college? Why was it so important?

When a valued employee who is not motivated to leave for more money at another company quits, you have failed as a manager.

Your initial reaction is crucial. In many cases, they’re quitting because they feel unimportant to you. If you don’t deal with the situation now, you confirm their feelings. 

Drop what you are doing and ask them why they are quitting. When the employee has provided their reasons for leaving, ask more questions. After the prepared points are delivered, the real issues come out. 

You need to convey to them that what they do is essential to you, and find out what is troubling them. Don’t try to change the employee’s mind; buy time. 

An employee quitting is a significant problem, so ask your manager for help. As this impacts you more than your supervisor, it’s up to you to communicate the need for help and get them participating in a solution. 

Pursue every avenue you can to keep the employee at the company, even if it means transferring them to another department. If this happens, don’t make them feel guilty about the new arrangement. 

Your subordinate may push back and say they’ve accepted the job and can’t back out. Some people feel pressure to comply with their commitment, but they’ve made two commitments: first to the potential employer and second to you, their current employer. And commitments made to people you work with daily are more durable than those made to new acquaintances. 

If none of this works, accept it, be happy for them, and work together to ensure a smooth handover of responsibilities.

Information gathered during interviews

The information gathered from these questions tends to fall into four categories:

1. Technical knowledge: What they know about performing the job they want

2. What they did: How they performed in the past using their skills and knowledge

3. Discrepancies: Any differences between their capabilities and their performance

4. Operational values: What guides them on the job

Unlike performance reviews, interviewing is about assessing potential. Don't worry about being blunt, direct questions tend to bring direct answers, and when they don't that also reveals something about the candidate.

Asking a candidate about a hypothetical situation can also provide valuable information, and the questions they ask you can reveal a great deal about their capabilities, skills, and values.

Finally, ask them what they want to know about you, the company, and the job. The questions they ask will tell you what they already know about the company based on their research, what they want to more about, and how well prepared for the interview.

References

Reference checks are another tool you should use. When you're talking to references, you're after the same information that you tried to get from the candidate.

If you know the reference personally or can find a personal connection to them, you'll have a much better chance of getting real information. The longer you talk, the more likely you'll be able to uncover a common experience or association, which is also why the final minutes of a reference call tend to be the most fruitful.

If possible, try to talk to the applicant again after checking references.

At the end of the day, careful interviewing and reference checking doesn’t guarantee anything. It just increases your odds of getting lucky. 

Talking an employee out of quitting

Now that we've covered interviewing, it's time to address how to talk an employee out of quitting.

When a valued employee who is not motivated to leave for more money quits, you have failed as a manager.

Your initial reaction is crucial. In many cases, the employee is quitting because they feel undervalued. If you don't deal with the situation right away, you confirm their feelings.

Drop what you're doing and ask them why they are quitting. Let them talk and don't argue. After they've finished going through their reasons, ask more questions. After the prepared points are delivered, the real issues come out.

Don't argue, don't lecture, and don't panic.

You need to convey to them that what they do is essential, and find out what is troubling them. Don't try to change their mind yet, but buy time.

An employee quitting is a significant problem, so ask your manager for help. As this impacts you more than your supervisor, you need to communicate the need for help and work with them to get their participation.

Pursue every avenue you can to keep the employee at the company, even if it means transferring them to another department.

When you're ready to go back to your report with a solution, you need to make them comfortable with the new arrangement.

They might feel like they force your hand and that you'll resent them or push back and say they’ve accepted a job and can’t back out.

Some people feel pressure to comply with the new commitment, but they’ve made two commitments: first to the potential employer and second to you, their current employer. And commitments to people you work with daily are more durable than those made to new acquaintances. 

If none of this works, accept it, be happy for them, and work together to ensure a smooth handover. Other employees will watch how you treat good leavers and their morale and commitment to the company will hinge on how you treat them.

Compensation as task-relevant feedback

Money is vital at all levels of Maslow's hierarchy. People need money for food, housing, and insurance, which form the base of the pyramid. As they move up the hierarchy, money becomes a proxy for worth in a competitive environment.

There's a simple test you can use to determine the role money plays for someone.

If the absolute amount of a raise is significant, a person is motivated by physiological or safety needs. If it's a relative amount, they're likely driven by self-actualization. Money is a measure, not a necessity.

As compensation increases, each dollar provides less utility. Most knowledge workers are paid well enough that money doesn't have a material significance to them, but not well enough that it has no significance.

As a manager, your job is to elicit peak of performance from your team. You need to use money as a way to deliver task-relevant feedback.

Designing performance bonus schemes

You need to tie compensation to performance but as we've learned, assessing performance is difficult.

Base a portion of a person's compensation on performance. As total compensation increases, the bonus should grow to make up a significant portion of it. If the absolute dollar amount makes little difference, the bonus should be as high as 50 percent.

You need to determine how to weigh:

• The significance of the team's versus the individual's performance

• The period the bonus should cover; and

• Whether the bonus is based on solely measurable objectives or a mix of subjective and objective metrics

Compounding these issues is the fact that cause and effect are usually offset from each other, often by a long time, but the bonus needs to be paid as close to the work as possible so people understand why they received it.

Finally, you can't create a system that pays out lavishly when the company isn't growing.

Determining base salaries

It's equally important to think through how to determine base salary.

At one extreme, experience alone dictates salary, the other merit only. In the experience-only approach, a person's salary increases based on the time they spend in a job. In the merit-only approach, salary is independent of time.

Either way, a given job will have a maximum salary. 

In reality, neither extreme is practical so we compromise deriving salaries from a mix of experience and merit.

Deciding on who to promote

Promotions play a vital role in communicating the value system of a company and must be based solely on performance.

This is the only way to ensure performance is highlighted, maintained, and perpetuated at an organizational level.

But what about the Peter Principle, which describes the tendency for organizations to promote employees until they reach a level of incompetence?

There is no alternative. When a person receives more training and becomes more motivated, they will exceed the requirements of their position and they must be promoted where they will first only meet requirements. With more experience, they may again exceed requirements and again be promoted, and the cycle repeats itself.

People fluctuate between meeting and exceeding requirements throughout their career. If you don't promote them when they exceed requirements, they'll get bored, performance will worsen, and you'll indicate to others that performance isn't what is valued.

Recycling failed promotions

There are times when a person is promoted into a position that is too difficult for them. The solution is to not fire them but to recycle them into the position they did well before the promotion.

You know they can do the job, they’ll be excellent candidates for another promotion later, and the second time they’re promoted likely to succeed. 

Unfortunately, people view this as a personal failure when in reality, management is at fault for misjudging the employee's readiness for more responsibility. Don’t force people to leave the company because of your mistake. 

If you recycle openly, you’ll be pleasantly surprised about how smooth the transition is. 

Training employees: Why training is the boss’s job

Poorly trained employees, despite their best intentions, produce inefficiencies, excess costs, and unhappy customers.

Many managers feel that training employees is best left to others. The best managers strongly believe that they should do it themself.

Remember, as a manager, your output is equal to the output of your organization plus the neighboring organizations under your influence. Your productivity depends on eliciting more output from your team.

Recall that there are two ways to raise the level of performance. By increasing motivation, which we covered in a previous episode, or by increasing individual capability through training.

It's generally accepted that motivating employees is a key task for any manager, and one that can't be delegated to someone else. So why shouldn't the same be true for training?

Training is high-leverage

Training is one of the highest-leverage activities you can perform.

Say you spend twelve hours writing a paper on management best practices and ten people read it.

Next year, those ten people will work a total of about twenty thousand hours. If that paper results in a one percent improvement in performance, your company gains two hundred hours of additional productivity as a result of your twelve hours.

Now imagine if a hundred or a thousand managers read the paper...

Making training effective

For training to be effective, it must be a process, not an event. Employees should be able to count on it being systematic and schedule, rather than a rescue effort designed to solve a problem.

If you accept that training, along with motivation, is the way to improve performance, and what you teach must be closely tied to what you practice, and that it must be a process rather than an one-off event, it becomes clear who must do the training: you, the manager.

You should instruct your team and perhaps a few ranks below them. Your subordinates should do the same, as should every manager at every level of the organization.

There is another reason why you must fill the role of the teacher. Training must be done by a suitable role model. Proxies, no matter how good they are at teaching cannot assume the role. The person teaching should be seen as a believable, practicing authority on the subject.

Developing training programs

Understand that there are two different types of training. The first is focused on teaching new employees what they need to do their jobs. The second is to teach new ideas, principles, or skills to existing team members.

Make a list of the things you feel your team should be trained on. Don't limit the scope, items can range from simple (like how to format a document) to abstract (like the objectives and values of the company).

Don't forget to ask people what they feel they need. You'll be surprise by what they tell you.

Once you've got your list, assign priorities among the items.

If you haven't done training in the past, start small. Create one short course on the most urgent subject.

You'll find that things you find easy are much harder to explain than to practice.

Just get started, create an outline, develop the first part of the course, and go. Develop the second part once you've delivered the first. Accept that the first time you teach something, it won't be that good.

To avoid causing damage, teach the course to your most knowledgable team members who can help shape it.

Asking for feedback

Always ask for anonymous feedback. Prompt people for numerical ratings and pose open-ended questions. Study and consider the responses, but understand that you'll never be able to please everyone.

If you receive feedback that the content was too detailed, too superficial, or just right, in about equal balance, you’re on the right track. 

What matters is that you accomplish what you set out to do.

Learning from teaching

Training is hard. Preparing materials and getting yourself ready to answer questions is difficult. Even if you're great at your job, you'll be amazed at how much you don't know.

It takes a lot more knowledge to teach than to do, but guess who learns the most? You, and that alone is worth the effort.

Putting what you’ve learned into practice

Answer the following questions. If you reach at least 100 points, you’ll be a better manager.

Production

• Identify operations in your work akin to process, assembly, and test production. (10 points)

• Identify the limiting step in a project you are working on and map out the production flow. (10 points)

• Define where you should perform the equivalents of receiving inspection, in-process inspection, and final inspection in your work. Decide whether these inspections should be monitoring steps or gate-like. Identify where you can move to variable review. (10 points)

• Identify six indicators that measure the quantity and quality of your group’s output. (10 points)

• Check these indicators daily and regularly review them in staff meetings. (20 points)

• What is the most important strategy you are pursuing now? Describe the environmental demand that creates it and your current status. Is your plan likely to result in a positive outcome for you or your organization if successful? (20 points)

Leverage

• Simplify your most tedious, time-consuming task. Eliminate at least 30 percent of the steps involved. (10 points)

• Define your output: What are the outputs of the organization you manage and the organizations you can influence? List them in order of importance. (10 points)

• Analyze your information-and knowledge-gathering system. Does it properly balance breadth and depth? Is redundancy built-in? (10 points)

• Outline how you’ll monitor the next thing you delegate. What will you look for, and how frequently? (10 points)

• Generate an inventory of projects you can work on when times are slow. (10 points)

• Have a one-on-one with each direct report. (Explain to them in advance what a one-on-one is and have them prepare.) (20 points)

• Look at your calendar for the last week. Classify your activities as low-/medium-/high-leverage. What will you stop doing so you have more time for more high-leverage activities (10 points)

• Forecast the demand on your time for the next week. How much time will you spend in meetings? Which of these are process-oriented vs. mission-oriented meetings? If the latter is over 25 percent of your total time, what should you do to reduce them? (10 points)

• Define the three most important objectives for your team for the next three months. Support them with key results. (20 points)

• Have your team do the same for themselves, after a thorough discussion of the set generated above. (20 points)

• Generate an inventory of pending decisions you are responsible for. Take three and structure the decision-making process for them, using the six-question approach. (10 points)

• Evaluate where you are in Maslow’s hierarchy. Do the same for each of your subordinates. (10 points)

• Define a set of performance indicators for each direct report. (20 points)

• List the task-relevant feedback your subordinates receive. How well can they gauge their progress through them?

• Classify the task-relevant maturity of each of your subordinates as low, medium, or high. Which management style is most appropriate for each? Compare what your own style is with what it should be. (10 points)

• Evaluate your last performance review and the last set of reviews you gave to your team to deliver task-relevant feedback. Did these reviews improve performance? (20 points)

• Redo one of your reviews as it should have been done. (10 points)

Additional resources

If you found this post interesting, I recommend reading High Output Management in its entirety. Below are some additional resources you may find interesting. 

Books

• The Score Takes Care of Itself, by Bill Walsh

• High Growth Handbook, by Elad Gil

• 7 Powers, by Hamilton Helmer

• Essentialism, by Greg McKeown

• Deep Work, by Cal Newport

• Creative Selection, by Ken Kocienda

• The Innovator's Dilemma, by Clayton M. Christensen

• The Innovator's Solution, by Clayton M. Christensen

• Andrew Carnegie, by David Nasaw

• The Outsiders, by William N. Thorndike

• Creativity Inc., by Ed Catmull

• Valley of Genius, by Adam Fisher

• Dealers of Lightning, by Michael Hilztik

• The Dream Machine, by M. Mitchell Waldrop

• Loonshots, by Safi Bahcall

• The Everything Store, by Brad Stone

• The Hard Thing About Hard Things, by Ben Horowitz

• Only the Paranoid Survive, by Andrew S. Grove

• Measure What Matters, by John Doerr

Posts

• How to Operate with Keith Rabois

• 7 Powers: The Foundations of Business Strategy by Hamilton Helmer

• Andy

• Maker's Schedule, Manager's Schedule

• How To Be Successful

Videos

Thanks to Jordan Hughes for creating the images you see in this post, check out his projects Good Books and Untitled UI. And thanks to Anna Cheng and Guy Proops who reviewed drafts and provided feedback that made this post much better than it would have been.

“Silicon Valley correctly places enormous value on execution and on culture. However, I think this sometimes leads to insufficient importance being placed on strategy. Hamilton Helmer’s deeply incisive work will hopefully help correct that.” – Patrick Collison, CEO and Co-Founder of Stripe.

Execution is what helps us find product/market fit. To get there, we follow Superhuman’s lead and listen to those who have come before to tell us when we’ve arrived.

We echo John Doerr’s famed line: Ideas are easy. Execution is everything.

But what we often overlook is product/market fit is necessary, but not sufficient, to build an enduringly valuable business. Operational excellence alone does not ensure success.

So, what are the secrets to making a company enduringly valuable?

If we are to believe Hamilton Helmer, the answer is strategy and execution. Every celebrated business is underpinned by decisive strategy choices and operational excellence, in the midst of uncertainty, which leads to some form of competitive advantage.

And what is strategy? To borrow from Nathan Baschez introducing Divinations: Every strategy is really just a theory: “We bet if we do x, then y will happen.” 

7 Powers by Hamilton Helmer provides real-world examples grounded in decades of experience as a strategy advisor, active equity investor, and Stanford University professor that help us build a set of mental models to think through strategy.

The ideas in 7 Powers cannot, and were never intended to, replace operational excellence. But rather to create a prepared mind that can point toward enduring value.

Strategy as an idea may be easy. But we should never confuse simple with simplistic.

Reed Hastings, CEO and Co-founder of Netflix, echoes the importance of strategy in his foreword to 7 Powers:

Most of my time and that of everyone else at Netflix must be spent achieving superb execution. Fail at this, and you will surely stumble. Sadly, though, such execution alone will not ensure success. If you don’t get your strategy right, you are at risk.

Beyond Reed and Patrick, other fans of 7 Powers include:

• Spotify CEO and Co-Founder - Daniel Ek;

• Entrepreneur and investor - Peter Thiel;

• Former CEO of Adobe - Bruce Chizen; and

• Chairman of Sequoia Capital - Mike Moritz.

The crux of 7 Powers is a business can try to improve its strengths, mitigate its weaknesses, eliminate competitor risk, better serve its customers, maximize shareholder value, or take advantage of its pricing power.

But there are very few times when a business can do all at once.

7 Powers provides a comprehensive strategic framework that can help every business choose what to focus on next.

Strategy Statics: Defining the 7 Powers

We begin with our central concept, the seven Powers. Each Power provides a benefit to its owner and a barrier to competitors.

Power 1: Scale Economies

• Definition: A business where per unit costs decline as volume increases

• Benefit: Reduced costs

• Barrier: Prohibitive costs of share gains

Power 2: Network Economies

• Definition: A business where the value realized by a customer increases as the userbase increases

• Benefit: Ability to charge higher prices or monetize more due to additional value created

• Barrier: Hard to gain market share as users don’t want to switch due to the lower value provided

Power 3: Counter Positioning

• Definition: A business adopts a new, superior business model that incumbents cannot mimic due to the anticipated cannibalization of their existing business

• Benefit: Lower costs and/or higher prices due to more valuable product

• Barrier: Cannibalization of existing business

Power 4: Switching Costs

• A business where customers expect a greater loss than the value they gain from switching to an alternate

• Benefit: Ability to charge higher prices for the same product

• Barrier: Competitor has to compensate the customer to switch

Power 5: Branding

• Definition: A business that enjoys a higher perceived value to an objectively identical offering due to historical information about them

• Benefit: Ability to charge higher prices due to perceived higher quality or reduced uncertainty

• Barrier: The significant time and uncertainty needed to build a brand

Power 6: Cornered Resource

• Definition: A business that has preferential access to a coveted resource that independently enhances value

• Benefit: Ability to charge higher prices, reduce costs, or create better products due to access to a cornered resource

• Barrier: Ranges from property and patent law to personal preference, e.g. retention of key talent

Power 7: Process Power

• Definition: A business whose organization and activity set enables lower costs and/or superior products that can only be matched by an extended commitment

• Benefit: Improved product and/or lower costs due to superior process

• Barrier: The significant time and/or investment needed to create the process

Power 1: Scale economies

• Scale Economies: A business where per unit costs decline as volume increases

• Benefit: Reduced costs

• Barrier: Prohibitive costs of share gains

Scale economies are straightforward, well-known, and often overlooked.

If your business has a low fixed cost base, it can benefit enormously from scale. This is why many technology companies try to get big fast. Not for scale itself but because being big is one of the best ways to protect their profits.

Hamilton uses Netflix and their move toward original content as his example of a scale economy.

Prior to creating their own content, Netflix had to negotiate the rights to each TV show or film on a case-by-case basis, often multiple times across geographies, each with changing terms. This is part of the reason why the US Netflix and the Australian Netflix have different content.

Netflix was at the mercy of its suppliers who could, and would, charge more based on the increasing number of subscribers Netflix had. In short, Netflix’s cost base wasn’t fixed.

Contrast this with Netflix Originals. With Originals, Netflix pays a flat fee that can be divided across its global subscriber base who will all get access to the same content, on the same day, regardless of geography.

And Netflix never has to pay another cent once the content is made.

The more subscribers the company amasses and the higher it can push its pricing, the more content it can produce – which in turn drives more subscribers, more engagement and more pricing power.

This flywheel is endemic to SVOD, but unique in the history of television. Linear television networks have always been bound by a finite number of primetime slots and the size of the total primetime audience.

Accordingly, a network with 15 viable slots would only add a new show (say #16 or #22) if it replaced one of its existing 15; the only thing that matters is relative outperformance.

Netflix faces none of linear’s limitations – any series that can meet the company’s target cost per hour watched contributes to its penetration and engagement (as do those that don’t, albeit less efficiently).

Source: Netflix Isn't Being Reckless

So that’s our Benefit, but what’s our Barrier? Scale is a chicken and egg problem. Netflix had to get big enough that it made sense to create its own content rather than leasing it from others, but in order to get big, it needed content from other providers.

Contrast Netflix, which has ~207 million subscribers, to Stan who has 2.3 million.

If both companies were to produce an original with a budget of $100 million, Netflix pays ~48 cents per subscriber while Stan pays $43.47 per subscriber.

If Stan wants to compete with Netflix’s original content, they not only have to make the large initial outlay to produce the content, but they also have a far smaller base of users to spread the cost across.

Even if they were willing to do that, Netflix’s superior cost advantage allows them to spend more on marketing, improving their user experience, hiring more people, or just undercutting the competition.

As you can see, Netflix’s move toward original content significantly improves its margins and creates a compelling moat that is not easily emulated.

If you want to dive deeper into why Originals are an important strategic decision for Netflix, I highly recommend Matthew Ball’s posts:

• Part 1 explains that Netflix spends far more on content than is typically reported.

• Part 2 explains how (and why) Netflix uses product and technology to economically outspend its competitors. 

• Part 3 explains why Netflix risks so much. 

• Part 4 explains why the term ‘Original Series’ is often a lie – and how Netflix uses this fact to beat its competitors. 

• Part 5 explains why 2019 and 2020 don’t represent significant threats to Netflix despite the volume of new entrants and their impact on Netflix’s library.

• Part 6 explains that quality in SVOD is a distraction, if the concept is even real.

• Part 7 explains why Netflix has been so resilient over the past decade – and why this is likely to continue even as competition intensifies.

• Part 8 explains why Netflix won't be stopped by recession, bond markets, debt, or cash losses

Not to mention…

Source: Screenshot Essays

Power 2: Network economies

• Network Economies: A business where the value realized by a customer increases as the userbase increases

• Benefit: Ability to charge higher prices or monetize more due to additional value created

• Barrier: Hard to gain market share as users don’t want to switch due to the lower value provided

If you work in technology, there’s a good chance you’re already familiar with network economies. They’re behind many of the most successful technology businesses including PayPal, Facebook, and Twitter.

That said, many people continue to confuse virality with network economies. This is because they’re often seen together, but don’t necessarily have to be. You can go viral without having a network effect and vice versa.

A network economy is a business where the value realized by a customer increases as the userbase increases.

Facebook is a canonical example of a service that benefits from network economies.

As the user base grows, you’re able to connect with more of your friends.

We all use Facebook because we all use Facebook. This is why Facebook’s major focus, in the beginning, was retention. It’s why they rolled out school by school, and why the entire Facebook product experience focused on getting users to seven friends in ten days.

Once a user had seven friends, they were much less likely to churn and thus added value to Facebook’s network.

More importantly, these strong network effects have made it near impossible for anyone to compete with a generalized social network.

The idea of leaving Facebook to join an alternative network that most of your friends aren’t using reduces the potential value of the alternative, even if it’s a better experience.

This is why network economies often lead to winner-take-all dynamics.

Compare this to virality, virality means each user that signups up invites more users. There are many examples of viral games on the Facebook platform that quickly declined in popularity because the additional users provided no additional value and the existing users churned.

The most important thing to understand is businesses can be viral and not have a network effect. Virality can disappear and churn can eat away at the user base.

I highly recommend diving deep into network effects by reading this a16z deck and listening to Anu Hariharan from Y Combinator (who wrote the a16z deck prior to moving to Y Combinator).

Power 3: Counter positioning

• Counter Positioning: A business adopts a new, superior business model that incumbents cannot mimic due to the anticipated cannibalization of their existing business

• Benefit: Lower costs and/or higher prices due to more valuable product

• Barrier: Cannibalization of existing business

Counter positioning is when a new business adopts a new, superior business model that incumbents can't mimic due to anticipated damage to their existing business. It’s similar to Clayton Christensen’s disruption theory but lacks the emphasis on moving from the low to the high end of the market.

Counter positioning tends to come from outsiders and startups rather than existing market leaders. This is because the business environment does not allow market leaders to pursue disruptive innovations when they arise for one of the following reasons:

• The total addressable market looks small;

• It's difficult to see if the total addressable market will grow (and by how much); or

• Their current cost structures are based on serving their existing market and not what is required to sell the disruptive technology, which may have lower margins.

When I first read about counter positioning, the first story I thought of was Kodak and the invention of the digital camera in 1975. I was wrong, digital cameras were disruptive technology to film, but they aren’t an example of failure to adopt counter positioning. Let’s dig into why.

Kodak of all places was best positioned to be the leader in digital cameras, after all, they were the leader in film. But what Hamilton (and Clayton) offer us is an opposing explanation. The reason Kodak wasn’t able to react was precisely because of its expertise and existing value chain.

Kodak as a business was designed to profit from selling film cameras that needed an endless supply of film and services like film development. Adopting digital would have done collateral damage to Kodak’s business, but would not offer lower costs and/or higher prices due to the reduced costs for consumers and if Kodak could adopt it so could its competitors.

As Hamilton told me over email:

for film there was no attractive business model (for media) in digital – so it goes down the first left branch in the decision tree: Stand-Alone Attractive - No. This is another example as well as to why there is many-to-many mapping of CP to Christensen’s disruptive technologies (that is to say they mean entirely different things). Digital photography was disruptive but not CP.

Pair this with the fact that film cameras were used by professionals while digital cameras were seen as toys.

But we shouldn’t forget what Aaron Harris wrote in Toy Markets:

Investors should know how important it is to bet on toy markets. They’re the ones who have learned, again and again, that the biggest companies start out looking like toys. Amazon was for selling books online when relatively few people used the internet. Google was a search engine in a landscape crowded with search engines that weren’t themselves gigantic businesses. eBay was for selling beanie babies.

Either way, digital cameras didn’t need film and photos didn’t necessarily need to be developed, at least not in the way that Kodak was used to. This meant that if Kodak were to develop a better digital camera, they would cannibalize their existing profits while offering a less profitable business in the future.

Things that look like toys improve and often become good enough for professional users. But Kodak’s management probably knew this.

Do you take the money and your best people away from what makes you money (film) to work on something that will likely make you far less in the future (digital) when your customers aren't currently using it?

Or do you keep profiting from what you've got until it's gone?

It’s a difficult decision to make even for the best executive.

Now let’s contrast Kodak to Vanguard, a real example of counter positioning.

Vanguard took on the world of active investing with its low-cost passive index funds. While it’s easy to spout the benefits of low-cost passive investing today, it was a highly contrarian approach.

Why would anyone accept the average returns of the market instead of choosing an active equity investor who wanted to beat the market?

Well, it turns out, it is exceptionally rare for an active fund to beat the market over the long-term. As this became apparent, Vanguard saw its assets under management balloon.

And while competitors could have easily introduced their own low-cost passive index funds, they were extremely slow to do so, for the exact same reasons that Kodak had for not adopting the digital camera. It made them less money and their cost base wasn’t set up to support it.

Power 4: Switching costs

• Switching Costs: A business where customers expect a greater loss than the gain from switching to an alternate

• Benefit: Ability to charge higher prices for the same product

• Barrier: Competitor has to compensate the customer to switch

Switching costs occur when it’s easier to stay with a product or service than it is to switch (even if the alternative is objectively better). Additional products, features, integrations, consulting, and training can make it even harder to switch.

This is part of the reason why many companies provide universities with free access to their software to teach students. It means when they get to the workforce they’ve already learned the software and would incur switching costs when they have to learn an alternative. Examples of this include Adobe, Matlab, Mathematica, and Atlassian products at Australian universities.

In 7 Powers, Hamilton uses the example of SAP which over time becomes the backbone of organizations. I personally think Salesforce is a more interesting example.

Once you rely on Salesforce’s CRM, it’s often not worth switching platforms. It can take months to retain your workforce and costs hundreds of thousands or even tens of millions of dollars to recreate business logic on the new platform.

Beyond CRM, Salesforce now has fourteen other products that when used, make it even harder to switch.

A key thing to consider is that switching costs often increase with the number of people who rely on it. They’re never that large for a single person but when an organization has hundreds or thousands of users on a certain platform, it can mean thousands of hours of productivity lost.

Even if a new solution is 10% better, it may not make sense to switch.

This is a big reason for vendor lock-in and why enterprise software can be terrible, overpriced, and have outdated UX - but still succeed.

Other reasons include:

• The person who uses the software is not the person buying the software. Managers prefer to go with the tried and tested rather than something new that may be better – they don't want to risk their reputation on a new shiny solution. Nobody ever got fired for buying IBM.

• The software is optimized for compliance, complicated workflows, permission management, etc and the user experience takes a back seat.

• No one is in charge of re-evaluating the old software solutions, so companies just keep paying for what they have.

Power 5: Branding

• Branding: A business that enjoys a higher perceived value to an objectively identical offering due to historical information about them

• Benefit: Ability to charge higher prices due to perceived higher quality or reduced uncertainty

• Barrier: The significant time and uncertainty needed to build a brand

Brand is arguably the most powerful and least understood of the 7 Powers. It’s so durable because of how complicated, time-intensive, and lucky you have to build a brand.

If you look at Warren Buffett's Berkshire Hathaway portfolio you see a lot of powerful brands: American Express, Wells Fargo, Apple, Heinz, Coca Cola, Bank of America, Goldman Sachs, IBM, and John & Johnson.

The key thing to understand is many of these brands sell commoditized goods. There is nothing different between them and their competitors beyond the company or product name.

With brands, customers are paying for consistent experience and years (even decades) of consistent messaging.

Tiffany & Co, the American luxury jewelry retailer, is a classic example of brand. The brand is so strong people want to buy used Tiffany's boxes on eBay.

When Tiffany’s introduced the diamond engagement ring in 1886, the Tiffany Blue Box became as desired as what was inside.

When you buy Tiffany’s, you’re not just buying jewelry, you’re buying over 100 years of consistent advertising that says when your partner buys you Tiffany’s they love you.

It’s an incredibly powerful growth loop, the more people who know the brand, the more they expect to see it, and the more people want it.

In the past, the Tiffany Blue Box meant better placements in malls and the store itself acted as a channel for new people to discover Tiffany’s.

One caveat to this is that the Internet may be changing how brands are developed and how powerful brand is as a Power. Just look at how quickly Airbnb was able to build a brand that often surpasses the likes of Hilton and other famous hotel brands.

Placement and brand may not be as important in the Internet’s infinite mall with unlimited shelf space. Near zero cost of distribution means new ways to build brands that can be much faster than traditional methods, see Email Addresses and Razor Blades from Stratechery for more on this topic.

Or, what I think is more likely, is that brand becomes more important than ever before. You only have to look at Hermès or LVMH to understand why.

Power 6: Cornered Resource

• Cornered Resource: A business that has preferential access to a coveted resource that independently enhances value

• Benefit: Ability to charge higher prices, reduce costs, or create better products due to access to a cornered resource

• Barrier: Ranges from property and patent law to personal preference, e.g. retention of key talent

A company has a cornered resource when it has preferential access to limited resources or talent.

A good example for talent is Amazon, who according to The Information, has a 17-person senior leadership team, called the S-team, who have an average tenure of 15 years – the majority of the company's 23-year lifespan.

The S-Team has a deeper understanding of what it takes to scale a business like Amazon’s than anyone else in the world and have shared almost two decades of experience.

Other famous examples:

• Apple's design team led by Jony Ive;

• The PayPal Mafia; and

• Pixar's Braintrust (this cornered resource is exactly why Disney bought Pixar!).

Power 7: Process power

• Process Power: A business whose organization and activity set enables lower costs and/or superior products that can only be matched by an extended commitment

• Benefit: Improved product and/or lower costs due to superior process

• Barrier: The significant time and/or investment needed to create the process

Process Power is one of the hardest powers to copy, Toyota is famously transparent about the Toyota Production System (TPS) processes but other companies haven’t been able to replicate it. I suspect some of the struggle to copy TPS is because most companies haven’t read the work of W. Edwards Deming and or the books he influenced like Understanding Variation.

The key difference between Process Power and operational excellent is hysteresis, even if you know what to do it takes significant time and effort to replicate.

Toyota’s Kanban just-in-time manufacturing system allows any worker from Janitor to CEO to stop the production line if they see something dangerous or something that could be improved.

Despite Toyota providing tours and creating a joint venture with GM, they still couldn’t replicate the Process Power of Toyota’s Japanese factories.

While checklists are great, see The Checklist Manifesto, Process Power is embedded in the organization’s culture. It’s learned by osmosis and feels natural to employees.

To outsiders, it looks like magic – and it sort of is.

Other good examples of Process Power are Instagram and WhatsApp, both were worth billions of dollars with less than 60 full-time employees, 13 employees for Instagram, and 55 employees for WhatsApp.

Dynamics: Getting to Power

“Hamilton Helmer understands that strategy starts with invention. He can’t tell you what to invent, but he can and does show what it takes for a new invention to become a valuable business.”
– Peter Thiel, entrepreneur and investor

The path to Power begins with invention. To understand why we return to Netflix.

When Netflix decided to take on streaming, they had Powers in their DVD-rental business (counter positioning and process power) that wouldn’t transfer over to streaming.

And to develop the streaming offering they had to license content which increased in cost as subscribers increased. There was no path to creating something that would produce persistent differential returns.

But thanks to invention, the streaming service we know today, Netflix created compelling value and had one Power (counter positioning) that allowed them to get to scale. Scale then allowed them to shift their attention to developing original content that was exclusive and produced at a fixed cost leading to a second power, a scale economy.

Over time, we may see Netflix develop a powerful brand and process power around their original programming too.

In general:

• Change creates new threats and opportunities;

• The company invents a new product or service; and

• Once the new product creates compelling value for customers, the company can develop a strategy (a route to Power).

What I take from 7 Powers is it's as important to decide what to work on as it is to work hard. Netflix would never have become Netflix without its counter positioning and scale economy.

Power Progression: When can you establish a Power?

There are three distinct phases when a company can craft a strategy to create Power:

• Origination: Before creating compelling value counter positioning and cornered resource can be created.

• Takeoff: The rapid growth phase that takes place when compelling value is attained is when scale economies, network economies, and switching costs can be attained.

• Stability: After growth slows brand and process power become the name of the game.

Summary

Hamilton develops strategy from first principles and uses specific nomenclature throughout 7 Powers that is useful to walk away with:

• Power: the central concept of the book, is the set of conditions creating the potential for persistent differential returns. Power is what makes a business durably valuable and is created if a business line is simultaneously superior (improves free cash flow), significant (the cash flow improvement is material), and sustainable (the improvement is largely immune to competitive arbitrage).

• Strategy (with a capital S): The intellectual discipline, is the study of what conditions create the potential business value.

• strategy (with a lowercase s) or The Mantra: The specific approach of a single business, is a route to continuing Power in significant markets.

The notion of Power (and the impact of its lacking) is what underlies Warren Buffett’s famed line: When a manager with a reputation for brilliance tackles a business with a reputation for bad economics, the reputation of the business remains intact.

Like John Doerr, Warren is correct for his stage of investing. A lack of Power in public companies is rarely fixed by brilliant management.

At the same time, brilliant management is often required to create Power in the first place.

This is why Strategy as an intellectual discipline must be further broken down into two underlying topics:

1. Statics–i.e. “Being There”: Statics are what makes a business durably valuable, the specific conditions that result in materially improved cash flow, while simultaneously inhibiting competitive arbitrage.

2. Dynamics–i.e. “Getting There”: Dynamics are the decisions that result in the creation or attainment of Power.

This is a good place to flag that Hamilton’s definition of strategy, the specific approach to a single business, is considerably narrower than the traditional definition of strategy. It is solely focused on creating a persistent Power, Power that leads to persistent differential returns.

It’s important to understand that the fundamental assumption behind good strategy is unchanging, differential margins. This is because the bulk of a business’ value comes in the future. For high-growth companies, this reality is accentuated. Remember, Power is reserved for conditions that create durable differential returns.

Each Power is associated with a magnitude and duration:

• Benefit: The conditions created by Power must materially augment cash flow. It can manifest as any combination of increased prices, reduced costs, and/or lessened investment needs. This is the magnitude aspect of Power.

• Barrier: The Benefit must not only augment cash flow, it must also persist. There must be some aspect that prevents existing and potential competitors from engaging in value-destroying arbitrage. This is the duration aspect of Power.

Benefits are common, but alone bear little positive impact on company value as they are generally subject to arbitrage. This is why Power must have a Barrier to prevent such arbitrage.

Power is a relative concept. It’s about your advantages relative to specific competitors. Good strategy involves assessing Power with respect to each competitor, including potential, direct, and functional equivalents.

Remember, Power is not a panacea. All the Power in the world won’t save you from an industry that is rapidly contracting.

Additional resources

If you found this post interesting, I recommend reading 7 Powers in its entirety. Below are some additional resources you may find interesting. 

Books

• High Output Management, by Andy Grove (summary and review)

• High Growth Handbook, by Elad Gil

• The Score Takes Care of Itself, by Bill Walsh

• Essentialism, by Greg McKeown

• Deep Work, by Cal Newport

• Scale, by Geoffrey West

• Creative Selection, by Ken Kocienda

• The Innovator's Dilemma, by Clayton M. Christensen

• The Innovator's Solution, by Clayton M. Christensen

• Andrew Carnegie, by David Nasaw

• The Innovators, by Walter Isaacson

• American Prometheus, by Kai Bird and Martin J. Sherwin

• Titan, by Ron Chernow

• The Outsiders, by William N. Thorndike

• Creativity Inc., by Ed Catmull

• Valley of Genius, by Adam Fisher

• Dealers of Lightning, by Michael Hilztik

• The Facebook Effect, by David Kirkpatrick

• In the Plex, by Steven Levy

• Behind the Cloud, by Marc Benioff

• The Dream Machine, by M. Mitchell Waldrop

• The House of Morgan, by Ron Chernow

• Loonshots, by Safi Bahcall

• The Everything Store, by Brad Stone

• Superforecasting, by Philip E. Tetlock and Dan Gardner

• Expert Political Judgement, by Philip E. Tetlock

• Thinking in Bets, by Annie Duke

• The Book of Why, by Judea Pearl and Dana Mackenzie

• The Elephant in the Brain, by Robin Hanson and Kevin Simler

• Competitive Strategy, by Michael Porter

Podcasts

Videos

Thanks to Jordan Hughes for creating the images you see in this post, check out his projects Good Books and Untitled UI. And thanks to Anna Cheng, Guy Proops, and Jack Walsh who reviewed drafts and provided feedback that made this post much easier to read.

Keith is a general partner at Founders Fund. Throughout his venture career, he has led investments in DoorDash, Affirm, Forward Health, Faire, Thoughtspot, and Stripe. He also founded Opendoor, which transforms the process of buying and selling a home in the United States.

Keith has a unique and unparalleled track record as an entrepreneur, executive, and investor. He started his career in Silicon Valley by working with Peter Thiel at PayPal and has subsequently worked with Reid Hoffman at LinkedIn, Jack Dorsey as Square's former COO, and Vinod Khosla at Khosla Ventures.

This post is my attempt to summarise his How to Operate talk at Stanford University's How to Start a Startup class.

Stanford CS183B: How to Operate (Keith Rabois)

Build an engine

Creating a company is harder than creating a product for the basic reason that people are irrational.

Somebody in your life, probably somebody close to you is irrational.

Building a company is basically taking all the most irrational people you know and concentrating them in the same place for hours a day.

What you want to do is build an engine. The first engine you build won’t be pretty, it might not even work. That’s why people work so hard at the start to bring something together.

Eventually, you’ll construct a high performance machine that nobody has to worry about.

We used to joke about at eBay, that if the Martians took over eBay it would take six months for the world to notice. That’s eventually what you want to get to.

The end goal is to, as Peter Lynch said, "go for a business that any idiot can run, because sooner or later, any idiot probably is going to run it."

I recommend watching David Sacks’ talk called The Cadence to learn more about engine building.

The role of a leader

There is only one book that outlines how to lead.

High Output Management by Andy Grove, says the role of a leader is to maximize the output of the organization, meaning "the output of a manager is the output of the organizational units under his or her supervision or influence."

Read my summary of High Output Management here

The CEO is responsible for everything, and a VP is accountable for their part of the organization, as well as the organizations around them.

This means a VP of Engineering is responsible for the performance of not only the engineering team, but also product and marketing as they have influence there.

The inverse is also true.

You want to measure your people with a focus on output and not input.

Don't measure motion and confuse it with progress. Spend time judging the quality of the ideas, not whether you can move revenue 3x this quarter.

While this might sound glamorous and people get excited about managing a whole large organization, it's about being responsible for the output. More on this later.

Editing: simply and don't accept complexity

One of the most essential things Keith learned at Square was the concept of editing.

This is the best metaphor I have ever seen in 14 years of running stuff, of how to think about your job. It’s a natural metaphor, so it’s easy to take with you everyday and it’s easy to transmit to each of your employees so they can figure out if they are editing or writing.

Whenever you’re doing something, ask yourself: Am I writing or am I testing?

As a manager, your primary job is to edit.

The first thing most editors do is take out a red pen, eliminating useless words and phrases, and omitting things that aren’t important.

That’s your job too. The more you simplify, the better your organization will run.

It's hard to keep track of and understand a complicated set of initiatives.

The end goal is to distill the business down to one, two, or three things and teach people to think about it through that framework so they can repeat it without thinking.

Simplify. Don't accept complexity.

You can change the world in 140 characters.

You can build the most important companies in history with a very simple to describe concept.

You can market products in less than 50 characters.

There is no reason you can’t build your company the same way. So force yourself to simplify every initiative, every product, every marketing, everything you do.

Basically take that red and start eliminating stuff.

Ask clarifying questions

You need to ask a lot of questions, and they can be simple questions like, "should we try this seven days a week? Or six days?" or fundamental questions like "where's our competitive advantage?"

But the best thing you can do is find the one, two, three, or four things that matter most and focus there. That'll allow you to make decisions rapidly and won't distract you from your real job. Building the company.

Allocating resources

People who work with you should be coming up with their initiatives.

Your end goal is to use less red ink every day.

A good way to measure that is to see how well you are communicating with your colleagues about what’s important and what’s not.

It’s okay to have a bad day with red ink all over the place, but you shouldn’t have a month where you’re editing more than the previous one.

Consistent voice

Your job is to ensure a consistent voice. When you pick up The Economist, there is one voice.

Every article feels like the same person wrote it. Ideally, your company should feel the same, whether it's the marketing site, press release, or the physical packaging on your product.

It should feel like one person wrote it.

Over time, you don't want to be doing all the consistent voice editing yourself. You want to train people so they can recognize it themselves. Stripe is world-class at this.

It may be helpful to read Style Guide by The Economist too.

Delegating

Editors aren’t the ones writing the publication. That will be true for your company too, the way you get most of the work done is by delegating.

When you delegate, you are responsible for everything.

The CEO, founder, there is no excuse.

There is no, there is that department over there, this person over there screwed up.

You are always responsible for every single thing, especially when things go wrong.

So how do you both delegate but not abdicate? 

Keith is a fan of the management technique called Task Relevant Maturity. All that means is, "has this person ever done this before?"

The more mature they are in a task, the more rope you give them. If someone is new to something, you're going to instruct and continuously monitor them. And if they've done it before you're going to delegate and trust them to get it done.

Your employee dictates your management style.

It's actually a good thing if you do reference checks on somebody and half the people you call say they are a micromanager and the other half say they actually give me a lot of responsibility.

That’s a feature not a bug.

When to delegate

Keith learned how to delegate from Peter Thiel.

You want to draw up a two by two matrix that has your level of conviction about a decision on one axis from extremely low to extremely high and the other axis with the consequence of the decision from little impact to catastrophic.

If the decision is a low consequence and you have little confidence in your own opinion, you should delegate and delegate completely, let people make mistakes and learn.

And if the consequences are dramatic and you have incredibly high conviction you are right, you shouldn't delegate. But you should explain your thinking about why you are making the decision.

Okay, you're the boss

When Keith was at LinkedIn, he had a talented colleague who would occasionally get annoyed if he didn't agree with his opinion on something.

So Keith would spend his time trying to persuade him about why he was making the decision a certain way.

And the colleague's trump card would be, "Okay, you're the boss."

And that to me was like I was burning a lot of social capital.

Every time he said that I knew I was creating a really thin line and ultimately that was going to backfire if I did that too often.

You want to track the times that you are doing that.

Let people make mistakes

An example of this is at Square, one of my favourite people in the world and my second hire, first marketing hire, had this program he wanted to run called Inner Square which allowed Square merchants to give out, imagine a food truck outside put out ten Squares on the counter and people could just grab them.

And Kyle had this great idea that this would be an awesome marketing program. Squares would spread Squares to other people and to some extent it was on brand. So it didn't have catastrophic consequences.

Each of these ten Squares didn't cost that much money, so financially we could afford to do it.

But at that time, my ten years of experience said it was not going to work on a meaningful enough scale for our metrics and I preferred not to do it.

Kyle was so excited about this that I decided to just let him do it. He learned that when you measure this thing, it's not massive. It doesn't create massive value for the company.

It did require a fair amount of operational complexity to ship all these Squares to people and figure out how to get them, etc, etc. But it allowed him to be excited about his job and to learn how to filter future ideas.

So it was totally worth letting him make the “Mistake."

Editing the team

It's impossible to have a perfect team, and you definitely won't have one at the start.

Keith has a good analogy for hiring that he calls barrels and ammunition.

When most people hire a lot of people, they expect that more people = more horsepower = higher velocity of shipping.

But it turns out it doesn't generally work that way. When you hire more engineers, you won't automatically get more done. The same is true for designers, business people, and customer success staff.

And that's because "most great people are actually ammunition. But what you need in your company are barrels. And you can only shoot through the unique barrels you have."

When you stock your barrels with ammunition, you increase the velocity of the company. Barrels are tough to find. When you have them give them lots of equity, promote them, take them out to dinner, because they are irreplaceable and culturally specific.

A barrel at one company may not be a barrel at another company.

Finding barrels

A barrel is a person who can take an idea from conception through to shipping and bring people along the journey with them. And that's a culturally specific skillset.

How can you tell who is a barrel and who is not? One is you start with a very small set of responsibilities, it can be very trivial.

It can be something like, I want to reward the engineers in my office at nine o'clock every night with a nice cold, fresh smoothie. This is actually a real example.

I was frustrated, our engineers were working really hard, and maybe 20%, 30% would stay late in the evening and we had already served them dinner but I wanted to give them something cool to reward them. You can think about alcohol but that's a little complicated.

So smoothies were probably a little bit better than pizza, which drains you of energy.

But nobody could get smoothies to show up in my office at nine o'clock sharp, that were cold, that tasted good, and that were delivered in the right place that the engineers would find them.

You would think this is simple but in fact it took two months to get this done. So we had an intern start, and I think on his second day I was explaining this problem, and he said, well I will do it.

And I was looking at him like there was no way.

I have seen my office manager fail, my assistant fail, who were actually pretty good. This just isn't going to happen.

And low and behold they show up. On time, cold, delivered at the right place, and my first instinct was great. Nothing about the smoothies, but now I can actually give him something more important that is more complicated to do.

If you've hired someone and people go up to their desk, especially if they don't report to them, it's a sign they believe the person can help them.

If you see this consistently, those are your barrels.

Expand the scope of responsibility

The story above is exactly what you want to do with every single employee, every single day until they can't handle any more responsibility.

And what you want to do is keep expanding it until you see where it breaks and that's the role they should stay in. That level of sophistication.

But some people will surprise you. There will be some people that you do not expect.

With different backgrounds, without a lot of experience that can just handle enormously complicated tasks. So keep testing that and pushing the envelope.

Only talk to Peter Thiel about your number one priority

Peter Thiel would insist everyone at PayPal could only do one thing. He would say, "I will not talk to you about anything else besides this one thing I assigned you."

The insight here is that people will only try to solve problems they know how to solve.

No one wants to bang their head against the wall day after day. And this cascades down the organization, so no one is solving the most important thing.

Try giving everybody one thing to prioritize.

Don't make all decisions yourself

Give people the tools to make decisions at the same level you would make them.

Build a dashboard that simplifies the company's value proposition into a set of metrics that the whole company can rally behind.

Then measure what fraction of your employees are using that dashboard every day.

If you've done your job well, it should be close to 100%. This dashboard needs to be as intuitive to your internal users as your product is to external users.

Pairing indicators

When you measure one thing the company tends to optimize towards that. Often at the expense of something important.

The example Keith uses is fraud rates. It's easy for the risk team to lower the fraud rate by treating every user as a suspect.

Then you have the lowest fraud rate in the world, you also have the lowest level of customer satisfaction score.

In this example, you want to pair the fraud rate with your false positive rate. This forces the team to innovate.

With your hiring team, you might pair the number of interviews with the quality of people being interviewed.

Look for anomalies

You don't actually want to look for the expected behaviour.

So a famous example was at PayPal. None of the top ten markets that the company was planning on going after included eBay.

One day, someone noticed that 54 of the sellers actually handwritten into their eBay listings, please pay me with PayPal and brought this to the attention of the executive team at the time.

The first reaction from the executive team was, what the hell is going on? Let's get them out of the system, that is not the focus.

Fortunately, David Sacks came back the next day and said, I think we found our market. Let's actually build tools for these power sellers instead of forcing them to write into their listing, pay me with PayPal. Why don't we just have an HTML button that they can just insert? And that actually worked.

Then he thought, why should we have them insert it each time? Why don't we just automatically insert it for them? They can just insert it once, then every listing they have forever will have it automagically appear there.

So that became the success for PayPal.

Be transparent

Metrics are the first step. Everyone should have access to what's going on.

The other thing Keith likes to do is take everyone through board decks and review every single slide after each board meeting. If you can remember the feedback you got from your board, you should pass that on too.

Every meeting should have notes that are sent around to the entire company. This means everyone can keep track of what is interesting and not feel excluded.

When Keith was at Square, every company room had glass walls so people could see exactly who was in the meeting and who was meeting who.

At Stripe, everyone has access to everyone's emails.

The Score Takes Care of Itself

The other book Keith mentions in the lecture is The Score Takes Care of Itself, by Bill Walsh.

The point of the book is if you can get all the small things right, you don't have to worry about the big things.

The big things are a byproduct of what you do every day to make the details excellent. As an example, Walsh wrote a three-page memo on how to answer the phone.

And that may sound absurd but what his point was organisation as a whole does everything exactly the right way.

Every detail matters.

That is the sort of detail obsession that building this sort of company requires.

Examples that may be a bit more practical for you instead of circuit boards may be, what sort of foods do you serve people?

It actually matters more than you might guess. When people don't like the food you serve them, what do they do?

They go gossip, they go complain to their friends, they walk over to someones desk. Then all of a sudden that lunch that they are complaining about is what they are spending most of their time gossiping and complaining instead of brainstorming.

You don't have this serendipities idea matching another serendipities idea that creates a spark instead they are all wallowing and whittling around.

The best thing you can do is give people the food they want or the food that’s good for them that makes them more productive.

Ending thoughts

My favorite quote of the talk:

So it may seem like this glorious job you thought you had is more like running around being a TaskRabbit for people.

But it is to take things off their plate that is a distraction so they can be high performance machines.

And if you take enough things away from people to distract them and give them the tools to be successful, all of a sudden your organisation produces a lot more.

Additional resources

These are resources that are related to Keith's talk. Hopefully, you find something useful.

More lessons from Keith Rabois

• How to become a Venture Capitalist

• How to Interview an Executive

• How to be an Effective Executive

• How to run an Effective Board Meeting and make an Effective Board Deck

• How to Become a Magnet for Talent and Assess Talent

• How Keith Rabois Builds Billion-Dollar Businesses: Radical Simplicity

• Keith Rabois on the Role of a COO, How to Hire and Why Transparency Matters

• Genius.com's full transcript of How to Operate

Irrationality

This is a super interesting topic. I recommend you get started with the following books:

• We Learn Nothing, by Tim Kreider

• Thinking, Fast and Slow, by Daniel Kahneman

• Nudge, by Richard H. Thaler and Cass R. Sunstein

• Misbehaving, by Richard H. Thaler

• Influence, by Robert B. Cialdini

• Fooled by Randomness, by Nassim Nicholas Taleb

• The Art of Thinking Clearly, by Rolf Dobelli

• Predictably Irrational, by Dan Ariely

• Rationality, by Eliezer Yudkowsky

• Superforecasting, by Philip Tetlock and Dan Gardner

• Hooked, by Nir Eyal

• Expert Political Judgement, by Philip E. Tetlock

• Thinking in Bets, by Annie Duke

• The Book of Why, by Judea Pearl and Dana Mackenzie

• The Elephant in the Brain, by Robin Hanson and Kevin Simler

Company building

I recommend reading the following books to get a deeper understanding of what Keith is talking to:

• Andrew Carnegie, by David Nasaw

• The Innovators, by Walter Isaacson

• American Prometheus, by Kai Bird and Martin J. Sherwin

• Titan, by Ron Chernow

• The Outsiders, by William N. Thorndike

• Creativity Inc., by Ed Catmull

• Valley of Genius, by Adam Fisher

• Dealers of Lightning, by Michael Hilztik

• The Facebook Effect, by David Kirkpatrick

• In the Plex, by Steven Levy

• Behind the Cloud, by Marc Benioff

• Who Is Michael Ovitz?, by Michael Ovitz

• The Dream Machine, by M. Mitchell Waldrop

• The House of Morgan, by Ron Chernow

• Loonshots, by Safi Bahcall

• Shoe Dog, by Phil Knight

• Hatching Twitter, by Nick Bilton

• The Everything Store, by Brad Stone

• The Hard Thing About Hard Things, by Ben Horowitz

Management and work

These are books that have changed the way I think about management and work.

• High Output Management, by Andy Grove

• High Growth Handbook, by Elad Gil

• The Goal, by Eliyahu M. Goldratt and Jeff Cox

• The Score Takes Care of Itself, by Bill Walsh

• Essentialism, by Greg McKeown

• Deep Work, by Cal Newport

• The Inner Game of Tennis, by W. Timothy Gallwey

• The Art of Learning, by Josh Waitzkin

• Flow, by Mihaly Csikszentmihalyi Ph. D.

• Skin in the Game, by Nassim Nicholas Taleb

• Why We Sleep, by Matthew Walker

• Ego is the Enemy, by Ryan Holiday

• The Obstacle is the Way, by Ryan Holiday

• Getting Things Done, by David Allen

• Scale, by Geoffrey West

• Sapiens, by Yuval Noah Harari

• Creative Selection, by Ken Kocienda

• Atomic Habits, by James Clear

• Extreme Ownership, by Jocko Willink and Leif Babin

• How Will You Measure Your Life, by James Allworth, Clayton M. Christensen, and Karen Dillon

• Grit, by Angela Duckworth

• Principles, by Ray Dalio

• The Innovator's Dilemma, by Clayton M. Christensen

• The Innovator's Solution, by Clayton M. Christensen

• 7 Powers, by Hamilton Helmer (review)

2024

1. The Snowball, by Alice Schroeder

2. Kochland, by Christopher Leonard

3. Understanding Variation, by Dr. Donald J. Wheeler

4. Murder at First Principles, by Ben Wiener

5. The Almanack of Naval Ravikant, by Eric Jorgenson

6. How to Make a Few Billion Dollars, by Brad Jacobs

7. Slow Productivity, by Cal Newport

8. Measures of Success, by Mark Graban

9. HBR Guide to Buying A Small Business, by Richard S. Ruback and Royce Yudkoff

10. Dream Big, by Cristiane Correa

11. The Algebra of Wealth, by Scott Galloway

12. Nick and Zak’s Adventures in Capitalism, by The Rational Cloner

13. Digital Minimalism, by Cal Newport

14. Actionable Gamification, by Yu-Kai Chou

15. The Gambler, by William C. Rempel

16. What It Takes, by Stephen A. Schwarzman

17. Never Enough, by Andrew Wilkinson

18. Men and Rubber, by Harvey S. Firestone

19. The Lessons of History, by Will and Ariel Durant

20. The Engines That Move Markets, by Alasdair Nairn

21. Am I Being Too Subtle?, by Sam Zell

22. What the CEO Wants You to Know, by Ram Charan

23. Unreasonable Hospitality, by Will Guidara

24. Playing with Reality, by Kelly Clancy

2023

1. How to Decide, by Annie Duke

2. The Definitive FDR, by James MacGregor Burns

3. How the World Really Works, by Vaclav Smil

4. 33 Strategies of War, by Robert Greene

5. The Founders, by Jimmy Soni

6. Outlive, by Peter Attia and Bill Gifford

7. The Creative Act, by Rick Rubin

8. Quit, by Annie Duke

9. Elon Musk, by Walter Isaacson

10. The Rebel Allocator, by Jacob Taylor

11. Chip War, by Chris Miller

12. A Man for All Markets, by Edward O. Thorp

13. Same as Ever, by Morgan Housel

2022

1. Dopamine Nation, by Dr Anna Lembke

2. The Power Law, by Sebastian Mallaby

3. Lessons from the Titans, by Scott Davis, Carter Copeland, and Rob Wertheimer

4. Rest, by Alex Soojung-Kim Pang

5. Tao of Charlie Munger, by David Clark

6. Build, by Tony Fadell

7. Four Thousand Weeks, by Oliver Burkeman

8. $100M Offers, by Alex Hormozi

9. The Dhandho Investor, by Mohnish Pabrai

10. Discipline Is Destiny, by Ryan Holiday

11. Hackers & Painters, by Paul Graham

2021

1. Exercised, by Daniel Lieberman

2. The Molecule of More, by Daniel Lieberman and Michael Long

3. The Story of the Human Body, by Daniel Lieberman

4. Obviously Awesome, by April Dunford

5. Atomic Habits, by James Clear

6. Tiny Habits, by BJ Fogg

7. The Box, by Marc Levinson

8. The Design of Everyday Things, by Don Norman

9. Indistractable, by Nir Eyal and Julie Li

10. Obviously Awesome, by April Dunford

11. Livewired, by David Eagleman

12. Think Again, by Adam Grant

13. The Art of Impossible, by Steven Kotler

14. Man’s Search for Meaning, by Viktor E. Frankl

15. Genome, by Matt Ridley

16. How Innovation Works, by Matt Ridley

17. Mastery, by George Leonard

18. Working Backwards, by Colin Bryar and Bill Carr

19. The Sovereign Individual, by James Dale Davidson and William Rees-Mogg

20. The Defining Decade, by Meg Jay

21. The Blind Watchmaker, by Richard Dawkins

22. Amazon Unbound, by Brad Stone

23. Project Hail Mary, by Andy Weir

24. The Scout Mindset, by Julia Galef

25. Storyworthy, by Matthew Dicks

26. 48 Laws of Power, by Robert Greene

27. Wanting, by Luke Burgis

28. Product-Led Growth, by Wes Bush

29. The Great CEO Within, by Matt Mochary, Alex MacCaw, and Misha Talavera

30. To Pixar and Beyond, by Lawrence Levy

31. Great by Choice, by Jim Collins and Morten T Hansen

32. The Contrarian, by Max Chafkin

33. The Culture Code, by Daniel Coyle

34. The Millonaire Fastlane, by MJ DeMarco

35. The Daily Laws, by Robert Greene

36. A World Without Email, by Cal Newport

37. Courage is Calling, by Ryan Holiday

38. How the Mighty Fall, by Jim Collins

39. Radical Candor, by Kim Scott

40. The Cold Start Problem, by Andrew Chen

41. Richer, Wiser, Happier, by William Green

42. Red Notice, by Bill Browder

2020

1. The Joy of Movement, by Kelly McGonigal

2. The Ride of a Lifetime, by Robert Iger

3. A Gentleman in Moscow, by Amor Towles

4. The Botany of Desire, by Michael Pollan

5. Why Nations Fail, by Daron Acemoglu and James A. Robinson

6. Algorithms to Live By, by Brian Christian and Tom Griffiths

7. That Will Never Work, by Marc Randolph

8. Tim Cook, by Leander Kahney

9. Hacking Darwin, by Jamie Metzl

10. Breakfast of Champions, by Kurt Vonnegut

11. Pre-Suasion, by Robert Cialdini

12. Mastering the Market Cycle, by Howard Marks

13. Facebook, by Steven Levy

14. Surprise, Kill, Vanish, by Annie Jacobsen

15. Let My People Go Surfing, by Yvon Chouinard and Naomi Klein

16. Setting the Table, by Danny Meyer

17. The Checklist Manifesto, by Atul Gawande

18. Solve for Happy, by Mo Gawdat

19. Working, by Robert A. Caro

20. A Bigger Picture, by Malcolm Turnbull

21. The Wedge, by Scott carney

22. Breath, by James Nestor

23. The Path to Power, by Robert A. Caro

24. Personal History, by Katharine Graham

25. Cable Cowboy, by Mark Rocihaux

26. Tools of Titans, by Tim Ferriss

27. 7 Powers, by Hamilton Helmer (summary)

28. High Output Management, by Andy Grove (summary)

29. Genghis Khan and the Making of the Modern World, by Jack Weatherford

30. The Biggest Bluff, by Maria Konnikova

31. Good Strategy/Bad Strategy, by Richard Rumelt

32. Means of Ascent, by Robert A. Caro

33. Mastery, by Robert Greene

34. Talent Is Overrated, by Geoff Colvin

35. No Rules Rules, by Reed Hastings and Erin Meyer

36. The Psychology of Money, by Morgan Housel

37. Tuxedo Park, by Jennet Conant

38. Tribal Leadership, by Dave Logan, John King, and Halee Fischer-Wright

39. Hooked, by Nir Eyal and Ryan Hoover

40. How Asia Works, by Joe Studwell

41. A Promised Land, by Barack Obama

42. Invent and Wander, by Jeff Bezos and Walter Isaacson

43. Snow Crash, by Neal Stephenson

44. Delivering Happiness, by Tony Hsiesh

45. The Alchemist, by Paulo Coelho

46. Good to Great, by Jim Collins

47. Beyond Entrepreneurship 2.0, by Jim Collins

48. The 7 Habits of Highly Effective People, by Stephen R. Covey

49. Inspired, by Marty Cagan

50. The Charisma Myth, by Olivia Fox Cabane

51. Black Box Thinking, by Matthew Syed

52. Guns, Germs and Steel, by Jared Diamond

2019

1. Superhuman, by Rowan Hooper

2. The Laws of Human Nature, by Robert Greene

3. Little Black Stretchy Pants, by Chip Wilson

4. How to Get Rich, by Felix Dennis

5. Digital Minimalism, by Cal Newport

6. The Messy Middle, by Scott Belsky

7. Loonshots, by Safi Bahcall

8. Death’s End, by Cixin Liu

9. What Doesn’t Kill Us, by Scott Carney

10. The Sports Gene, by David Epstein

11. Leading, by Alex Ferguson and Michael Moritz

12. The Return to the Little Kingdom, by Michael Moritz

13. Endure, by Alex Hutchinson

14. Reboot, by Jerry Colonna

15. Disciplined Entrepreneurship, by Bill Aulet

16. Range, by David Epstein

17. The First 90 Days, by Michael Watkins

18. Trillion Dollar Coach, by Eric Schmidt, Jonathan Rosenberg,
    and Alan Eagle

19. 12 Rules of Life, by Jordan B. Peterson

20. Alchemy, by Rory Sutherland

21. I Am a Strange Loop, by Douglas R. Hofstadter

22. Super Thinking, by Gabriel Weinberg and Lauren McCann

23. Educated, by Tara Westover

24. The Language Instinct, by Steven Pinker

25. Talking to Strangers, by Malcolm Gladwell

26. Lifespan, by Dr David A. Sinclair and Matthew D. LaPlante

27. Super Pumped, by Mike Isaac

28. Stillness Is The Key, by Ryan Holiday

29. Capitalism Without Capital, by Jonathan Haskel and Stian Westlake

30. High Performance Habits, by Brendon Burchard

31. Can’t Hurt Me, by David Goggins

32. The Untethered Soul, by Michael A. Singer

33. I Will Teach You to Be Rich, by Ramit Sethi

34. What It Takes, by Stephen A. Schwarzman

35. Loserthink, by Scott Adams

36. The Man Who Solved the Market, by Gregory Zuckerman

37. How to Fail at Almost Everything and Still Win Big, by Scott Adams

38. The Challenger Sale, by Matthew Dixon and Brent Adamson

39. Dopesick, by Beth Macy

40. Disrupted, by Daniel Lyons

41. Mistakes Were Made (But Not By Me), by Carol Tavris and Elliot Aronson

42. American Kingpin, by Nick Bilton

2018

1. Michael Jordan, by Roland Lazenby

2. Deep Work, by Cal Newport

3. The Art of Thinking Clearly, by Rolf Dobelli

4. The Art of Learning, by Josh Waitzkin

5. The Inner Game of Tennis, by W. Timothy Gallwey

6. Predictably Irrational, by Dan Ariely

7. Superforecasting, by Philip Tetlock

8. Rationality, by Eliezer Yudkowsky

9. A Moveable Feast, by Ernest Hemingway

10. Born to Run, by Christopher McDougall

11. Flow, by Mihaly Csikzentmihalyi

12. The Fighter’s Mind, by Sam Sheridan

13. Chop Wood Carry Water, by Joshua Medcalf

14. For Whom the Bell Tolls, by Ernest Hemingway

15. Why Moats Matter, by Elizabeth Collins

16. Modern Monopolies, by Alex Moazed

17. The Human Brand, by Chris Malone

18. Titan, by Ron Chernow

19. The Hero with a Thousand Faces, by Joseph Campbell

20. Learn Better, by Ulrich Boser

21. Stealing Fire, by Steven Kotler and Jamie Wheal

22. Skin in the Game, by Nassim Nicholas Taleb

23. Beyond the Goal, by Eliyahu M. Goldratt

24. Ready Player One, by Ernest Cline

25. When Breath Becomes Air, by Paul Kalanithi

26. Hooked, by Nir Eyal

27. Drive, by Daniel H. Pink

28. The Power of Habit, by Charles Duhigg

29. Peak Performance, by Brad Stulberg and Steve Magness

30. Brave New World, by Aldous Huxley

31. Surfaces and Essences, by Douglas Hofstadter and Emmanuel Sander

32. Why We Sleep, by Matthew Walker

33. Deep Thinking, by Garry Kasparov

34. Incognito, by David Eagleman

35. The Gene, by Siddhartha Mukherjee

36. Play On, by Jeff Bercovici

37. How to Change Your Mind, by Michael Pollan

38. Origin Story, by David Christian

39. Ego is the Enemy, by Ryan Holiday

40. Why Information Grows, by Cesar Hidalgo

41. Bad Blood, by John Carreyrou

42. Measure What Matters, by John Doerr

43. Factfulness, by Hans Rosling

44. So Good They Can’t Ignore You, by Cal Newport

45. The Three-Body Problem, by Cixin Liu

46. The Outsiders, by William N. Thorndike

47. Creativity Inc., by Ed Catmull

48. The Dark Forest, by Cixin Liu

49. Altered Traits, by Richard Davidson and Daniel Coleman

50. Own the Day, Own Your Life, by Aubrey Marcus

51. Thinking in Bets, by Annie Duke

52. Expert Political Judgement, by Philip E. Tetlock

53. The Courage to Be Disliked, by Fumitake Koga and Ichiro Kishimi

54. Valley of Genius, by Adam Fisher

55. A Guide to the Good Life, by William B. Irvine

56. The Barefoot Investor, by Scott Pape

57. The Spider Network, by David Enrich

58. Dealers of Lightning, by Michael Hiltzik

59. High Growth Handbook, by Elad Gil

60. Getting Things Done, by David Allen

61. Frenemies, by Ken Auletta

62. The Facebook Effect, by David Kirkpatrick

63. In the Plex, by Steven Levy

64. Scale, by Geoffrey West

65. Dune, by Frank Herbert

66. Subscribed, by Tien Tzuo and Gabe Weisert

67. The Selfish Gene, by Richard Dawkins

68. The Talent Code, by Daniel Coyle

69. Behind the Cloud, by Marc Benioff

70. 21 Lessons for the 21st Century, by Yuval Noah Harari

71. The Book of Why, by Judea Pearl

72. The Beginning of Infinity, by David Deutsch

73. Angel, by Jason Calacanis

74. Who Is Michael Ovitz?, by Michael Ovitz

75. Creative Selection, by Ken Kocienda

76. The Warren Buffett Way, by Robert Hagstrom

77. The Success Equation, by Michael J. Mauboussin

78. Atomic Habits, by James Clear

79. Blitzscaling, by Reid Hoffman and Chris Yeh

80. Conspiracy, by Ryan Holiday

81. The Mindful Athlete, by Phil Jackson

82. The Elephant in the Brain, by Robin Hanson and Kevin Simler

83. The Dream Machine, by M. Mitchell Waldrop

84. The House of Morgan, by Ron Chernow

85. The Revolt of the Public, by Martin Gurri

86. Extreme Ownership, by Jocko Willink and Leif Babin

87. Eat Move Sleep, by Tom Rath