Insights from High Output Management


High Output Management is clearly a great book for leaders but it’s overrated to be the bible book for middle management. I know some director made it a point that all managers in his org study this book and hold regular meetings to share their learning. I think this practice is problematic in itself, regardless of what book they study.

That being said, I like the topics it discussed about typical manager’s tasks day to day. I’ve summarized with comments in this post the insights that

  1. Resonate most with me
  2. Not obvious to most people
  3. Are core to leadership

Enabling the team


The physiological, safety/security, and social needs all can motivate us to show up for work, but other needs — esteem and self-actualization — make us perform once we are there. At the upper level of the need hierarchy, when one is self-actualized, money in itself is no longer a source of motivation but rather a measure of achievement. Money in the physiological- and security-driven modes only motivates until the need is satisfied, but money as a measure of achievement will motivate without limit.

Majority of the tech workers are esteem driven. This insight will put the motivation task in perspective, which is the most important task for leadership. In a nutshell, leadership is about enabling, which translates to motivation and coaching(see below). Tasks like team organization design are means for motivation.

When a person is not doing his job, there can only be two reasons for it. The person either can’t do it or won’t do it; he is either not capable or not motivated.

This should be the beginning when thinking about how to fix performance problems. It’s much more efficient than starting with how to find an interesting project(he might be not capable).

In almost all such cases, the employee is quitting because he feels he is not important to you. If you do not deal with the situation right at the first mention, you’ll confirm his feelings and the outcome is inevitable.

I think employees quit for four reasons:

  • They are not capable to do the job
  • They can’t find any sponsor for their success in the company
  • The managers didn’t appreciate their performance enough by promotion, bonus, etc
  • They’ve got more appealing offers

The first three should be fixed in regular 1:1 rather than last minute. When the employee already decided to quit, it’s not practical to persuade her otherwise.

The move from blaming others to assuming responsibility constitutes an emotional step, while the move from assuming responsibility to finding the solution is an intellectual one, and the latter is easier.

As no living human beings ever truly believe they are wrong(at least we should assume so), it’s extremely hard for anyone to still admit his flaw in a framework that doesn’t hurt the ego. That’s why most people stay in denial mode forever and end up a cynic.

1:1 meetings

As a rule of thumb, a manager whose work is largely supervisory should have six to eight subordinates; three or four are too few and ten are too many. This range comes from a guideline that a manager should allocate about a half day per week to each of his subordinates.

I find the so-called ‘flat’ team is hard for individual contributors to be successful. Because everyone needs a sponsor(ideally also a mentor) to succeed. A single manager simply can’t provide enough sponsorship for 20 people. So the majority of such teams are stuck in mediocrity.

A 1:1 should last an hour at a minimum. Anything less, in my experience, tends to make the subordinate confine himself to simple things that can be handled quickly.

All companies I worked at have the tradition of weekly or bi-weekly 30 minutes 1:1. To make it worse, weak managers dominate this 30 minutes for project management or technical discussion. 1:1 should be mostly driven by subordinates. The goal should be exclusively about providing transparency for the IC:

  • What’s the current trajectory of perf rating?
  • How is IC’s performance evaluated based on what they do?
  • What’s the expectation from leadership?
  • The areas for IC’s development
  • The plan and execution for the IC’s career growth

Other topics are wrong topics for 1:1.

Performance review

There is also a time factor to consider. The subordinate’s output during the review period may have all, some, or nothing to do with his activities during the same period. Accordingly, the supervisor should look at the time offset between the activity of the subordinate and the output that results from that activity.

Performance review should be based on the performance of the reviewed period, not the indicators of the reviewed period. Many indicators lag more than one year, so shouldn’t be used in perf.

In my experience, the best thing to do is to give your subordinate the written review sometime before the face-to-face discussion.

None of my managers ever did this. Maybe it’s company policy. But I think giving subordinates time to read the perf result will make the discussion most efficient. The subordinates will have time to compose the questions to ask.

Building culture

No single managerial activity can be said to constitute leadership, and nothing leads as well as example. By this I mean something straightforward. Values and behavioral norms are simply not transmitted easily by talk or memo, but are conveyed very effectively by doing and doing visibly.

No action communicates a manager’s values to an organization more clearly and loudly than his choice of whom he promotes. By elevating someone, we are, in effect, creating role models for others in our organization.

A great guidance about how to promote culture or behavior norms in a team. For example, engineering excellence should be taught by role models(leaders) or promotion.

As a manager, you should remind yourself that each time an insight or fact is withheld and an appropriate question is suppressed, the decision-making process is less good than it might have been. A related phenomenon influences lower-level people present in the meeting. This group has to overcome the fear of being overruled, which might mean embarrassment: if the rest of the group or a senior-level manager vetoed a junior person or opposed a position he was advocating, the junior manager might lose face in front of his peers. This, even more than fear of sanctions or even of the loss of job, makes junior people hang back and let the more senior people set the likely direction of decision-making.

A very good practice is for senior leaders to ask the questions that junior people probably don’t know but are too intimidated to ask, especially when the question seems basic or even naive. This will usually encourage everyone to participate more actively.

A supervisor should never use staff meetings to pontificate, which is the surest way to undermine free discussion and hence the meeting’s basic purpose.

As a rule, leaders should never criticize in public. This is the best way to de-motivate employees and force them to leave the job. See Larry Page’s bad example.

One of the reasons why people are reluctant to come out with an opinion in the presence of their peers is the fear of going against the group by stating an opinion that is different from that of the group. Consequently, the group as a whole wanders around for a while, feeling each other out, waiting for a consensus to develop before anyone risks taking a position. If and when a group consensus emerges, one of the members will state it as a group opinion (“I think our position seems to be…”), not as a personal position. After a weak statement of the group position, if the rest of the mob buys in, the position becomes more solid and is restated more forcefully. Note the difference between the situation described earlier by the auto executive and the one John describes. In the former instance, the people were expected to wait for their supervisor to state his opinion first. In the latter, members of the group were waiting for a consensus to develop. The dynamics are different, but the bottom line in both is that people didn’t really speak their minds freely. That certainly makes it harder for a manager to make the right decisions.

People invest a great deal of energy and emotion in coming up with a decision. Then somebody who has an important say-so or the right to veto it may come across the decision later. If he does veto it, he can be regarded as a Johnny-come-lately who upsets the decision-making applecart. This, of course, will frustrate and demoralize the people who may have been working on it for a long time. If the veto comes as a surprise, however legitimate it may have been on its merits, an impression of political maneuvering is inevitably created. Politics and manipulation or even their appearance should be avoided at all costs.

Project Management

Indicators(metrics) is a key tool for managers to understand what’s going on. There should be at least one indicator for each possibility to defend against.

Important stuff should have metrics for its health and metrics measuring if it’s over done.

I have to confess that the information most useful to me, and I suspect most useful to all managers, comes from quick, often casual verbal exchanges. This usually reaches a manager much faster than anything written down. And usually the more timely the information, the more valuable it is.

This is just human nature. Our system 1 is easily dominated by information more available. This is carefully demonstrated in Thinking, fast and slow.

Information gathering is the foundation of a manager’s work and they should spend large chunk of time doing it.

I think this is true for all work.

Keep in mind that a meeting called to make a specific decision is hard to keep moving if more than six or seven people attend.

This is especially true if before the meeting, the driver doesn’t know how each person thinks. Such meetings usually end up as pure information sharing in the good cases or huge chaos in bad ones.

I tend to avoid driving consensus with more than two parties at the same time, unless the decision is very straightforward and uncontroversial. Trying to win an argument is stupid enough. Trying to win an argument with multiple players is incompetent communication skill.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store