High Output Management

by Andrew S. Grove

A second principle applies to the frequency with which you check your subordinates’ work. A variable approach should be employed, using different sampling schemes with various subordinates; you should increase or decrease your frequency depending on whether your subordinate is performing a newly delegated task or one that he has experience handling.

To use quality assurance principles effectively, the manager should only go into details randomly, just enough to try to ensure that the subordinate is moving ahead satisfactorily. To check into all the details of a delegated task would be like quality assurance testing 100 percent of what manufacturing turned out.

First, we must identify our limiting step: what is the “egg” in our work? In a manager’s life some things really have to happen on a schedule that is absolute.

if we determine what is immovable and manipulate the more yielding activities around it, we can work more efficiently.

A second production principle we can apply to managerial work is batching similar tasks. Any manufacturing operation requires a certain amount of set-up time. So for managerial work to proceed efficiently, we should use the same set-up effort to apply across a group of similar activities.

set aside a block of time and do a batch of them together, one after the other, to maximize the use of the mental set-up time needed for the task.

What is the medium of a manager’s forecast? It is something very simple: his calendar.

You should say “no” at the outset to work beyond your capacity to handle.

It is important to say “no” earlier rather than later because we’ve learned that to wait until something reaches a higher value stage and then abort due to lack of capacity means losing more money and time.

Remember too that your time is your one finite resource, and when you say “yes” to one thing you are inevitably saying “no” to another.

allow slack—a bit of looseness in your scheduling.

A manager should carry a raw material inventory

this inventory should consist of things you need to do but don’t need to finish right away—discretionary projects, the kind the manager can work on to increase his group’s productivity over the long term.

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. (Two days a week per subordinate would probably lead to meddling; an hour a week does not provide enough opportunity for monitoring.)

anyone who spends about a half day per week as a member of a planning, advisory, or coordinating group has the equivalent of a subordinate.

Interruptions—The Plague of Managerial Work

The next important production concept we can apply to managerial work is to strive toward regularity.

we should try to make our managerial work take on the characteristics of a factory, not a job shop.

the same blocks of time must be used for like activities.

if you can pin down what kind of interruptions you’re getting, you can prepare standard responses for those that pop up most often. Customers don’t come up with totally new questions and problems day in and day out,

many interruptions that come from your subordinates can be accumulated and handled not randomly, but at staff and at one-on-one meetings,

If such meetings are held regularly, people can’t protest too much if they’re asked to batch questions and problems for scheduled times, instead of interrupting you whenever they want.

To make something regular that was once irregular is a fundamental production principle,

The two basic managerial roles produce two basic kinds of meetings. In the first kind of meeting, called a process-oriented meeting, knowledge is shared and information is exchanged. Such meetings take place on a regularly scheduled basis. The purpose of the second kind of meeting is to solve a specific problem. Meetings of this sort, called mission-oriented, frequently produce a decision.

At Intel we use three kinds of process-oriented meetings: the one-on-one, the staff meeting, and the operation review.

there is an enormous difference between a casual encounter by a supervisor and a subordinate, or even a meeting (mission-oriented) to resolve a specific problem, and a one-on-one.

how do you decide how often somebody needs such a meeting? The answer is the job- or task-relevant maturity of each of your subordinates.

you should have one-on-ones frequently (for example, once a week) with a subordinate who is inexperienced in a specific situation and less frequently (perhaps once every few weeks) with an experienced veteran.

I feel that a one-on-one 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.

Where should a one-on-one take place? In the supervisor’s office, in the subordinate’s office, or somewhere else? I think you should have the meeting in or near the subordinate’s work area if possible.

A key point about a one-on-one: It should be regarded as the subordinate’s meeting, with its agenda and tone set by him.

The supervisor with eight subordinates would have to prepare eight times; the subordinate only once. So the latter should be asked to prepare an outline, which is very important because it forces him to think through in advance all of the issues and points he plans to raise.

The meeting should also cover anything important that has happened since the last meeting: current hiring problems, people problems in general, organizational problems and future plans, and—very, very important—potential problems. Even when a problem isn’t tangible, even if it’s only an intuition that something’s wrong, a subordinate owes it to his supervisor to tell him, because it triggers a look into the organizational black box.

“The good time users among managers do not talk to their subordinates about their problems but they know how to make the subordinates talk about theirs.”

A real time-saver is using a “hold” file where both the supervisor and subordinate accumulate important but not altogether urgent issues for discussion at the next meeting.

If the supervisor uses a set schedule for a one-on-one, such as every second Wednesday morning, and if the subordinate’s vacation happens to fall on that date, the meeting is not going to occur. By scheduling on a rolling basis, this can be easily avoided.

Ninety minutes of your time can enhance the quality of your subordinate’s work for two weeks, or for some eighty-plus hours, and also upgrade your understanding of what he’s doing. Clearly, one-on-ones can exert enormous leverage.

To digress a bit, I also think that one-on-ones at home can help family life. As the father of two teenage daughters, I have found that the conversation in such a time together is very different in tone and kind from what we say to each other in other circumstances. The one-on-one makes each of us take the other seriously and allows subtle and complicated matters to come up for discussion.

Staff meetings also create opportunities for the supervisor to learn from the exchange and confrontation that often develops. In my own case, I get a much better understanding of an issue with which I am not familiar by listening to two people with opposing views discuss it than I do by listening to one side only.

a group discussion on any subject tended to get more detailed and more heated, but always more rewarding, than an exchange between me and one other specialist.

What should be discussed at a staff meeting? Anything that affects more than two of the people present. If the meeting degenerates into a conversation between two people working on a problem affecting only them, the supervisor should break it off and move on to something else that will include more of the staff, while suggesting that the two continue their exchange later.

It should be mostly controlled, with an agenda issued far enough in advance that the subordinates will have had the chance to prepare their thoughts for the meeting.

the supervisor’s most important roles are being a meeting’s moderator and facilitator, and controller of its pace and thrust.

The format here should include formal presentations in which managers describe their work to other managers who are not their immediate supervisors, and to peers in other parts of the company. The basic purpose of an operation review at Intel is to keep the teaching and learning going on between employees several organizational levels apart—

As a rule of thumb, I would recommend four minutes of presentation and discussion time per visual aid, which can include tables, numbers, or graphics.

before calling a meeting, ask yourself: What am I trying to accomplish? Then ask, is a meeting necessary? Or desirable? Or justifiable? Don’t call a meeting if all the answers aren’t yes.

An estimate of the dollar cost of a manager’s time, including overhead, is about $100 per hour. So a meeting involving ten managers for two hours costs the company $2,000. Most expenditures of $2,000 have to be approved in advance by senior people—like buying a copying machine or making a transatlantic trip—yet a manager can call a meeting and commit $2,000 worth of managerial resources at a whim.

even if you’re just an invited participant, you should ask yourself if the meeting—and your attendance—is desirable and justified. Tell the chairman—the person who invited you—if you do not feel it is. Determine the purpose of a meeting before committing your time and your company’s resources. Get it called off early, at a low-value-added stage, if a meeting makes no sense,...

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

Decision-making is not a spectator sport, because onlookers get in the way of what needs to be done.