Effectiveness Is a Habit

The Core Insight

Peter Drucker consulted to executives for twenty years and never met a natural. The effective ones ran from extroverts to the morbidly shy, and one of them had no more personality than a frozen mackerel. All they had in common was getting the right things done.

That observation carries the book. Effectiveness is a habit, a complex of practices, and practices can be learned. A knowledge worker counts as an executive when his position or his knowledge makes him responsible for results that reach the whole organization.

Knowledge work is why the habit is required. Manual work needs efficiency and produces a countable thing like a pair of shoes. A knowledge worker produces ideas, useless until another knowledge worker turns them into his own output. Nobody supervises that closely, so he directs himself and gets judged by results.

Most organizations answer weak performance by raising ability, and write job specifications that describe a universal genius. Drucker points out that the only person in abundant supply is the universal incompetent. When you cannot increase the supply of a resource, you increase its yield. You will never play Mozart the way Arthur Schnabel does, and nothing stops you from playing your scales that way.

The Framework

The executive works against four realities he does not control, and each one pushes toward nonresults.

• Your time belongs to everybody else, and everyone who can move in on it does.
• You keep operating unless you act, so the flow of events picks the work for you.
• Your output counts only when other people use it, and most of them do not report to you.
• You sit inside the organization, and all the results are outside it.

The fourth reality gets worse with size and with computers, which handle only quantifiable data. Inside events quantify easily, and an organization holds effort centers and costs alone. The United States escaped thalidomide because one physician noticed a minor skin tingling. Every quantitative figure said the Ford Edsel was right, and no count caught the shift in buyer taste.

The five practices answer those realities in order. Record where the time goes. Ask what you can contribute. Make strength productive. Do first things first and second things not at all. Then decide by system, which rests on dissent rather than on agreement about the facts.

Key Ideas

Record Your Time Before You Plan It

Effective executives do not start with planning. They find out where the time goes, then manage it, then consolidate it. The supply of time is inelastic, and you cannot rent or buy more of it.

Memory does not survive a log. One chairman was certain he split his time between senior men, customers and the community. A six-week record showed almost none of it there, and he needed two or three more logs before he believed it. Drucker's minimum is three to four weeks at a stretch, twice a year.

Three questions cut the log. The first asks what happens if an activity stops, and Drucker never met an executive without a quarter of his demands to throw away. The second asks who else handles it as well. The third asks other people what you do that wastes their time.

Senior executives rarely have a quarter of their time at their own disposal. A first draft takes six or eight hours in one stretch, and the same seven hours split into fifteen-minute pieces produce doodles. One bank president booked ninety minutes with a single agenda item, and every crisis he met waited that long.

Contribution Outranks Effort

The second practice is one question. What can I contribute that affects the performance and the results of this institution? Most executives answer a different one, and say they run the accounting department or that 850 people work under them. A man who stresses his downward authority is a subordinate whatever his title reads.

Every organization needs performance in three areas: direct results, the building of values, and the development of people for tomorrow. Losing any one of them kills it.

Nurse Bryan's rule tested every action against the best the hospital can do for this patient. It outlived her retirement by almost ten years. The most common cause of executive failure is refusing to change with the demands of a new position.

Staff From Strength and Make Weakness Irrelevant

Told that Grant was fond of the bottle, Lincoln asked for the brand, to send a barrel to his other generals. Grant's appointment turned the Civil War. Lincoln first appointed three or four generals whose main qualification was the absence of major weaknesses. The North made no headway from 1861 to 1864.

Lee staffed the same way, and his generals carried monumental weaknesses with one real strength each. Asked why he kept a general who wrecked his plans, Lee answered that the man performs.

Four rules run the staffing. Any job that defeats two or three good performers in succession is unfit for human beings and needs redesign. Make each job demanding and big, because a knowledge worker's first job sets the standards he holds for the rest of his career. Appraise on strength, and accept the weakness that comes with it, because two mediocrities achieve less than one.

The appraisal asks what he did well, what that predicts next, and whether you accept your own son or daughter working under him. Only the last treats a weakness as disqualifying.

George Marshall relieved any general officer short of outstanding, and read the relief as a verdict on his own appointment. His sixtieth birthday closed the job of Chief of Staff to him, and the appointment beat it by four months.

Posteriorities Are the Hard Half

The one secret of effectiveness is concentration, and it means one thing at a time. Mozart is the only known exception. One pharmaceutical chief executive did three large jobs in sequence across the eleven years before he retired.

Sloughing off yesterday is the first rule. The test on every running activity is one question: knowing what we know now, do we go into this today? Anything short of an unconditional yes gets dropped. Total failures liquidate themselves, and yesterday's successes linger long past their productive life. Drop the old before starting the new, and staff the new thing with proven people.

Drucker guessed that at least half the federal bureaus are obsolete. The Interstate Commerce Commission still guarded the public against a railroad monopoly that disappeared thirty years earlier.

Setting priorities is easy and everybody does it. The hard job is setting posteriorities, deciding which tasks to leave alone and holding to that. What you postpone, you abandon. Every posteriority is somebody else's top priority, so the call takes courage rather than analysis. Under pressure you pick the crisis over the opportunity and the urgent over the relevant. The rules that fight back take the future over the past and the opportunity over the problem.

Decisions Run on Boundary Conditions

Effective executives make few decisions. They work the important ones and aim to be sound rather than clever. A decision that never degenerates into work is a good intention. Five elements make the system.

• Classify the problem as generic or as a real exception, because a generic problem needs a rule.
• Define the boundary conditions, the minimum goals and specifications the answer has to satisfy.
• Think through what is right before you think about what is acceptable.
• Build the action commitments into the decision itself.
• Set up the feedback that tests the decision against the actual course of events.

The common mistake treats a generic situation as a run of unique events. The rare exception and the first case of a new genus look alike.

Boundary conditions are the specifications the answer has to satisfy. Sloan asked in 1922 whether removing the autonomy of the division heads met General Motors' needs, and the answer produced decentralization. They also say when to abandon a plan. The Bank Holiday changed the conditions under Roosevelt's 1933 recovery plan, so he switched to reform. They catch the decision that works only while nothing goes wrong, like the Bay of Pigs in 1961, whose two specifications were incompatible.

Start from what is right rather than from what is acceptable, because the compromise arrives at the end anyway. Sloan told Drucker at the start of the 1944 General Motors study to write down what he thought was right. Nobody makes the right compromise until somebody says what right is.

A decision becomes real when carrying it out is somebody's work assignment. One company sold a discontinued model as a replacement for another three years, and nobody told the purchasing clerk. The plant ended with parts for eight to ten more years of production. Theodore Vail changed the yardsticks at Bell to measure managers on service rather than profit.

Even the best decision has a high probability of being wrong, so feedback belongs inside it. The officer who gave the order goes and sees for himself, having learned to distrust reports.

Disagreement Is the Entry Fee

The first rule of decision-making bars a decision when nobody disagrees. Sloan opened one committee meeting by noting complete agreement on the matter in front of them. Everyone nodded. He postponed the discussion to give the group time to develop disagreement.

Dissent guards against capture by the organization, where everybody is a special pleader in good faith. It also produces the alternative, and a decision with no alternative is a desperate gambler's throw. The German General Staff weighed no alternative to the Schlieffen Plan for twenty-five years. Roosevelt had his Brains Trust ready in 1933, and nothing for the court-packing fight after his 1936 landslide.

Treat the dissenter as intelligent and fair-minded, and ask what he has to see for his position to hold. Understanding comes first, and who is right comes after. One starts with opinions, and facts do not exist without a criterion of relevance.

The traditional measurement reflects yesterday's decision, so the effective decision-maker assumes it is the wrong one. About 4 percent of military inventory items by number accounted for 90 percent or more of procurement dollars. McNamara ran that list in minute detail and sent the other 95 percent to management by exception.

One alternative is always doing nothing. Act or do not act, and never hedge, because a surgeon who removes half an appendix risks the same infection as the full job.

Practical Applications

Keep a time log for three weeks, written as the day happens. Read it against your own account of the week, and the gap is the diagnosis.

Run three cuts on the log. Drop what changes nothing when it stops. Hand off what somebody else does as well. Ask three colleagues what you do that wastes their time.

Consolidate what survives into blocks a real task fits inside, and defend one ninety-minute block a day.

Write the contribution sentence for your own job, in results rather than in headcount. Ask each person you depend on what he needs from you, and in what form.

Set posteriorities in writing each quarter, and name what you drop rather than what you postpone. On the next decision that matters, write the boundary conditions before the options and appoint somebody to argue the other side.

Who This Is For

The book aims at anyone whose output is a decision rather than a countable unit. Engineers, researchers and founders with nobody reporting to them all qualify, because the definition runs on knowledge.

Skip it if you supervise measurable manual output, where efficiency is the whole job. Skip it too if you want tools or a system to install. The book hands over five practices and an argument about why each is hard.

Read it with the vintage in view. The copyright reads 1967, and the vantage is the large institution: the United States Army, General Motors, the Bell System, federal agencies, big hospitals. The smallest unit Drucker examines is a department, and a small company runs on different physics. He argues from observation rather than from data, with no study and no counts, and every case arrives already knowing how it ended.

The Decision

The cheapest test in the book costs three weeks and no money. Log your time as the day happens, not from memory at the end of it. Then run three questions over every entry. Each entry answers what dies if it stops, who else does it as well, and whose time it wastes.

Drucker's prediction is that a quarter of the entries go in the bin and nobody notices. If the log matches your account of the week, put the book down. It will not match.