In the 20-odd months since In Search of Excellence was published, I have spoken to some 400 audiences. They were in Maui, Hawaii; and Geneva, Switzerland. They were representatives of old industries like steel and forest products, and of new industries like software and gourmet chocolate-chip cookies. They included everyone from a top wholesaler of plumbing and electrical fixtures in Sweden to the owner of the world's largest dairy store, in Norwalk, Conn. Among them were 5,000 city managers, 1,000 bakers, and the senior management of Bell Telephone Laboratories Inc.

Surprisingly, despite the diversity of the audiences, they all asked similar questions. What follows is a distillation of those questions -- and of how I try to respond.


Excellence focused on large companies. But is there anything you learned that relates to the particular problems and opportunities of smaller companies?

There is a lot of advice and writing on the transition a small company must go through as it grows. It suggests an inevitable cycle of deterioration, an evolution from a highly charged, entrepreneurial, customer-intensive enterprise to a bureaucracy. The implied message is: "You're going to become calcified; it's just a matter of how soon."

There is a germ of truth in this. There is no doubt that the $15-million company, no matter what the business, needs a more highly articulated set of procedures for controlling things than the $100,000 company does.But most of the scenario is hogwash. The magic of W. L. Gore, Dana, Emerson Electric, 3M, Hewlett-Packard, Procter & Gamble, PepsiCo, Milliken, and Johnson & Johnson -- in other words, the companies we hold up as excellent -- is that they have all retained a much higher share of their "small company" simplicity and vitality than their less effective competitors have.

The message of Excellence is that the key to corporate success lies in superior customer service, continuing internal entrepreneurship, and a deep belief in the dignity, worth, and potential of every person in the organization. There is no iron law that says you must lose those virtues as you grow; in fact, the principal job of senior officers of a small corporation is to maintain them. The tools may be different for a Disneyland than for a one-park, 20-acre establishment. But the intensity of customer focus can and must be maintained. The personal touch -- we call it Management By Wandering Around, courtesy of Hewlett-Packard -- can also be maintained. Giving up MBWA is not an imperative of size. But the sad news is that small companies can lose touch very easily and that the basics of Excellence are at issue from the start.


Excellence also focused on the concerns of people who are building companies, using models of enterprises that have been people-oriented and innovative since inception. But what can you do if you inherit a company that has been bureaucratic or lackluster for decades?

Successful turnarounds are a tough act, but what is perhaps most interesting about them is that they follow exactly the same form as the new-company building efforts. The greatest turnaround artists work repetitiously on creating distinctive strategic skills. They are tireless.Each, above all, believes in the power of a thousand little things done just a bit better.

A handful of stories come to mind. Rene McPherson inherited, as chairman, a then-$1-billion (now more than $3-billion) business in Toledo -- Dana Corp., a manufacturer of such products as brass propeller blades and axles, a company he described as "having the rottenest product line ever granted by God to a Fortune 500 company." In the 1970s he turned it around to the point at which it became the number 2 company among the 500 in return to investors. McPherson radically decentralized. He gave the factory managers autonomy and the tools to do the job: finance, personnel, purchasing, computers, etc. He created an exciting, no-nonsense, competitive environment with a ceaseless focus on practical productivity improvements. His focus was the people on the line -- "the boss of their 25 square feet," in McPherson's terms.

More important was McPherson's approach to change. When he is asked to comment on the importance of any particular program in his productivity-improvement process, he shoots back with: "It is damned important. Another feather on the scale."

A different sort of turnaround is PepsiCo Inc. Fifteen years ago it was a sleepy, second-rate competitor to The Coca-Cola Co. Today it is a vital, entrepreneurial, $7.5-billion company. The magic has been brute persistence, never assuming that the job is complete.President Andy Pearson's strategy, while in the field with a division, is to avoid the executive suite like the plague. He first zeros in on the young assistant brand manager, and he always asks the same damned questions. "What have you learned in the last 96 hours? What's going on in the test market? What are you up to?" That's the PepsiCo magic. Every one of Pearson's talks is, at best, another feather on the scale.

The theme is incredibly consistent, regardless of the company, yet each variation of the theme is unique.Successful turnabout artists are, pure and simple, no more and no less, broken records. No stone is left unturned. No stone is especially important. Yet no stone is unimportant.


One of Excellence's premises is that a company's attitudes about people, products, and its way of doing business are defined by its founder. The book holds up Tom Watson, the founder of IBM Corp. and the inspiration for many of IBM's characteristic values and practices, as a model. But what if the founder of your company lacks Watson's stature?

There are several ways to answer this question. Most important is to challenge the notion that the people who run the socalled excellent companies are 17 feet tall with well-honed tap-dancing skills. It's not so. Some are extroverts. Many are introverts. Tom Watson Sr. was, in his own words, an "avowed showman." On the other hand, if you're more than seven inches away from Bill Hewlett, you have a tought time hearing him. But they all have one thing in common.Almost all are obsessives.

The issue is not one of personality. It is one of vision, persistence, and consistency. Yes, the desert is littered with the bones of those who had similar ideas and similar obsessiveness, yet failed. Obsessiveness and consistency, even with vision, don't make an automatic winner. On the other hand, we virtually never find a winner who doesn't exhibit these traits. So while we can't guarantee that obsessiveness or consistency surrounding a solid vision will produce a winner, we can almost certainly guarantee that the absence of these traits will result in a loser.


Excellence is written to and about the top managers at some of the world's largest companies. But what can be done to make your company more excellent if you're not at the top?

In the worst of companies, there tend to be good divisions, great factories, inventive product-development groups, superb accounting departments -- "pockets of excellence," my colleague Nancy Austin calls them. General Motors Corp. has a raft of problems, but the old Packard Electric division, under the guidance of Jim Rinehart (now chief executive officer of Clark Equipment Co., a manufacturer of forklifts and other industrial equipment) was a winner, and the Livonia, Mich., Cadillac engine plant, formerly under the guidance of Bob Stramy, is a star-studded operation. Pockets of excellence can occur at lower levels of operation, as well. I find superb accounting departments within hopelessly bureaucratic organizations, superb sales branches in demoralized companies, superb stores in lousy retailing operations. There is no doubt that a gutsy, obsessive, consistent visionary can bring about radical change in his or her part of the organization, even if it is at odds with the view of the "old man."

One caveat: I don't extend the argument to suggest that the excited plant or accounting department or store will be the precursor of a corporate turnaround. Pockets of excellence don't foreshadow automatic diffusion throughout the organization. We are certain that the leadership for a true corporate turnaround must come from the top person and the top team.


Excellence often seems to value just plain hoopla -- songs and parties, wandering around giving Fourth of July speeches about values, firing the accountants, and hiring the cancan girls. Surely that's not enough -- don't you need controls?

I'm not suggesting that formal control systems, budgets, and procedures be thrown out the window. But the more intriguing and important issues surround the form of control.

J. Willard Marriott Sr. had an odd habit: For well over 50 years, in what eventually became a multi-billion-dollar enterprise, he read every single customer complaint card -- raw and unsummarized. One of his senior officers summed up the impact: "You had a hundred or so property managers working 26 hours a day, eight days a week, to make sure the boss had a very, very light reading load."

A longtime Procter & Gamble Co. manufacturing manager -- the company is renowned for its rigid, yet simple, control system -- responded to the claim of a financial service company president that I was talking about balance between formal and informal systems of control. "That's B.S.," he said. Then, with great passion, he related a supporting story: "We don't have complex, mandated control systems in P&G manufacturing. We have no mandated MRP [material requirements planning] systems. I'll tell you how I learned about P&G's view of quality. It wasn't from a manual. I got a call one night, about 1:30 a.m., from a district sales manager. I was a young manufacturing supervisor. He told me, 'George, you got a problem with a bar of soap down here. "Down here' was about 200 miles away. 'How about coming down here tomorrow morning. I'd like to see you about 6:30 or 7:00, so you can get a look at it.' The tone of his voice suggested something more than an invitation. After you've once driven 200 miles through tennessee mountains to look at one damned (his word was stronger) 39? bar of soap, it suddenly sinks in that Procter & Gamble is deadly serious about product quality. They don't need to supplement that lesson with a 300-page manual."

Another friend relates a tale from Du Pont Co., known for its safety record."I was doing some work in Wilmington. I stepped into the executive suite's coffee room to grab a cup of coffee from a machine. It was one of those infernal machines where the little plastic door drops down, and three times out of four hits the coffee cup. Sure enough, I spilled some on the floor. Out of nowhere -- I swear I didn't see him before -- a Du Pont vice-president (and, she adds, they only have about 25, in spite of the company's size) swooped in on me, grabbed a paper towel, and almost frantically wiped up the coffee. I stood there gaping. He looked up at me almost incredulous: 'Don't you know people can trip and hurt themselves on those?" In spite of the vast size of Du Pont, every accident that occurs anywhere, no matter how minor, is the subject of a report dispatched to the chairman's desk within 24 hours of its occurrence -- in a $35.4-billion company. Du Pont's safety record is 17 times better than the chemical-industry average.

All of the above are examples of the most rigid control systems I've ever observed. Violate them and you're in hot water. Their simplicity invites understanding and ownership by all hands, and leaves space for creative improvements. Yet none are paper-driven, formal control devices. Their word-of-mouth nature, the attendant all-hands ownership, the constant all-hands tinkering, and the strict and regular senior involvement are what make them work so well.

We are strongly in favor of controls. Rigid controls. But control and paper, control and complexity, control and formality are not related.


Excellence speaks rather glibly about the desirability of "failure." But in many businesses -- airlines, medicine, public-sector management -- failure is catastrophic, or public. Some say a small business can't afford failure. What are the limits to the concept?

We principally, almost exclusively, learn from failure and failure alone. That is as true for the $1-million company as for the $60-billion one. As true for the airline pilot as for the hamburger chef. As true in the public sector as in the private.

Now I don't mean the $100-million failure of failing to install a control system where failure is likely to have catastrophic results. I mean inducing a strong desire in people to take an idea and test it -- now. Test it in the marketplace. Test a new procedure in a department, a bit or piece of a new product with a customer. Live tests always surprise the experts. The marketplace -- a plant for an accounting procedure, a single lead customer for a product -- is naive. It likes what it likes. The objective is to get to hard, tangible evidence quickly.

Many corporate philosophies focus on failure. Emerson Electric Co.'s "Ten Commandments," under chairman Chuck Knight, include one touting failure and calling it that. Johnson & Johnson's founder, General Robert Wood Johnson said, "If I wasn't making mistakes, I wasn't making decisions." Don Estridge, president of IBM's Entry Systems Division, says, "If I don't let people make mistakes, there is no ownership whatsoever." A Silicon Valley CEO who runs a highly successful computer peripherals company of moderate size has even quantified it: "We tell our people to make at least 10 mistakes a day. If you are not making 10 mistakes a day, you're not trying hard enough." That's not his comment after a third vodka-martini, it's part of an explicit, brief, written statement of corporate philosophy.

Managers have to assure people that failure is normal. That they have been there and back. I don't mean applauding shoddy work. The tiniest of experiments should be well thought out, designed for learning something quickly and efficiently. The only true scientific process depends on empirical progress, based in turn on the rapid collection of as many fragments of live data as possible.

For the small company, the major advantage of thoughtful failure is that it is more efficient that any other avenue to success. Failures will come -- no matter how hard or how well you plan. The issue is: Do we plan and plan and plan (and spend lots and lots of time), and then fail? Or do we test, fail, test, learn a little, test, fail, learn a little more, succeed partially, test, succeed?


Listen to your customers, build high-quality products, take care of your employees -- many of the book's key phrases sound like platitudes. Aren't these things common sense?

I often comment quite seriously that, five years after this research, there is only one question that truly interests me now: Why is it, in a roughly $20-billion industry, that only Walt Disney Productions knows how to keep a theme park really clean? "Keeping the park clean," or its analogue in other businesses, is indeed the height of effective strategic management. Common sense says listen to your customer. But few companies do. And those that do -- obsessively, over the decades -- tend more often than not to be among the winners.

The more annoying point hinted at by the question is: Why don't more companies do it? Few businesspeople are out to screw the customer. But we have come to focus on the wrong things -- technique, rather than product or service.

Business is not "management by objectives" and strategic portfolios. It is inventing things, making things, selling things, and servicing things. It is products that work and courtesy to customers. Common sense it is. But somehow we lose it. And once lost, it is extraordinarily difficult to get back


Excellence extols the "do-it, fix-it, try-it" approach, that of getting even a crude model of a product or procedure out among potential users as quickly as possible. But that seems the antithesis of planning. Is central planning really unnecessary?

Absolutely not. But large, central corporate staffs that generate 35 notebooks, each 300 pages long, are a waste of time. Many give lip service to "the right stuff," like decentralized, bottoms-up planning. But look at their apparently innocuous corporate "planning-guidance document," and you'll see that they have carved up space that should be left to the divisions. I'm in favor of sound strategic thinking and a well-drawn strategic plan -- at the division level.

Let's begin with General Electric Co., the Mecca of strategic planning, the company to which every American corporate executive came on bended knee, in the 1960s and '70s, to learn. Then Jack Welch became chairman. Jack, a former premier "skunk" and champion, drastically reduced the central strategic-planning department. Does he think that strategic planning is unimportant? No. But the plans are done by the division, within a general corporate framework. Why was Welch led to this? Because GE got behind in many major technologies. The company was well planned and under-entrepreneured. Moreover, most of its big successes in the past, even during the days of planning dominance, had come from skunk-entrepreneurs. He simply needed more of them. Welch's prime objective was to rekindle the traditional GE spirit of entrepreneurship.

Second only to decentralization is the magic word "short." Procter & Gamble has no central strategists, either. Each of its brand managers is a strategist. The P&G one-page memo rule holds for strategic planning: Each brand manager, as a part of a lengthy and intense process that occurs once a year, must boil his upcoming year's strategy down to one paragraph. That's right, one paragraph.

The reality -- I think all of us who have ever been forced through it realize it full well -- is that a strategy that is one paragraph long is harder to develop than a 50-page monograph. In 50 pages, you can be all things to all people. In a paragraph or two, you have to figure out just what makes you distinctive.


Excellence touts skunk works, but surely you have to have a master plan, not just random experiments. IBM may be able to afford seven teams working in parallel on the same project, but a small or medium-size company can't. Aren't skunk works and experiments just for the big guys, and in any size company, don't they amount to just a small share of the research and development budget?

There is truth in those statements.Development funds in the most innovative companies -- 3M, Hewlett-Packard, Johnson & Johnson, for example -- are not devoted only to skunk works. And, if you are in a $5-million company, you can't have seven parallel teams at work on the next new product. But that's where my willingness to accede ends. Our analysis of financial service firms, hamburger-makers, pharmaceutical houses, software producers finds that all successful innovation -- product or process -- starts and is nurtured in skunk works, big company or small.

The proper role for central R&D is threefold. First, central R&D gives a critical signal, the company's directional bet on the haystacks it wants to search in for its needles. This means that technical people are trained and aimed at exploiting roughly the right sciences or technologies. Second, R&D is a perfect place for the "real people," the people in the operating divisions, to steal (and the word "steal" is chosen carefully) ideas from. Finally, and most importatly, central R&D is the ideal place from which to steal people to put in the operating divisions. What central R&D is not is the place in which specific products appear at specific times, per plan.

Finally, the issue of parallel operations and experimentation in the small company. I feel almost violent about this: Experimentation and parallel projects and skunk works are vital to the $5-million company. The top problem for the fast-starting company is almost always coming up with the second or next series of products. The main block is often the founder/CEO or master designer who invented the first one. After succeeding once, he decides that he was mainly smart, rather than mainly lucky. The odds are that when a product is a commercial success, the stars were perfectly aligned the first time out. Innovation, small or large, hamburgers or computers, is a chancy activity at best. So it is very important to allow playfulness and multiple experiments from more sources than the genius-founder very early on, at $5-million sales levels or less.

The size of the skunk work or experiment or parallel project in a $4-million plumbing company is not that of Western Electric Co. The viable skunk work in the $10-million company may be just two-thirds of a person or one to two people. The experiment might take just 20 days and cost $15,000. But the principle still holds. Experimenting -- trying a new twist on a piece of software, or an accepted fashion, or a store format -- is vital from the start. We learn only in the real world.


The lessons of Excellence sound great. But do they really make a difference to the business? Do they have an impact on the bottom line?

The impact naturally varies. But top performers do hundreds of times better than the norm. Before Christmas of 1983, I visited Lockheed's kelly Johnson in Burbank, Calif. Johnson brought the expression "Skunk Work" to industry from the comic strip Li'L Abner. He designed much of what flies today -- the P-80 (F-80), the F-104, the U-2, the SR-71. Kelly's band of 125 regularly, over a period of decades, outperformed groups of three to four thousand on similar tasks. Outperformed in terms of higher efficiency, higher quality, lower budgets, and shorter development times. In one case, when he was called in to bail out an ailing operation, he reduced the number of people involved in a major aspect of it from 1,271 to 35 in the space of a year. At the same time, he raised the success rate from 12.5% to 98%.

Frank Perdue, in the mundane business of peddling fryers and roasters, gets a margin that is 800% above the industry averages -- and maintains market shares in the 50% range in tough urban markets. The new-product launch rates at Hewlett-Packard, 3M, Wang Labs, and Johnson & Johnson are miles ahead of their peers. Mervyn's, a star of the Dayton-Hudson Corp. retail operation, can remerchandise its stores in two weeks; competitors take more than three months. Raychem Corp., a highly successful $535-million company in Menlo Park, Calif., develops sophisticated, state-of-the-art, super-high-technology products from cold start to installation in periods as short as a week.

I grew up in Baltimore. The Baltimore Orioles, who for years ranked 26 out of 26 on salary, have put together teams that have executed exceptionally well: Over the past 27 years, Baltimore is well over 100 games ahead of the number two team in all of major league baseball. Those forms of dominance should be our goal -- not 2% here and 2% there.