Featured Classic: Good To Great
A conversation with Jim Collins
By Joe Flower
From Health Forum Journal July/August 2002
In the 20-year
history of these interviews, Jim Collins is the first person we have talked to twice. Here’s why: In 1994, with Jerry Porras, Collins produced
Built To Last, a remarkable study of what it took for an organization to survive for generations.
Soon after that publication, a McKinsey consultant buttonholed Collins after a presentation to say, "Great book. Really interesting. Totally useless."
Collins was taken aback. "Why?"
"Because all the companies you looked at were already great. What most of us really want to know is, how do you become great?"
The question gnawed at Collins. It led to over five years of research, and finally some answers - intriguing, surprising, even confounding, and deeply useful for anyone who runs an organization. Collins told us the story form his home tucked up against the mountains in Boulder, Colorado:
Can good become great?
Can an organization that has never been better than average shift to become something far superior – superior to what it was in the past, and superior to everyone else. Can good become great? And if so, how?
This question hooked me deeply. I couldn’t shake it. So I began to assemble a research team to see if we could find the answer.
How do you measure it?
First, we had to ask: How are we going to select companies that have done this? Early on, with the guidance of Harvard’s Morton Hansen, I decided to use a sound, scientific method.
There is little good science in management studies. Science calls for two things — predictability, which is not possible in the real world, and repeatabilty, which you can get. Any other researcher, using the same criteria, should end up with the same study set.
We said, “Let’s look for a specific pattern: If you put a dollar into this company’s stock at one point, for at least 15 years they would have done no better than the general market. Then, after some transition point, it shoots upwards and substantially out-performs the market for at least 15 years.”
That is long enough that you can be confident that it’s not an Enron, not somebody manipulating the numbers, not a lucky event, and not dependent on a single CEO. It would have to be something pretty substantial to last that long. But 15 years is short enough that we would not have to be looking at ancient history.
We decided that three times the market was enough. We picked GE as a marker and asked, “How well has GE done at its best? About 2.5 to 2.8 times the market. Let’s up the ante from there.”
Would we find any companies? Or too many?
Would we find any companies? Or too many? I fund my own research. I began writing checks, and the checks said, “James C. Collins and Joanne A. Ernst.” They had our home address on them. And they were big. Eventually I would spend over half a million dollars on research. When Joanne realized the scale of the project, she said, “Gee, I sure hope you find something.”
The concern was more than academic.
And if we found no companies that went from good to great, we would have to publish that result: “You’re all doomed.”
A nine-month death march of financial analysis revealed a set that would withstand scrutiny, 11 companies that not only beat themselves, they beat the market, and they beat their
own industry.
Sexy? Snoresville.
The first interesting thing we noticed was how uninteresting the companies are. They are not sexy high fliers with fascinating new technologies riding on a wave of media hype. I still find astounding, for instance, the story of Walgreens. People don’t think of it as glamorous. But when we ran the numbers, they were so unbelievable that we ran them three times on different data sets to make sure (a tactic we eventually used for all the chosen companies). An investment in Walgreens from 1975 to 2000 would have beat an investment in Intel two to one, GE five to one, Coca-Cola eight to one, and the market 15 to one.
Kroger beat the market 10 to one from ’73 to ’98 – a grocery chain, of all things. A steel company, Nucor, out-performed the market by five to one, in an industry whose performance ranked 70th out of 71 industries. Wells Fargo beat the market nearly four to one at a time when its industry fell 40 to 60 percent behind the general market.
What this meant was clear: Becoming great is not circumstance-dependent. If you can do it in steel, in grocery stores, in drug stores, in banking in the middle of de-regulation
when everyone is getting hammered, if you can do it at Fannie Mae when interest rates are against you, you could do it anywhere. All these companies had tremendous adversity and
constraints, yet they attained some of the most spectacular results of any corporations in the last generation.
What doesn't work
What leapt out of the data most clearly was what does not work. Many of the assumptions that we have about what it takes to make things better simply are not supported by the evidence.
Making something great is not harder, does not involve more work or suffering, than making something that is merely good – in fact it may involve less suffering and labor.
How you pay people — having substantial executive compensation in order to motivate executives to do the right thing – does not play a role at all.
The idea that you can get there through artful use of technology — this is not supported by the data. Technology helps, but is never the cause of either greatness or decline.
Mergers and acquisitions cannot ignite greatness, they can only accelerate what is already happening.
The idea that the most effective change comes from outsiders bringing in “fresh blood” turns out not to be true. Less than 5 percent of the good-to-great CEOs were outsiders, versus 31 percent of the direct comparison company CEOs. Ten of the 11 good-to-great companies never used an outsider CEO.
The usefulness of programmatic change, the “change program,” was not supported by the data. One of the most wonderful moments in the research process happened when a CEO of a company came to our lab in Boulder. I asked him, as we always did, “What would you want to know the answer to?”
They had no name for it. No tag line. No launch event.
He said, “I’d like to know: What did they call what they were doing? How did they talk about it at the time?” We went back to the data and looked. We found no evidence that they talked about it at all. They had no name for it. No tag line. No launch event. No self-consciousness. It was just a process of doing.
A CEO of one of the 11 said, “We were going through a period of total transformation, but I am not sure we were aware of it then. It was only later, that we could see it.”
If you were there during one of these transitions, each day was a deeper understanding, a key decision, a key piece of data that helped you see things, a whole series of decisions that built upon one another, until at some point there is this crystalline breakthrough. It only looks sudden to outsiders.
I asked Kurt Walgreen, “Can you identify the moment, the day, the incident that would indicate that you had made the leap?” He thought about it for a minute and said, “I’d say it happened sometime between 1971 and 1980.” He was CEO when it happened. And he can’t identify when it happened in less than a 9-year spread.
A self-conscious, large, noisy, hoopla-filled, programmatic change effort is not likely to work.
The hedgehog concept
But what did work? What was the core?
For months, years, I sat here on the porch of my bungalow in Boulder, reading the books, the data, and the 6,000 articles we had collected about the 11 companies and their direct comparisons, over and over, trying to get at the essence of the differences between them. Companies in the same industry, facing the same dynamic, with the same set of facts, would seem to be looking at two different worlds. Fannie Mae versus Great Western, Bethlehem Steel versus Nucor, Kroger versus A & P.
What came through to me was this: The people in the “good to great” companies did things that seemed so incredibly obvious, straightforward, simple. Finally that maddening simplicity began to make sense. The comparison companies may have been very smart people, brilliant, but they saw things as complex, and they had elaborate plans and strategies.
One night I was reading some Tolstoy, and I came across the Isaiah Berlin essay on Tolstoy, “The Hedgehog and the Fox,” based on an ancient Greek adage: “The fox knows many things, but the hedgehog knows one big thing.” Suddenly I saw how completely this applied to the people who built the good-to-great companies. They know one big thing. Their world is not terribly complicated. The comparison people were smart, but they couldn’t grab onto anything simple that would actually work.
But then we began to wrestle with the obvious fact that “simple” doesn’t mean “right.” Did we just happen to have a handful of cases in which the strategy was simple, and they were just lucky enough to be right? Or was there something deeper to it? That’s where the whole idea of the “Hedgehog Concept” came from. This is one of the things that really made my head hurt.
We stared at it for years, asking ourselves what these different corporate strategies had in common. In our first stab at it, we came up with eight variables. Then we boiled that down to six, and eventually to three circles. The companies each had a simple concept, but that simple concept reflected a deep understanding of three variables.
They understood what they were passionate about, what they were best in the world at, and what drove their economic engine.
One: They came to understand what they were genuinely passionate about. They would never do a business just because it was a business. They would do things that they felt truly excited about. That’s what drove them in different directions. Kimberly Clark wasn’t passionate about paper; it was more passionate about the consumer business, so it sold the paper mills.
Two: They confronted the reality of what they could be best in the world at. A number of our companies made dramatic changes over time by recognizing that what they had been doing for most of their history was something at which they were competent, but not something at which they could be the best in the world. The only way to attain greatness as an organization is to do something at which you can be the best in the world.
Three: They gained a piercing understanding about the drivers of their economic engines. Every one of the companies ended up with an elegant, simple economic denominator, Profit per X.
Wells Fargo came to understand that, “We always talk about profit per loan, but in a de-regulated world it’s really about profit per employee.” It was a simple idea that happened to be right, and it drove their entire system.
When they found the place those three circles came together, they began to get breakthrough results.
This is not something you can jump into. You can’t do an off-site with flip charts and come up with your three circles in six hours of breakouts. It took these companies, on average, four years to get it right. Insight and understanding take time.
The leadership problem
Good research gets you results you don’t expect. Really good research gets you results that you don’t like. In this research, I found a number of findings challenging — especially the “Level 5 leadership” concept.
I’m biased against leadership answers to start with. It was hard for me even to have a chapter title with the word “leadership” in it.
But this particular leadership answer I find even more challenging. I can see how to be a Level 4 leader, but becoming a Level 5 would be really hard. The discipline they have, and the ways they are able to transfer their ambition into something far beyond themselves, I find an inspiring and daunting standard.
On the surface, the Level 5 leader is the antithesis of the heroic, charismatic, outgoing, egocentric change leader. Most have had a charisma bypass. The comparison CEOs, in contrast, were often high-profile, charismatic change leaders, brought from the outside in many cases, who in fact had tremendous classical leadership capabilities — people like Lee Iaccoca at Chrysler, Stanley Gault at Rubbermaid, or Henry Singleton at Teledyne.
But what’s below that surface? The real question is not what is your style, it’s what are you ambitious for? Are you ambitious first and foremost for yourself, or for the organization, the cause, the work? Coupled with that: Do you have the will, the discipline, to execute on that ambition? That is the essence of it: a ferocious will on behalf of an ambition that is not about you.
Culture of Discipline
The “culture of discipline” turned out to be, for me, one of the more profound ideas in the research. It’s a complicated idea with many parts. The essence of the culture of discipline is self-disciplined people who operate with a lot of freedom within a general framework. They understand that they do not have a job, they have a responsibility. Imagine an air traffic controller. You could define their job in terms of specific actions, but their responsibility is to see that the planes don’t crash.
The medical professions are based on the idea that you don’t just have a job, you have a responsibility for this patient’s life.
The Stockdale Paradox
What helped Admiral Jim Stockdale, the highest ranking prisoner of war in Vietnam, survive those years, is what I call the Stockdale Paradox, and it is desperately needed in healthcare. You have to live with the absolute faith that you can prevail in the end, regardless of the difficulty — and at the same time have the discipline to confront the most brutal facts of your current reality, whatever they are.
If healthcare feels harder than other industries, that’s because it is. Embrace that fact. At the same time, keep the faith that you can create pockets of greatness despite the great systemic forces working against you.
In order to create pockets of greatness, all I need is the ability to influence or make the decisions about who is on my bus, who is off my bus, and who is in what seat. I have seen people do it in school systems, in divisions of companies, even in parts of large bureaucratic government agencies. It is entirely possible.
Not-for-profits
The three circles are even more relevant to the not-for-profit world. They are also harder to get right.
We have little rigorous understanding of the not-for-profit world, but it is clear that the most effective social-sector organizations are the most focused. If you look at the best-run social sector organizations, such as the Boys and Girls Clubs, the Girl Scouts, or the Cleveland Orchestra, one of the ways in which they made themselves sound institutions was to be rigorous and clear about what they should not do.
I always go back to those three circles and think about them. Number one: finding out what you’re really passionate about. That’s an easy one for the not-for-profit sector. But the second circle is harder: What can you do better than any other institution? It requires tremendous discipline to say, “Yes, we could do this well, but we can’t really make a unique contribution. We should leave it to others.” Third: It doesn’t matter whether you’re for-profit or not-for-profit, you need a powerful economic engine. One of the things that will be very hard to make sense of, as a not-for-profit organization is not how to make money but how to build an economic engine that allows us to fulfill our mission so that we can stay in all three of those circles.
All the ideas that emerged from this research can be helpful to the not-for-profit world.
