Robert Sutton’s Guide to Excellence
The Stanford professor’s latest research explores the practices that enable companies to scale what they do best. And watch the video “How to Scale Up Excellence in an Organization.”
Professor Robert Sutton is, above all else, a humanist. That makes him something of a rare breed in the academy of elite management thinkers. Whereas many of the most noted business scholars of the late 20th and early 21st centuries have made their mark by advancing the fields of strategy, finance, and broad organizational dynamics, Sutton has focused on revealing the role of the person. He studies how policies drive people and how people drive companies. When so much management theory pays lip service to individuals, assuming they are ultimately replaceable parts in a system—and that as long as they are treated well and suited to their duties, they will perform adequately—Sutton’s work often flips the equation.
In his delicately titled The No Asshole Rule: Building a Civilized Workplace and Surviving One That Isn’t (Business Plus, 2010), for example, the Stanford professor of management science and engineering and cofounder of Stanford’s Institute of Design (known as the “d.school”) makes a compelling argument for one critical element that distinguishes high-performing companies: They don’t tolerate jerks. Even when he is exploring his passion for evidence-based management, as he does in the two books he has written with longtime Stanford colleague Jeffrey Pfeffer, Sutton seeks and extols the virtues of practices that directly affect and are affected by the behavior of unique human beings. As Sutton and Pfeffer wrote in their January 2006 Harvard Business Review article on evidence-based management, “Many practitioners and their advisers routinely ignore evidence about management practices that clashes with their beliefs and ideologies, and their own observations are contaminated by what they expect to see.” Even in the cold, hard world of evidence-driven theory, Sutton understands, it all comes down to the person.
How to Scale Up Excellence in an Organization
Stanford's Robert Sutton discusses the mind-set and strategies of companies that are most adept at building and spreading high standards.
S+B: Let’s begin by talking about the mind-set of leaders who are adept at scaling excellence.
SUTTON: Leaders who are adept at scaling have an almost paradoxical mind-set. On one hand, they’re always focusing on cleaning things up—smoothing and clarifying policies and processes so they are more spreadable. They have a clear sense of the excellence they’re trying to spread and of where they’re going. On the other hand, they understand that life is messy, and there are going to be long stretches when they and their people are confused and don’t know exactly what needs to be done. But they urge people to keep moving forward in spite of that. That’s the paradox: They’re always trying to clean things up, but they embrace chaos and don’t let it stop them.
Take IDEO founder David Kelley, whom I’ve known for many years. David has this ability, when people are all worked up about how difficult and uncertain things are, to calm them down and say, “We’ve just got to keep going one step after another.” But then at the same time he tends to be quite picky and specific about what achieving excellence looks like. In some respects, his leadership style turns a lot of traditional management wisdom on its head—especially the direction we typically give to people with operational responsibilities. We’re often coached to drive for clarity in the near term while embracing the idea that the future is full of ambiguity.
How David and his team face the world is completely different. On a day-to-day basis, they accept that there’s mess and confusion, but they balance that with a clear vision of where they are going. Sometimes you just don’t know what you’re doing in the short term even while you’re quite certain of the long term, and you’ve just got to keep muddling forward.
S+B: One of the examples that you cite when talking about what you call “the problem of more”—simply the challenge of how to grow the scale of what you do well—is Pulse News. And I love this anecdote because you talk about the challenges they face, increasing from an organization of two people to eight people. And at first blush that doesn’t—
SUTTON: No big deal, right?
S+B: Exactly. So, can you talk a little bit about Pulse News?
SUTTON: Pulse News is a great organization. I’ve known the two founders, Ankit and Akshay, for about three or four years, which is a long time, given their youth. They had this idea for one of the first news aggregation apps for the iPad. They write an app, and it’s just the two of them. Then, there’s four of them. And then eight. Just a tiny little startup. But soon they start fighting and having setbacks.
What they realized, and this is one of the important takeaways of the book, is they hadn’t just gone to an eight-person organization, but they’d gone to an eight-person work team. Suddenly, there was all this confusion and lack of coordination. What they did [next] was one of the hallmarks of great scaling. They cut themselves into three smaller teams, and they got a little bit more effective. Then, they added just a little bit of process. It was simple: Each team maintained a bulletin board—they’re all in one room—that showed what they were doing. Every afternoon somebody from the team would stand up and say, “Here’s what we accomplished, and here’s what we need help with.”
You think of “scaling” as this giant thing: We’re going to replicate this process that a few people are doing to thousands and thousands of people. But Pulse just grew from two to four to eight, and even within those minuscule numbers, they had problems with both size and process that needed to be addressed before they could scale further.
That pressure is always there when you’re scaling, whether you’re scaling General Motors or you’re scaling Pulse. By the way, Pulse had a similar problem when they hit 11 people and 14 people. But they dealt with it, and the story has a happy ending: They recently sold themselves to LinkedIn for $90 million.
S+B: Surely not every organization encounters problems at that level. What is the size of a manageable team? When is a team too big?
SUTTON: Well, the classic question of when is a team too big depends a lot on what you’re doing. I like to back up and ask people to imagine that they’re going out to dinner with three or four people versus nine or 10. It’s virtually impossible—we’ve all had this experience—to have a conversation with nine other people. That’s why the average restaurant reservation worldwide is for roughly four people.
Teams of up to five generally work pretty well in the workplace. When you go too far beyond that, team members spend more time coordinating than doing, and you start having more interpersonal problems. And it’s very difficult to have a functional team that has more than 10 members. There are just too many human variables to manage. You’ve got to break them into subgroups.
At the beginning of World War II, the U.S. Marines for some reason had decided that a combat team should be 12 people. But they had all sorts of problems with camaraderie and, more seriously, high mortality rates. So they split up into four-person combat teams, and things improved rapidly. Today, throughout the world, whether it’s the U.S. Navy SEALs or virtually any combat group, the standard fire team in battle consists of one leader and three people being told what to do. And that’s as much as the leader and the team can reasonably take.
Here’s an important takeaway: If you have a team that you think has a lousy leader, the first question to ask is, Is the team too big? It’s amazing how crummy leaders become great when they go from leading, say, 11 people down to five. And the reverse is true as well. You may think someone has reached their limits as a manager when they’re asked to take on a larger team and the results get worse. But the problem may be more the size of the team than their ability to lead.
S+B: How does this play out when many organizations are trying to become flatter?
SUTTON: Right, we keep hearing this mantra: Go flat. “We’re going to have flat organizations with few levels, and we’re going to be decentralized.” But the challenge is that in the process, very often you get teams that are so big that they’re just out of control. An example that we use in the book was the point in the early days of Google when Larry Page decided to get rid of all the middle managers—and they ended up with 100 people reporting to a single manager.
It’s not that I’m opposed to flatter organizations. It’s just that you have to be careful that you have enough layers to maintain some control. This is sort of unavoidable. Hierarchy is not an inherently evil concept; it’s gotten a bad name because it’s become associated with barriers and ineffectiveness.
Hierarchy has gotten a bad name because it’s become associated with barriers and ineffectiveness.
Here’s what Google does quite effectively: They’re still fairly flat, but they have a social contract that everyone understands. The more that you base your governance on an agreement about what’s sacred and what’s taboo, the less closely you have to supervise people.
S+B: You label 2014 the “Year of Subtraction,” as a way to focus people on cutting out the clutter of what doesn’t matter before they scale what does. So let’s talk about three of the components of subtraction. The first is, “If you only had five rules, what would they be?”
SUTTON: That’s to help force constraint. The most widely cited article in the history of psychology was written by George Miller in 1956 and was called “The Magical Number Seven, Plus or Minus Two” [Psychological Review]. As human beings, we can focus our attention on only so much stuff, and five is a pretty good number.
One of the cases of scaling we talk about in Scaling Up Excellence is the spread of electronic patient records and operations at Kaiser Permanente (KP), the largest healthcare system in the United States. As they scaled their electronic operations, called KP HealthConnect, rather than adding thousands of new rules and processes, they put in place five simple guardrails that gave clarity to what really mattered in the change effort. For instance, one was that physicians should return patients’ emails within 24 hours. Another was that no part of the vast KP network could introduce software changes that hampered the network’s ability to maintain a single, integrated system. It was just that simple. Instead of overloading people with complex processes, apply simple constraints that keep people focused on what matters. Our brains can handle only so much.
S+B: Second rule of subtraction: “Stop doing dumb things just because everybody else does.”
SUTTON: Well, this is a classic mantra. Leaders often talk about this notion of traditions that have been done because everybody else does them. Well, there was a point in medical history when everybody did bloodletting. Turns out, it was not a good thing. In that light, one question that I’d pose for your readers is “How’s your performance evaluation system working for you?” There’s very good evidence to suggest it is ineffective. If performance evaluations were a drug, they would not get FDA approval—at least as they’re done in most organizations. About 20 percent of the time they make things better, 20 percent of the time they make things worse, and 60 percent of the time, meh. Yet we keep on doing them in pretty much the same ways we always have, based on nothing more than tradition. [Also see “Kill Your Performance Ratings,” by David Rock, Josh Davis, and Beth Jones.]
But there was a courageous woman we talk about in the book, Donna Morris. Donna is head of HR at Adobe, a Fortune 100 company. Adobe got rid of performance evaluations. They discovered the process took 80,000 person-hours each year and added no value. Overall employee satisfaction didn’t go up as a result of performance reviews and, worse, the best people would often leave. So they changed to a system they call check-ins, where managers are responsible for constantly checking in with their people and giving them feedback. They also place a lot of emphasis on employees evaluating their bosses about the quality and usefulness of the feedback.
Oh, and Donna did one other thing that I thought was a brilliant form of subtraction. She took the evaluation forms away from managers, because she said if you don’t have the forms, you’ve got to look people in the eye and actually talk to them.
S+B: And the third rule: “Subtract yourself.”
SUTTON: I learned this from [IDEO’s] David Kelley. There are lots of times in organizational life, especially if you’re the boss, when you’re doing two things without realizing it, just by being present. You’re both stifling the group and wasting your own time.
David avoids these traps by focusing on being a convener. He brings people together. Maybe it involves a client. Maybe it’s just two people who need to know each other. He stands up at the front of the room and gets a conversation rolling. Then, as the discussion takes off, he walks to the back of the room, and then he walks out completely. No goodbyes. He just leaves. Then he goes to another meeting and, soon enough, walks out of that one.
So, I joke, remember “management by walking around”? Kelley does “management by walking out of the room.” And I’ve seen him do this about a thousand times in my life. Partly, he’s just kind of an impatient guy. He has to move around from place to place. But the reason he says he does it is that “very often I’m—for better or worse—the authority figure, and by my being absent, other people become more creative.”
I would also add that when you’re a boss, and even just a boss of a small group, not the CEO, in some cultures it’s not so bad for you to be around. People may kind of ignore you anyway. But in other cultures, the boss can be stifling of people taking risks, making mistakes, speaking out of turn.
In one of the great IDEO projects from many years ago—this is 1994 or so—the company was working with Samsung. Some of us will remember that Samsung was sort of a bad imitation of Sony for many years, though now they’ve become this great, innovative company. So, Samsung brought a bunch of Korean managers and young folks to Palo Alto to teach them design thinking in the IDEO way. They were making no progress for months and months, and eventually what they did was to get rid of the old white guy from IDEO who was working on it and bring in a young Korean-American guy named Dan Kim, who understood the culture. The first thing Dan said is, “We’ve got to get the bosses out of the room.”
Once the bosses left, the remaining managers immediately felt much more comfortable and started to freewheel ideas right away.
S+B: You talk about how bad behavior often emerges from incentives meant to fuel good behavior.
SUTTON: This notion is as old as the hills. When you give people money or other rewards for doing something, you expect them to do something good, but they often do something bad instead. Psychologists label these perverse incentives.
One of my favorite examples is an old one involving the Albuquerque garbage collectors. They were paid [according to] how much garbage they collected. That was it. They were not paid for anything else or by any other measure. So they’d drive like maniacs—50 miles an hour through the city—and they’d run at full speed and they’d just grab up everything in their path. And if your garbage wasn’t ready, they’d ignore you. They were going absolutely crazy, grabbing everything that was easy to grab—because that’s what they had an incentive to do. And not only were they having lots of accidents, they weren’t actually picking up as much garbage as they should have.
Almost everyone has been in a situation where they’re paid to do one specific thing that is basically blind to their surroundings rather than to see and work within the system as a whole. It never works out well.
Huggy [coauthor Huggy Rao] and I use a very simple diagnostic to gauge whether a company is encouraging people to be selfish or to support the general good. We look at who is getting rewarded: the people who are known as the organizational “superstars.” Are they the individuals who are only out for their own success, and step on and damn everybody else, or are they the people who get ahead by helping others succeed?
It can be easy sometimes to bash General Electric and Jack Welch. Yes, there are some things he did that I didn’t like, but one thing that General Electric has done very well for 20, 30 years is defining a superstar as somebody who both delivers good individual performance and helps the people around him or her succeed. GE measures the impact a person has on others’ success within the organization. I think that if you start with the simple question of who are the superstars, a whole bunch of other issues come to the foreground about accountability, culture, and the kind of people an organization values.
S+B: We saw perverse incentives playing out with devastating effects in the mortgage crisis. People were paid to originate mortgages.
S+B: In defining its culture and setting a course for scaling, you say an organization needs to decide whether it’s going to be Catholic or Buddhist. Please explain that one.
SUTTON: OK, let me begin with a story. It was when we were in the process of setting up the d.school at Stanford. A group of us were talking with a guy named Michael Dearing, a great early-stage venture capitalist. He had been a senior vice president at eBay and had been raised a devoted Catholic.
We’re having a conversation about how to spread design thinking, and he asks this question, which I’m paraphrasing loosely from memory: Are we going to be Catholics or Buddhists? Are we going to have a relatively standardized set of approaches that we try to spread everywhere else, like Catholicism, or are we going to have a mind-set meant to be interpreted in various ways, depending on local conditions, which is more akin to Buddhism?
The difference between establishing standardization as a core organizational value and allowing for variation runs deep. And it’s one of the most important factors in guiding scaling decisions. Do we lean toward insisting that everybody must do things the same way, or do we allow them to be flexible for local cultures, for different sorts of clients? It’s one of those issues that doesn’t go away; it keeps recurring as you grow. You need to keep asking and answering this question again and again in scaling up.
S+B: Talk a little bit about how things look on the ground in a Catholic situation versus a Buddhist situation.
SUTTON: First, I would think of them as a continuum. But let’s take some pure types. A relatively clean approach to Catholicism out in California would be In-N-Out Burger. See’s Candies is another example. Both attempt to do essentially the same thing in the same way everywhere they open a store. If the store doesn’t work in the environment they’ve entered, they leave. Consistency and formula are what matters; they compete only where the formula works as is.
And for a more Buddhist philosophy, look to McDonald’s. No, most people wouldn’t think of McDonald’s as Buddhist, but it is Buddhist compared to In-N-Out Burger. When McDonald’s goes into a new geography, it flexes to the needs of that market. So, when they go to India, they do not serve hamburgers. They serve lamb burgers and more vegetarian dishes than you would see elsewhere. In Germany, they offer beer and wine, because that’s expected as part of the culture. With this notion, you start out with a basic formula and then you allow people to change it. There is a template, but it anticipates variation.
As you are scaling, there’s a lot of evidence to suggest that it’s easier to start with a fixed standardized template and then, as you gather information that suggests that it’s not working in all instances, to adapt it. This is what Howard Schultz did when he started Starbucks; it was actually quite brilliant. In the Seattle coffee shop that eventually became Starbucks, Schultz built kind of a lab to test out his original concept of an Italian espresso bar. It had a stand-up bar and blaring opera music, for example. Well, it turned out that Americans want to sit down with their coffee drinks and prefer soft pop to Italian arias.
Schultz started with a specific template, so he could see what parts didn’t work, and then he replaced the pieces that were not fitting. If you try to start out with an undifferentiated mass as opposed to a template, it’s much more difficult to figure out what works and what doesn’t. So it can be effective to be a purist to start, but then to be flexible as more information emerges.
You can extend this line of thinking to the challenges of scaling overall. While the particular circumstances of your company will ultimately drive the specifics of how you scale the best of what you do, there are a set of practices that have been proven time and again to add real value to the process. You need to start somewhere—and the more concrete the beginning, the easier it will be to eventually customize your approach in a way that people will understand, follow, and make their own.
Reprint No. 00276
- Paul Michelman is executive editor of strategy+business.