The amount of literature devoted to teaching people how to become a successful executive is vast. And that’s no surprise: People aspire to be in the C-suite, and they want to do a good job when they finally get to the top rung of the corporate ladder. Moreover, leaders are faced with great expectations and are usually the ones held responsible when things go wrong. Which raises the question: How do you stop things from going wrong?
Hal Gregersen, the executive director of MIT’s Leadership Center, has been researching whether corporate leaders are too comfortable once they get to the top. Gregersen interviewed more than 200 senior business executives—including Elon Musk of Tesla, Ed Catmull of Pixar, Sara Blakely of Spanx, and Marc Benioff of Salesforce—on how leaders become insulated from information they need, and why that can lead to a plethora of unexpected problems. His research was published in a new article in Harvard Business Review.
I spoke with Gregersen about the problems of insularity in the position of an executive, Uber’s recent controversies, and the revelatory potential of uncomfortable situations. A lightly edited transcript of our conversation follows.
Bourree Lam: Your article in Harvard Business Review explores what's "wrong" with CEOs and leaders. How did you become interested in this topic?
Hal Gregersen: I spent a little over a decade trying to figure out how a leader drops into a new company and effectively survives and ultimately succeeds. Then I spent another decade trying to figure out what it takes to create transformational change. The last decade has been devoted to [studying] cutting-edge kinds of innovation. The commonality across all three of those spaces is that they have to ask different questions in order to get new answers to the problem sitting right in front of them.
Lam: You refer to something called the CEO cocoon, which suggests that CEOs wind up being very isolated from crucial information. Why does that happen and is it the status quo?
Gregersen: I think the default strategy and direction with any organization is towards efficiency. It's all about the assumption that we already know what to do and we're doing the right thing, so let's just be more efficient with it. When that starts happening, people stop asking questions and it can easily turn into a situation where leaders don't want to get questions coming at them. A number of the leaders I interviewed said that they got hired, and promoted up the ranks because they were able to give quick, smart, relevant answers to people. The problem is, when people move into that CEO role, they not only need to know how to do things that are currently being done, but they're defining the future. That takes a completely different behavioral-set, mindset, and skill set. Many organizations don't realize that, and when they get into that pure efficiency mode, solving and poking and challenging the status quo is not part of the mantra.
Those CEOs asking better questions and speaking out about what they don't know are obsessive about client, customer, and user needs. When a leader takes that fundamental position of customers first, and they actually mean it, they're simultaneously taking the position that problems are first, not politics.
Lam: Why do corporate leaders get so caught up in politics?
Gregersen: The challenge is: The higher we go, the harder it is to figure out the “unknown unknowns.” We increasingly get isolated from the front line. The challenge becomes that once people move into leadership roles, they often spend too much time in offices and too little time out on the edge of their organizations where people are voicing legitimate, honest concerns about what's working and what isn't. For a leader in a formal position, if they don't get up and get out of their office and into the world with people they're not normally with, these forces of isolation tighten their grip often without them even knowing it.
We fail to ask new questions when we stop being in different places around different people. When that stops happening, we're crippled by lack of information. We start going down the path of being right and comfortable all the time. Unfortunately, it gets too loud to hear any new under-the-radar data coming in.
Lam: In your article you talk about the need for leaders to be quiet. That was so surprising to me. Can you tell me more about that?
Gregersen: What's intriguing is that even someone like Steve Jobs knew how to be quiet when he was on the hunt for new information. He would ask, and he would listen properly. Most leaders, if they're getting promoted to give answers, they're used to talking all the time. Unfortunately, a lot of people in organizations are more than happy to let leaders talk and take responsibility.
Think of the television show Undercover Boss. That's the classic situation when a leader gets into the kind of place where they're getting unfiltered information, either by what they see or what they hear. Some of that is going to be uncomfortable, and some of that is going to say, "Guess what, leader, you're doing something wrong. In fact, really wrong." That feeling of being uncomfortable and wrong is not normal. The natural tendency is to run from it or avoid it. When leaders embrace that, those conditions silence people. For example, Deval Patrick, who used to be the governor of Massachusetts, intentionally creates those pauses and just waits. Ed Catmull at Pixar, if you go to him and say that something might be wrong and uncomfortable, he almost always gives it a 24-hour gestation period before he'll really respond to it.
Lam: That discomfort and "you're doing something wrong" message from the front lines really brings to mind the recently leaked video of Uber's Travis Kalanick talking to one of his drivers.
Gregersen: You think about the leaders, like Walt Bettinger, who actually goes out of his way to have those conversations where people are saying "Guess what, things weren't what they thought they were." If leaders don't go out of their way to go after this passive data—the data that's there but doesn't, intentionally, systematically come at [them]—they'll get overwhelmed with all the information coming in and get blindsided like the Uber CEO did.
Lam: What else did you learn from talking to 200 senior executives?
Gregersen: Most of these people I interviewed were quite skilled at putting themselves in situations that forced them to ask better questions. What's intriguing is that the leaders who were most capable of explaining to somebody else how they go about surfacing the right questions, figuring out what they don't know before it's too late, they're also the best at teaching other people how to do it. They're the kinds of leaders that actually build a culture where it's not just one person doing all the talking, but it's a big part of the system to reward and support that.
Lam: As I read your article, I was wondering about whether it's really a good idea to have one person in charge.
Gregersen: There are a number of companies in the past who have had two or three CEOs. They're the exceptions. The challenge is, when there's more than one person guiding the organization from the top, they've got to be completely capable of resolving differences in a rapid enough way that the organization can keep moving forward.
I think there's value in having a CEO, but there are these natural dynamics that happen around a senior leader, especially the person at the top, that severely restrict information and perspectives. If a CEO doesn't figure out how to knock those walls down, it's arguable that the CEO is worth having there.
In innovative companies, it's not just the CEOs that are the innovators. There's a deep, fundamental expectation that everyone generates new ideas that create value. There's just an expectation that you're creating a future here no matter who you are in the system. That's uncommon. But when that's the case, it's a lot easier to check the CEO's power and keep them from falling into the dilemma of a bubble.
We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.
