In his acclaimed first book, Turn the Ship Around!, US Navy Captain David Marquet told the story of his time in command of the nuclear submarine USS Santa Fe. In it, he showed how switching from giving orders to giving control allowed him to turn one of the worst-performing ships in the Navy to one of its best.
Marquet’s new book, Leadership Is Language, dives even deeper into the essence of modern leadership. He picks apart the outdated vocabulary of command-and-control leadership, revealing how it still dominates the language of so many of today’s leaders. Marquet sat down with Texas CEO Magazine owner Joel Trammell to talk about evolving leadership language out of the Industrial Age, taking organizations from “can do” mindset to a “can think” approach, and how his work applies to the role of the CEO.
In Turn the Ship Around! you talk about being assigned to lead a ship you weren’t specifically trained on. That reminds me of the CEO role, where it’s very hard to be fully prepared for the job before you take it on. Did your experience change your view on how “qualified” a person needs to be before they take command?
Marquet: I don’t advocate not knowing your job, but that experience was the catalyst I needed to force me to become a different kind of leader. I always thought I wanted to be a “knowing and telling” leader—I thought I should know all the answers and give all the orders. When I got assigned to the USS Santa Fe, which I hadn’t learned, I became a “not knowing but still telling” leader. That didn’t work well. My solution was to commit to not give any orders.
Once I learned the ship, the question was, “Do I go back to giving orders now that I can?” I learned that just because you think you know the answer doesn’t mean you have to give the answer. Most leaders operate with the rule of “If I know it, I tell it.” But you have to exercise self-control. Telling people what to do can give you short-term wins; it can increase production and get stuff done. But letting people make their own decisions is what develops your team’s long-term capacity. You have to ask yourself the question: Do you want to be the decision maker, or do you want to build a factory that can make decisions?
Of course, as a leader, you get judged and evaluated on how well you make decisions. Every CEO innately feels the desire to hold on to that role, especially if they founded the company. But once you have 100 people, if you’re still the decision maker, you have one thinker and 99 doers. But the CEO should want everybody in the thinking game. You can’t afford not to have people thinking.
One way to keep people thinking is to not insist they be perfectly “qualified” for every role. If you come into a new job on with all the exact qualifications for that role—all the perfectly fitting knowledge, skills, and abilities—that might feel good, but it’s super boring. There’s no sense of growth or learning, no curiosity or excitement. You want some overlap between where you are now and what the role requires, but you also want to grow into the job.
From an organizational design perspective, you want people to not be fully qualified, to grow into their jobs. Then, at the point where they are fully qualified, they jump to the next role where they aren’t. Longtime CEOs can also lose that sense of progression. Submarine commanders are only there for three years, so there’s a sense of urgency that comes with that: “I have a thousand days to do what I can do.” In the corporate world, where people might stay in certain roles a lot longer, there’s a sense of arthritis. Sometimes leaders look at that situation in their organization and say, “Hey, how come no one’s acting like an entrepreneur here?” Well, it’s because there’s no growth happening.
In the new book, you describe the difference between redwork, or “doing” work, and bluework, which is that “thinking” work you’re talking about. Most CEOs got where they are because they were able to produce a lot of redwork. They weren’t asked to do much bluework early in their careers. That means you end up with leaders who are biased to redwork. Would you agree with that?
Marquet: Yes. Often, the whole organization is biased toward redwork. In the Industrial Age, we separated red and bluework by role. Leaders did the bluework and workers did the redwork. I describe what we want now with the phrase “Let the doers be the deciders.” The people who before were just doing redwork should be given some of that bluework. In fact, we should all flip to bluework every once in a while.
The problem is that bluework benefits from embracing variability, while redwork benefits from reducing variability. They require using your brain in two different ways. Most of us don’t have a language to say that we are in either bluework or redwork, so we tend toward redwork and just give lip service to embracing variability.
So, that’s one of the basic concepts in the book, that we oscillate between doing and thinking. When I tell people that, I get everything from “My God, that’s the most brilliant thing I’ve ever heard” to “That’s all you got? It’s obvious.” Personally, I’d never heard that distinction, and most companies don’t operate that way. We don’t have language for talking about experimentation and thinking. We go directly into performance mode, not learning mode.
There’s a point between 20 and 100 people where the CEO goes from knowing pretty much everything about the organization to not knowing everything. I thought your concept of moving authority toward the information was interesting. Can you explain how that works?
Marquet: That’s a concept that resonates with a lot of our clients. I had a light bulb moment once I made the fundamental decision to lean back and ask people to lean into me. That seems like “Well duh,” but most of us think that way. We think in terms traditional hierarchies with leaders and followers. We put people in categories accordingly—thinker/doer, leader/ follower, white collar/blue collar, salaried worker/hourly worker.
In that hierarchical setup, no one’s really happy. The “lower” person is just doing what they’re told, and the boss is having to run around all day long checking on people. Who wants that? Trying to separate the thinkers and doers, leaders and followers, and so on is a ginormous waste of time. And the bigger problem, the one you alluded to, is that hierarchies separate authority from information. The people at the frontlines—the coders, the people flying the airplane, the people conducting surgery—they have more nuanced information than the people with authority.
The Industrial Age solution to that problem was to aggregate information and channel it to the top. That’s why SAP makes billions of dollars with its software. These communication systems are designed for direction. We still call the layer below us our “direct reports.” I appreciate the clarity of the language, but too many of us act like that means “I direct and you report.” I like to see leaders switching that up, so the most common form of communication isn’t direction but intent. When you do that and allow people to do more thinking work—more bluework—you move the authority closer to the information.
One challenge of giving people more latitude to make decisions for themselves is deciding exactly how much latitude to give them. You can’t just let them do all the thinking work, right?
When thinking about how much control to give an employee, our two bins are competence and clarity. The idea is that you should attune the amount of control you give the employee based on their technical competence and their organization clarity. The equation goes like this: Control equals competence plus clarity, and then I write a “plus Z” at the end of the equation. That Z factor is trust—giving them a little bit more control than is probably warranted. Why? Because as we talked about before, you want people to grow into their roles.
In gauging a person’s competence and clarity, the first problem I see is that people don’t know what the competence of their team is, and they’re afraid to ask. At the [US Navy’s] Nuclear Power School, we have a culture where you get an exam a week. We’re used to demonstrating what we know. If you go into a corporation and say you’re going to give everyone a test to see if they know what they’re doing, people will take umbrage at that. That means that a lot of the time, we don’t know whether someone is competent. Our natural inclination as leaders is to assume that employees’ competence is low, when it’s probably not as low as we think. That’s why in fearful, non-transparent organizations, there’s a bias to over-control. Employees’ competence is unknown, so it must equal zero.
You talk a lot about language in this book. What drew you to that angle on leadership?
Marquet: Before I wrote my first book, I thought, “I’m an engineer. I can’t write books.” But now words and language are my new profession. I call myself a “word engineer.” I enjoy re-engineering the language.
I describe analyzing language as revealing the underlying patterns that are already there. Those structures and patterns are present and we sense them, but we can’t see them and don’t have labels for them. If you take the water low enough, you can start to see the rocks and shoals—the patterns that are already there. That’s super hard work, and I did it all wrong at first.
For me, there are so many interesting questions around leadership and language. Why do we call them “soft skills,” for example? I hate that—it makes it sound like they’re less important. Why does it sound more natural to say, “We’re a can-do organization, not a can-think organization?” Why does it sound more natural to ask, “Are you sure?” than “How sure are you?” Why do I go to a technology company and hear them say “We’re having an all hands meeting”? At a technology company!
An all-minds meeting might be better.
Exactly. Some of these are innocuous, but we often use the words we heard from our parents and our bosses, and they repeated those from their parents and bosses. It’s the same language that fits in Ford’s assembly line, and it doesn’t fit in our companies.
Another thing we say is “We act our way to new thinking.” Language can help there, too. Here’s a thought experiment I run with CEOs. I get them to fill in the sentences like these: “I would like our company to be more blank.” “I would like employees to be more blank and less blank.” “I’d like our culture to be more blank.” Filling in those is easy. Everyone fills in the same words: collaborative, thoughtful, accountable, and so on. But the next question is the important one: “What would it sound like if?,” or WWISLI. That invites the CEO to consider what it would sound like if people were more collaborative or thoughtful or accountable. The magic is that it forces you to think of a specific scene—two people at the coffee machine, six people at the weekly ops meeting, construction workers doing a morning toolbox briefing.
The magic of putting specific words in is that you can count the frequency of the words that signal collaboration. Those words also become the vehicle for advancing what you want. For example, on the Santa Fe, we said “Let’s practice saying we instead of they.” After a while of intentionally saying “we” instead of “they,” your brain gives up and rewires the synapses and you start thinking in terms of “we.” It’s like with running. You don’t think your way into a running routine. You put your running shoes on first and keep running until you start to think of yourself as a runner. That’s what it means to act your way to new thinking. We like that phrase.