Mediations with Candost
Software World with Candost
#21: Intent-Based Leadership with L. David Marquet

#21: Intent-Based Leadership with L. David Marquet

In this episode of The Software World, I welcomed L. David Marquet is a retired nuclear submarine Captain in the USS Navy. We talked about his book Turn The Ship Around!, the power of intent-based leadership, the importance of language, and how leaders can build the organization with the bottom-up decision-making approach.


Candost Dagdeviren: Hello everyone, welcome back to Software World with Candost. Today, I'm very excited, I have a great guest that I still cannot believe David Marquet is here. Well, let me tell you the story of David Marquet if you don't know him.

In 1981, David graduated top of his class from the U.S. Naval Academy—an institute renowned for developing “leaders to serve the nation.” Thereafter, he joined the submarine force.

Along his journey, one thing bothered him: the traditional leader-follower model. Used by the Navy and most companies around the world, the goal of leader-follower is to influence people to comply, not think. David experienced first-hand how this practice makes people feel marginalized. He knew in his gut that there had to be a better way.

He’d soon discover that to prove his theory he’d have to break some rules.

As engineer officer aboard the USS Will Rogers, a nuclear powered ballistic missile submarine, David tried empowering his team. He provided broad guidance, giving the team intent rather than orders. It was a disaster. His team made poor decisions that led to errors. He had to stop and revert to the traditional leader-follower method.

After a while, ultimately, David was selected to captain the USS Olympia, a nuclear powered attack submarine. He studied for over a year to take command, understanding on a deep level every detail of how that submarine operated.

Unexpectedly, David was diverted to take command of the USS Santa Fe when its captain quit. Santa Fe was the worst performing submarine in the fleet and a different type of submarine that he knew little about.

In a short time in Santa Fe, David realized that the leader-follower environment meant his crew would do anything he said—even if it was wrong. That could be catastrophic. He decided to try Intent-Based Leadership again.

Captain Marquet began treating his crew as leaders, not followers, and giving control, not taking control. It wasn’t long before operations took a dramatic turn. Santa Fe went from “worst to first,” achieving the highest retention and operational standings in the Navy.

Captain Marquet retired from the Navy in 2009 and authored the Turn the Ship Around! the book that we are gonna talk about it today. The book is a true story of turning followers into leaders. Fortune magazine called the book the “best how-to manual anywhere for managers on delegating, training, and driving flawless execution.” Captain Marquet’s Intent-Based Leadership model is turning around all types of organizations—from big manufacturers to start-ups and sport teams to government.

And today, we are gonna talk about this book because I read the book, I was pretty amazed, and I have many questions to ask to David.

Candost Dagdeviren: Without further ado, welcome, David.

L. David Marquet: Hey, thanks for having me on your show Software World, Candost

Candost Dagdeviren: yeah, thanks a lot. I really appreciate your taking the time and being here. So answer a couple of questions that I had after reading your create that turned the ship around. That was an amazing book. Thank you.

L. David Marquet: Yeah.

Candost Dagdeviren: So, in the book you mentioned about leadership style, it's called intent-based leadership.

Could you tell us briefly what is intent-based leadership. And what are the qualities that leaders will earn when they go to the this style and use this style?

L. David Marquet: Yeah, I think of it as a language and it's a language that we speak at work because what happened to me as a leader was I had to learn a new language and. Too many times, I think we approach leadership as some philosophical activity like you should empower your team. You should create motivated people. You should take care of your people. Oh, obviously. And we don't really talk about how you actually do it. So for us, the fundamental shift was from permission to intent and it was motivated because I was the captain of a submarine, which at the very last minute I was transferred to, which was not a submarine, which I had been trained to take over, it was a submarine I've never been on, kind of submarine I've never been on. So I was unfamiliar with all the equipment and the old model of me knowing all the answers and telling people what to do didn't work.

So the idea is the team says, here's what I intend to do. And then the leader can ask questions. The leader can say no, but it's initiated with the members of the team.

Now, when I say team, it's not a group effort, it's the individual owners. So if I'm say the senior vice president for product, it's not like five people come to me and they say this it's like thus, the product owner for a particular product comes to me and says in the next sprint cycle, this is what we intend to do or we intend to launch the product on this date, or we tend to delay the line, whatever it happens to be. And then they back it up with reasons why they want to do that.

Candost Dagdeviren: Okay. Interesting. Because like, when I hear, or like when I read the book the first time, and now I hear you, I had the different, a little bit different understanding. So for example, vice-president of product or any kind of leader that can come to the I don't know, executive, for example, and says, as we intend to do this in this sprint or in this cycle that we are having. But in my head, I thought about the individual contributors, like software engineers, for example, comes up and says, yeah, I tend to do that and not just in the mid level, but literally on the bottom level of organization chart.

L. David Marquet: Yeah, it can cascade all the way down, but what I was saying, the reason I was saying that was because people will ask me, oh, but I don't really have time for this consensus decision-making. It's not consensus decision-making. I never said that. What they would they say. The team tells the leader, what they intend to do. It's the head of the team tells the leader what they tend to do. And if it cascades down to the individual contributors, great. Today, at our daily stand up, we say, okay, we talked about what do you intend to do today?

And is there's two pieces. You can start with hey, what do you intend to do. Both the magic comes from what decisions are you making? We always think about our work in terms of what decisions are we making and what actions are we taking.

Candost Dagdeviren: Okay. Now, instead of a bit more clear, like the intentions and decisions.

Yeah. I I've seen a lot everybody is trying to either create a consensus or hate consensus. I was in both styles, but I think this intent-based relationship like someone's coming up and I intend to do this is also like having their own ownership and also bringing the accountability. If you're an intending to do that, which that means for you, that you are responsible for the results and that's your intention. If everything is good, just good and do it what are you intending to do.

L. David Marquet: Yeah, the key is it's initiated lower down. So you can't ever say, oh, well, I knew it was screwed up, but someone told me to do it. So the difference is one of our partners is a company called Crisp and they operate out of Stockholm and they consult with companies in the software space.

They don't talk about consensus because when we say consensus and the other phrase is everyone gets a voice. What happens is everyone thinks they can stop anything. So I need to make every beat. Every person vote. Yes. Before I can move forward, this is designed to prevent things from happening. So we say it's consent.

So I say, this is what I intend to do. And people can take an abstain. They can not vote. They could not respond, but as long as no one's saying no, then I'm moving forward. So I don't have to run around and convince everyone. It's the right thing to do. I just need to make sure no one sees any showstoppers.

Candost Dagdeviren: Okay, so this is a, I think it's a great point. Like in consent. Yes. Everyone's getting a say saying yes. Or getting an answer. Yes. Let's go with it. But in this style where if there is no objection or anything, if there is no red flag raising up saying that we should stop doing this, then you just got to go and do it whatever you want to do.

L. David Marquet: Yeah. Consent. So it's not, yeah, it's not consensus it's consent and it's not everyone gets a voice, but everyone should be heard. If you're the leader, you need to listen to everybody and then you make a decision. So that's number one. But number two is for most of these decisions, we push the decision down one level for almost every decision on the submarine and went down one level. So for decisions the captain used to make, they would be made by the engineer for the weapons officer or one of the department heads. Then for the decisions those guys would make, they would now we may buy junior officers. And so it was just like every decision got made one step lower, one step closer to where the information is.

We say, push the authority to the information, not information to authority. So we, in general, our system was set up so that people on the periphery of the organization would submit all these reports and stoplight charts to the people with authority to make decisions because they didn't, they were removed. They didn't know how many, how complex the software was. They didn't know how hard the job was. They didn't, they weren't looking into the face of the customer. And so they didn't have all the context and they say, well, then we spent all this time explaining the context. So that someone who, who is removed from the situation could make a decision.

That's a bad approach, in my opinion, because slows things down, decouples people from ownership and it creates distortion. So what you want to do is let the people who are closest to the information, have the authority to make the decision that they need. An easy example is a software engineer is in the code, they know exactly how complicated it is. They know how the testing is going. They know how many sort of little workarounds they put into it. And so they know what the risk is of having an intrusion or something like that.

To be able to explain all that to some executive far away. Is just takes a long, long time and they would never have the understanding that the person flying the airplane, unless the person conducting the operation, let's say the person selling the product that's a sales would make the decision.

Now it requires their motivations to be aligned with the organization, not for their own personal self, which is why we talk. It's easy to talk about, Hey, intent based leadership. But what you find is you'll spend a lot more time talking about what are you trying to achieve as an organization?

Candost Dagdeviren: Yeah, I, I love this. Like the quote actually, like this is one of my favorite quotes that I wrote down is like saying don't move information to authority, move authority to the information. But this also, like in my head, it's wanting is unclear about. It's the part that leaders can have a huge impact on changing the direction. Like information between the information and authority, but how non-leaders like, like software engineers, for example, can nudge their leaders on this point to make these changes, make this fundamental shift. This is where I struggled the most actually, when I was reading the book, I am a software engineer and I always thought, okay, this is being a leader. I mean, I don't need to hold the title of being manager or something. I'm not talking about that one, but how can I nudge my manager, for example, the person who has the authority, let's say to change their style in that direction.

L. David Marquet: Okay. So I think what you're saying is you were working for a person who makes decisions doesn't really listen to the team very well. Sometimes you agree with the decision. Sometimes you don't think is a very good decision and then they don't listen too well. So the first is I would be transparent and I would invite feedback. Here's the key. You can only control your own self. You only control your own act words and language.

So, I would say you got to make it safe for them to hear your idea. If you go and say. Hey, I don't think that's a good idea. They're going to feel threatened and it's going to feel unsafe. So we have to give them a choice. We always start with description. Hey, would you like to see what we see? Would you like to know what we know on the team?

Here's how we see it. Just describe what you're seeing. We have so many lines of code. The last time we did an update, it took a lot. You didn't compile properly the first 10 times, we had a lot of little glitches. It's it just feels like it's getting really, really complicated and we need to push towards simplifying some of the codes and then say, okay, so, so what do you want us to do?

And then you do that a few times, and then you say, okay, so here's what we see. Here's what we saying. Here's what we would do. If we were here, this is what I recommend. And you can sort of bootstrap yourself to having greater voice. It's hard to do, and I'm not sure there's any guarantees of success, but I do know that if you challenge the person's authority, it's almost assuredly not going to be a success.

Candost Dagdeviren: Okay. I think that's a strategy at least to give it a try, because I do think that the nudge from, like, for example, the software engineer here is in a good direction. And also like, I want to ask you one close and related to what we are talking about. Most probably you talk with a lot of leaders across different industries and with all these talks, what did you see that works well, especially, I mean, if, especially in software about shifting this managers mindset or position holding a position of privilege, to the accountability responsibility and doing the work.

L. David Marquet: The phrase we use is we act our way to new thinking. So what doesn't work is trying to convince people, oh, you need to be more like this, you need to be more of a servant leader. You need to be more inclusive. You need to be more, whatever. That's not going to work. I've never seen that work. What works is we say, Hey, you know what? Our protocol is that when we run meetings, what we're going to do is we're going to vote first and then discuss it.

So the typical way of meeting is run. Let's say the team's got to make a decision, or let's say the product owner has to make a decision. To launch the product it's scheduled for the launch in one week, we got to decide. 99% of all teams will get together. And then someone will say, oh yeah, well, you know, we're in good shape.

We've done all this testing. And then someone might say, well, yeah, but I'm not so sure about this, blah, blah, blah. And they talk about it. And then someone may say, well, let's vote, like show your hands. This is a terrible way to run the meeting because it just makes it really hard for people to express different opinions.

And we say, oh, well, we want to be really inclusive of diverse thinking, but you're running the meeting in a way that's exactly the opposite. So what you want to do is say, before we discuss and contaminate anyone with group thing, everyone write down on a card or show your hands. We do hands, we do cards one zero to five.

We call fifth to five. How ready is it for release? and then when people show their hand, and then now you're looking for the people who are real the fives and the zero. And if there's a whole bunch of fives and not too many zeros, you want to hear the zeroes first, you got to control the flow of the meeting.

That's what leaders do. They don't control the content. They say, okay, where are the outlying opinion? Okay. Candost, do you have an, I can see that you have an outlying opinion. You feel strongly, it's not ready. I need to hear from you. So let's let you get to talk first, not the majority or the loudest person, we don't need to hear from them.

And that's what the leader needs to do and needs to control the flow of information because humans are susceptible to those first things that they hear and they first four things they hear, oh yeah, it's good. Ready to go? Uh, blah, blah, blah, 737 max. It's a game changer, Blabla. Then the people who think differently will not speak up.

Candost Dagdeviren: That's a good perspective I think. Because like, especially I've seen many times that people are more focused on the content and like creating agendas. We create a lot of agendas for meetings and we try to stick to them. And we often overlook those situations. Like if someone is not very confident, they don't get these opportunities to talk out loud.

This strategy, in my opinion, has so in that direction to give the person who needs a voice, a voice, basically.

L. David Marquet: Yeah. And I wouldn't label them as they don't, I mean, it doesn't need to be someone with not a lot of competence. It's a human natural human thing. If you think that you're the only person who sees it some way, it's less likely I don't care how much competence you have, the probability that you're going to speak up, these are all probabilities. I mean, they should, no one should be thinking it's like a light switch to psych. It's just, we make it a little bit easier for people to speak up, but we make it a little bit harder for people to speak up. That's basically when you want to make it a little bit easier for them, the things that you want people to do and a little bit harder for the things you don't want people to do. As simple as that.

Candost Dagdeviren: Yeah. I want to like a bit switch to the part and talk about the, we touched a little bit at taking care of people. And we talked in this podcast with Susan Bond before about protective leadership. And you also mentioned in your book by saying that taking care of people in the organization doesn't mean keeping them safe from the results of their behavior. It means giving them tools, education, and advantage to achieve. Their personal and professional goals in the best way can be achieved in this direction. Can you elaborate on that a little bit? Because I'm curious to hear your thoughts more and also what are the most common mistakes leaders make? That in the end creates extra work and stress for these leaders, and then they fall into the old habits because in your book, I remember that you tried to implement this in a team that was not ready, and then you fall into making a decision by yourself as well.

L. David Marquet: Yeah, well, so protecting your team is a role for the leader, but what happens is people get confused and someone makes a mistake, and they say, oh, I'm going to protect you from the consequences of that mistake. That is not the way to do it because now what you're doing is you're decoupling the consequences from people's behaviors. So now what you do is get more   irresponsible behavior.

What you do is you want to protect them from things like if your boss is having a bad days, yells at you. You protect them from that. You don't even talk about it to the team. You don't get upset. You don't take it out on the team. You protect them from that. You also can protect them from undue stress and the feeling of time pressure. You say things like, Hey guys, let's, we're going to take a time out for the next hour. I just want to review where we are and talk about how it's going. Don't worry about the deadline. And protect them from those kinds of stress.

But, so for example, we had a situation over here where a college coach was the junior coach was molesting the boys in the locker room and the senior coach knew about it, but was "protecting him" from these consequences and it was horrible. That's not what we're talking about, that, that, and then this person, the senior coach was famous for "taking care of his people", but this is dysfunctional. This is not love. This is not any kind of responsible adult way to interact.

So people get confused. They'll just blur it out or I can tweet out a take care of your people, but it's a lot more nuanced than that. You want to protect them from external threats, not from their own behavior. We need to identify, if there's bad behavior on the team, someone shows up consistently late for meetings. You don't "protect them" from that. You challenge, you ask them, Hey, what's going on? There's a reason why you can't show up on time. It feels disrespectful for me and the rest of the team. So that that's where people get this wrong in  my mind.

Candost Dagdeviren: This reminds me like the type of protective leadership reminds me the leadership that you experienced and then you took, over the, the submarine. when you changed the submarine captain role to the, I think it was called Santa Fe. And at that time, your Commodore, what you mentioned in the book is that your Commodore told you "we aren't going to vote on and tell you what you need, but whatever you think you need we will support." And I think this, with this same with this, your Commodore, may, they protected you from other things that are on, they won't come to you and they, they didn't come to you most probably I don't know. But they didn't come to you and told you that you should do that or do this. These are the consequences or I don't know what's going on there, but the, the cases that you experienced as an employee while you were a captain, of course, but also you transferred this into your team in the submarine.

L. David Marquet: Yeah. Yeah, there was the word we use incubate. I call it incubate. He incubated us, which means that's what you do kind of when you got the eggs and the haven't hatched yet, and they're very fragile and you're kind of protecting them. And then they hatch and they go out into the world that's incubation. And then I, and I think that was very important because it allowed us to work through some things. Before we got too much scrutiny because if you get too much scrutiny, you tend to operate more in a fear-based mode. So that was very, very helpful. They incubated us.

Candost Dagdeviren: Yeah. I think that was like one of the critical points that I want to touch also is that this incubation had you through to implement this intent based leadership as this is how I see it.

And when I think more about, about your experience before being an engineering officer in your previous submarine, and you tried the similar structure and, but you were not successful. And at there you're, you're this failure of you shaped you in a different way. And when you got the opportunity or this level of protection incubation from your commodore at that time, this had you through to achieve intent based leadership, basically.

And now I'm curious about your tasks very shortly, about how like the mid-level leaders, for example, in software companies can get this protection to implement this intent-based leadership in their teams. Yeah.

L. David Marquet: I'm not sure. We have. So one of our stories is there's an executive working at a drug company and he was offered to take over the diabetes division, which was not doing very well, that this particular drug company and this person had very strong reputation. And so he was able to negotiate and he said, look, I'll do it. but I don't want anything. I'm not going to report any metrics for six months.

I'm not going to do weekly sales reports or anything like that for six months, we'll do them, but I'm not going to show them to you because I don't want to be, I don't want to have to react to this moment by moment thing. Cause we're going to retool the team. Now I, my experience was things happen very quickly within like a week.

We people were morale was up. People were signing up to stay in the Navy, whereas before they were leaving. But the performance on very in, on examinations was up. So very quickly things started getting better, but it will feel like there's this pause, because what happens is team comes to you and says, Hey, here's the situation? What should we do?

And in your head, you think, Hey, it's my job to tell them what to do. And B you know what the answer is? You think, you know what the answer is? So you say, okay, do the following, blah, blah, blah. Add the feature. Don't add the feature let's say, and then move on. And we're, and we're making progress and it's, and it's over biased towards progress, but the decision we didn't really spend much time on the decision.

In fact, the leader was involved in the decision to when you say, okay, well look, why don't you guys go away and come back tomorrow or come back in half an hour and tell me. What you would do here? It feels like that half an hour is wasted time or that whole day is wasted time. Because if you just told them what to do, you'd be, they'd be doing it already. But you say, come back and half an hour, then what you would do that is critical time though, because now that's like exercising in the gym for decision-making. Cause the team now is thinking about, okay, what did we do? And we have to weigh, what are the criteria what's really important.

What's not important. What does the client care about? What is the company here? How much capacity. So they're making all these calculations that you need to do to make good decisions. So when they come back to you yet, they "wasted half an hour" because we weren't doing anything. But once we get started, we spend the next week.

What we're doing is much more likely to be helping us in the right direction. Then something, we got to go, ah, then we lot now we lost a whole week and I would never, like, don't say, go away and think about it for two weeks. That's too long. A lot of these times, for me, it was literally, sometimes it was 30 seconds.

I would just say, don't say anything for 30 seconds. Just think about it. Do you think about that? Oh, that's a super short period of time, but when you stand there for 30 seconds, just looking at another person, usually I go right to, Hey, I'm going to get a cup of coffee. I'll come back. When I come back, tell me what you would do if you had my job.

So, so you're doing two things there. You're giving them a pause to think about it, and then you're taking them and you're changing their perspective. So instead of thinking about, okay, I'm the engineer, what I need to do for engineering? You're saying, oh, now I'm the captain. Now I'm the CEO. Now I'm the CTO from that.

Okay. What if I were the CTO? What would I. Sometimes we say also your future self. If you were six months from now, what do you think your six month your future self would want today's self to do? Another thing is to say, imagine you just got replaced by somebody who knows everything, you know, what would that person do?

So in other words, you, you getting out of your own, you're stuck in your own head too much. And we want to get out of our own.

Candost Dagdeviren: Yeah. These are like, I think all helpful. And I know like, I want to listen to you more and more, and, but I know that you have an event coming up. I'm thinking to come to Prague for that

L. David Marquet: Yeah.

Candost Dagdeviren: I'm coming to, could you please talk to us a bit about the event? I mean, what it will be about, and when?

L. David Marquet: yeah, of course, I'm super excited to coming to Prague on September 22nd is the Project Agile and Leadership event in Prague. And this is one of the first in-person conferences that I'm going to, and it was, it was about six months ago. They invited me and they say, well, we could do it virtual. And I said, you know what I said to the conference virtual. I said, no, we're going to try and do it. So look, yeah. And then I'm going to be there. And so I love Prague is a beautiful city, of course, and I'm got my vaccination and I'm ready to travel. So I'm coming to Prague, September 22nd, the project agile and leadership conference, check it out.

We're going to have some fun. And we're going to talk more about like these things that we should be saying, as opposed to what we've been programmed to say.

Candost Dagdeviren: I will try to be there. I mean, it's close to me, but I will try to be there. Maybe we can meet face to face.

L. David Marquet:  Yeah, definitely come say hi.

Candost Dagdeviren: Well, thanks a lot, David. It was a pleasure to me and thanks a lot for taking the time to join me in this podcast.

L. David Marquet: All right. Cheers. Thanks a lot. And uh, thanks all the listeners. Take care.

Candost Dagdeviren: Thanks.

L. David Marquet: Good luck.

Mediations with Candost
Software World with Candost
Software World is a podcast for software engineers hosted by Candost. Every second Tuesday, Candost uncovers the journeys of people and software systems. I interview the experts or talk alone about software architecture, system design, feedback, software engineering leadership, careers, team management, processes, product and customer-centricity, and more. Follow my blog at, for articles and a podcast, and subscribe to my newsletter.