“Turn the Ship Around!” (L. David Marquet)

After seeing a number of positive reviews and recommendations for this book, I asked the Melbourne Library Service to procure a copy – they agreed and I’ve recently enjoyed reading the fruits of their investment.

Marquet is a former nuclear submarine commander and the book details his moves to change the leadership on a poorly-performing submarine from leader-follower to what he calls “leader-leader”.

He starts out by describing what leadership meant in the (US) Navy, quoting from the “Naval Academy leadership book”:

Leadership is the art, science, or gift by which a person is enabled or privileged to direct the thoughts, plans, and actions of others in such a manner as to obtain and command their obedience, their confidence, their respect, and their loyal co-operation

As he points out “leadership in the Navy, and in most organizations, is about controlling people. It divides the world into two groups of people: leaders and followers.” His solution to the leader-follower pattern is the leader-leader model:

The leader-leader structure is fundamentally different from the leader-follower structure. At its core is the belief that we can all be leaders and, in fact, it’s best when we are all leaders. Leadership is not some mystical quality that some possess and others do not. As humans, we all have what it takes, and we all need to use our leadership abilities in every aspect of our work life.

The leader-leader model not only achieves great improvements in effectiveness and morale, but also makes the organization stronger. Most critically, these improvements are enduring, decoupled from the leader’s personality and presence. Leader-leader structures are significantly more resilient, and they do not rely on the designated leader always being right. Further, leader-leader structures spawn additional leaders throughout the organization naturally. It can’t be stopped.

Marquet details his journey of building the leader-leader model during his time turning around the flagging fortunes of the Sante Fe submarine. His passion, guts and honesty in making the changes he did shine through the narrative and results in a really simple but powerful model for changing the way we view leadership in organizations.

He argues that “the core of the leader-leader model is giving employees control over what they work on and how they work. It means letting them make meaningful decisions. The two enabling pillars are competency and clarity.”

One of the first things Marquet noticed on joining the Santa Fe was their focus on avoiding mistakes:

What happened with Santa Fe…was that the crew was becoming gun-shy about making mistakes. The best way not to make a mistake is not to do anything or make any decisions. It dawned on me the day I assumed command that focusing on avoiding errors is helpful for understanding the mechanics of procedures and detecting impending major problems before they occur, but it is a debilitating approach when adopted as the objective of an organization.

This observation led to his first mechanism for clarity: achieve excellence, don’t just avoid errors.

His discovery of the processes around signing off sailor’s leave led to his first mechanism for control: find the genetic code for control and rewrite it:

The first step in changing the genetic code of any organization or system is delegating control, or decision-making authority, as much as is comfortable, and then adding a pinch more. This isn’t an empowerment “program”. It’s changing the way the organization controls decisions in an enduring, personal way.

I like this idea of “Don’t move information to authority, move authority to information” in this area too.

The next control mechanism Marquet came up with was: act your way to new thinking:

When you’re trying to change employees’ behaviours, you have basically two approaches to choose from: change your own thinking and hope this leads to new behaviour, or change your behaviour and hope this leads to new thinking. On board Sante Fe, the officers and I did the latter, acting our way to new thinking.

The next control mechanism rings very true in software development, especially when adopting agile practices: have short early conversations to get efficient work:

Supervisors needed to recognize that the demand for perfect products the first time they see them results in significant waste and frustration throughout the entire organization. Even a thirty-second check early on could save your people numerous hours of work… a well-meaning yet erroneous translation of intent [could result] in a significant waste of resources.

In his mission to turn passive followers into active leaders, a “minor trick of language” turned into an effective means of control: use “I intend to…” Marquet would “avoid giving orders. Officers would state their intentions with “I intend to…” and I would say “Very well”. Then each man would execute his plan.” It turned out that this simple change “profoundly shifted ownership of the plan to the officers.”

Another control mechanism follows: resist the urge to provide solutions. This is a really hard habit to break and I’ve been working on this personally during my coaching and mentoring activities. I can see how it breaks down the leader-follower mentality, but it takes a deliberate effort to stop yourself from stepping in and solutionifying!

Marquet’s next control mechanism is: eliminating top-down monitoring systems:

Supervisors frequently bemoan the “lack of ownership” in their employees. When I observe what they do and what practices they have in their organization, I can see how they defeat any attempt to build ownership.

Worse, if they’ve voiced their frustrations out loud, their employees perceive them as hypocritical and they lose credibility. Don’t preach and hope for ownership; implement mechanisms that actually give ownership… Eliminating top-down monitoring systems will do it for you. I’m not talking about eliminating data collection and measuring processes that simply report conditions without judgement. Those are important as they “make the invisible visible”. What you want to avoid are the systems whereby senior personnel are determining what junior personnel should be doing.

More control follows in the shape of: think out loud:

When I heard what my watch officers were thinking, it made it much easier for me to keep my mouth shut and let them execute their plans. It was generally when they were quiet and I didn’t know what they would do next that I was tempted to step in. Thinking out loud is essential for making the leap from leader-follower to leader-leader.

Regular inspections were part of Navy life and Marquet and his crew decided to “be open and invite outside criticism”, in a control mechanism he calls: embrace the inspectors.

Embrace the inspectors can be viewed as a mechanism to enhance competence, but I think it fits even better in the discussion of control because it allowed us not only to be better submariners but also to maintain control of our destiny.

[It] also turned out to be an incredibly powerful vehicle for learning. Whenever an inspection team was on board, I would hear crew members saying things like, “I’ve been having a problem with this. What have you seen other ships do to solve it?” Most inspection teams found this attitude remarkable.

While Marquet started off by pushing decision making and control to lower and lower levels in the organization, he found that control by itself was not enough and he also needed to bolster the technical competence of his crew if this approach was to be successful.

The first mechanism for competence he outlines is: take deliberate action. Following an incident where a circuit-breaker was mistakenly closed on the submarine, the idea of taking deliberate action arose following a postmortem:

“Well, he was just in auto. He didn’t engage his brain before he did what he did: he was just executing a procedure.”

I thought that was perspective. We discussed a mechanism for engaging your brain before acting. We decided that when operating a nuclear-powered submarine we wanted people to act deliberately, and we decided on “take deliberate action” as our mechanism. This meant prior to any action, the operator paused and vocalized and gestured toward what he was about to do, and only after taking a deliberate pause would he execute the action. Our intent was to eliminate those “automatic” mistakes. Since the goal of “take deliberate action” was to introduce deliberateness in the mind of the operator, it didn’t matter whether anyone was around or not. Deliberate actions were not performed for the benefit of an observer or an inspector. They weren’t for show.

This particular mechanism reminded me of the ideas in Daniel Kahneman’s “Thinking, Fast and Slow” book.

The next competence mechanism is easy to relate to: we learn (everywhere, all the time). Embedding the idea that everyone in an organization needs to be constantly learning is a very good thing, be it in a military setting like Marquet’s or a software engineering setting. I actually think folks in IT are generally on board with this idea due to the high rate of change in technology, programming languages, etc. and the popularity of IT-related meetup groups, for example, are an indicator of a willingness to continue learning outside of the scope of the day-to-day in the office.

His next mechanism for competence is: don’t brief, certify.

A briefing is a passive activity for everyone except the briefer. Everyone else “is briefed”. There is no responsibility for preparation or study. It’s easy to just nod and say “ready” without full intellectual engagement. Furthermore, the sole responsibility in participating in a brief is to show up. Finally, a brief, as such, is not a decision point. The operation is going to happen and we are simply talking about it first.

We decided to do away with briefs. From that point on we would do certifications.

A certification is different from a brief in that during a certification, the person in charge of his team asks them questions. At the end of the certification, a decision is made whether or not the team is ready to perform the upcoming operation. If the team has not demonstrated the necessary knowledge during the certification, the operation should be postponed.

Another competence mechanism is presented next: continually and consistently repeat the same message:

Repeat the same message, day after day, meeting after meeting, event after event. Sounds redundant, repetitive, and boring. But what’s the alternative? Changing the message? That results in confusion and a lack of direction. I didn’t realize the degree to which old habits die hard, even when people are emotionally on board with the change.

This mechanism is one I’ve employed frequently in coaching testers around the world and it’s surprisingly effective (I say “surprising” since it surprised me that it is both necessary and valuable to do it).

Marquet’s last competence mechanism is: specifying goals, not methods. This arose from a fire drill in which the team members followed a prescribed response but failed to extinguish the fire within the safe time limit:

…[now] the crew was motivated to devise the best approach to putting out the fire. Once they were freed from following a prescribed way of doing things, they came up with many ingenious ways to shave seconds off our response time [to fires].

The problem with specifying the method along with the goal is one of diminished control. Provide your people with the objective and let them figure out the method.

Lovers of “best practices” please take note!

The third and final set of mechanisms Marquet introduces are around clarity:

As more decision-making authority is pushed down the chain of command, it becomes increasingly important that everyone throughout the organization understands what the organization is about. This is called clarity.

Clarity means people at all levels of an organization clearly and completely understand what the organization is about. This is needed because people in the organization make decisions against a set of criteria that includes what the organization is trying to accomplish. If clarity of purpose is misunderstood, then the criteria by which a decision is made will be skewed, and suboptimal decisions will be made.

His first mechanism for clarity is: building trust and taking care of your people:

It’s hard to find a leadership book that doesn’t encourage us to “take care of our people”. What I learned is this: Taking care of your people does not mean protecting them from the consequences of their own behaviour. That’s the path to irresponsibility. What it does mean is giving them every available tool and advantage to achieve their aims in life, beyond the specifics of the job. In some cases that meant further education; in other cases crewmen’s goals were incompatible with Navy life and they separated on good terms.

The next mechanism for clarity is: use your legacy for inspiration. This one helped to provide organizational clarity, explaining the “why” for the crewmen’s service:

Many organizations have inspiring early starts and somehow “lose their way” at some later point. I urge you to tap into the sense of purpose and urgency that developed during those early days or during some crisis. The trick is to find real ways to keep those alive as the organization grows. One of the easiest ways is simply to talk about them. Embed them into your guiding principles and use those words in efficiency reports and personnel awards.

Another mechanism for clarity comes next: use guiding principles for decision criteria.

Leaders like to hang a list of guiding principles on office walls for display, but often those principles don’t become part of the fabric of the organization. Not on Santa Fe. We did several things to reinforce these principles and make them real to the crew. For example, when we wrote awards or evaluations, we tried to couch behaviours in the language of these principles. “Petty Office M exhibited Courage and Openness when reporting…

Most of you have organizational principles. Go out and ask the first three people you see what they are. I was at one organization that proudly displayed its motto in Latin. I asked everyone I saw what it meant. The only one who knew was the CEO. That’s not good.

I’ve personally seen these working well within Quest – we have a set of Core Values and we refer to them regularly at all levels of the organization.

Another mechanism for clarity is: use immediate recognition to reinforce desired behaviours. I really like this one, it’s simple but easy to forget to do:

When I say immediate recognition, I mean immediate. Not thirty days. Not thirty minutes. Immediate.

Look at your structures for awards. Are they limited? Do they pit some of your employees against others? That structure will result in competition at the lowest level. If what you want is collaboration, then you are destroying it.

A mechanism for organizational clarity comes next: begin with the end in mind.

As you work with individuals in your organization to develop their vision for the future, it is crucial that you establish specific, measurable goals. These goals will help the individuals realize their ambitions. In addition, you as a mentor have to establish that you are sincerely interested in the problems of the person you are mentoring. By taking action to support the individual, you will prove that you are indeed working in their best interest and always keeping the end in mind.

His final mechanism for clarity is: encourage a questioning attitude over blind obedience. He asks “Will your people follow an order that isn’t correct? Do you want obedience or effectiveness? Have you built a culture that embraces a questioning attitude?” Reinforcing that asking questions is a good idea is so important in what we do as software testers (I recently heard Nick Pass define “QA” as “Question Asker” during his talk at the TiCCA19 conference) and there are sometimes personality and cultural barriers to overcome in encouraging people to question (the latter I have much experience of while working with our teams in China).

In summary, Marquet’s set of mechanisms for Control, Competence and Clarity are as follows.

Control

• Find the genetic code for control and rewrite it
• Act your way to new thinking
• Short, early conversations make efficient work
• Use “I intend to…” to turn passive followers into active leaders
• Resist the urge to provide solutions
• Eliminate top-down monitoring systems
• Think out loud (both superiors and sub-ordinates)
• Embrace the inspectors

Competence

• Take deliberate action
• We learn (everywhere, all the time)
• Don’t brief, certify
• Continually and consistently repeat the message
• Specify goals, not methods

Clarity

• Achieve excellence, don’t just avoid errors
• Build trust and take care of your people
• Use your legacy for inspiration
• Use guiding principles for decision criteria
• Use immediate recognition to reinforce desired behaviours
• Begin with the end in mind
• Encourage a questioning attitude over blind obedience

There are so many useful takeaways in this book; it’s a short but engaging read and the direct applicability to the way we manage the people in software development projects is very clear – especially if you’re aiming for truly self-organizing agile teams!

I highly recommend reading Turn the Ship Around to anyone interested in genuinely empowering people in their teams.