Leading Metamorphosis: How to Make an Expert Team Thrive in a Complex Environment (the answer is not ‘more autonomy’)

There’s a fundamental tension in leadership paradigms: should an expert team be provided full autonomy (the Wheel approach) or should a seasoned expert operate as a hierarchical leader (the Star approach). Currently, the Wheel approach is getting a lot of traction in software teams (e.g. in agile development), but there are good arguments against it (and for the Star approach). Further, a typical team doesn’t operate in either maxim (and thus takes a compromise approach), which I claim is worst of both worlds.

In this text I propose a fourth, a more dynamic approach to team leadership (the Metamorphosis approach) and claim that the role for a modern leader in a team is to enable the dynamism it requires by changing culture with role interventions.

Three leadership approaches to expert teams

In this chapter, I’m partly paraphrasing a HBR article by Kudaravelli, Faraj and Johnson and adding some ideas of my own in to the mix.

Most teams grow organically to the “Star” approach, letting the most experienced specialist to lead the team. Lifting a seasoned expert to a team leader position is a typical solution and a way to show admiration to the best specialist in an organization. The Star approach is most clearly visible in military leadership, and there is a reason why it’s still widely maintained in many critical contexts: the roles are clear and the leader is solely responsible in execution, dividing his work to smaller tasks for the team to conduct, and then putting the pieces back together. In a functioning Star team, there is very little excess work done and the team operates quite predictably. A hierarchical team can rely on its deep knowledge and expertise and doesn’t end up in an analysis paralysis. The flipside is that the hierarchical structure might make the leader a communication bottleneck and hinder new ideas and younger talent.

In the “Wheel” teams, a team self-manages. Currently, it’s quite popular to demand teams to be more autonomous and there indeed are benefits, such as ability to avoid biases and reduce conflict. The Wheel teams have less myopia in new approaches and can utilize larger number of tools. However, autonomous teams are no silver bullet. They increase the inter-team discussion links and might burden the specialists who are supposed to be doing the actual work. In addition, they cause unclear responsibilities and even avoidance of admitting mistakes.

Because of the pitfalls of both the Star and the Wheel approach, most team take a third, compromising approach, either consciously or unconsciously, and end up becoming pseudoautonomous or semihierarchical. This is oftentimes the worst approach, causing problems of both the Star approach and the Wheel approach to emerge.

Due to these challenges, Kudaravelli, Faraj and Johnson propose that teams should decentralize for design phase and centralize for execution phase, to get the best of the both worlds.

This is true, but it’s easier said than done. There is an inherit inertia in any team to change. Furthermore, a once-changed team is quite like Gregor Samsa, transformed to a large insect, and certainly never transforming back to what it used to be. So how to transform any team to a metamorphosing superteam that oscillates between the Wheel and the Star. How to lead metamorphosis?

Here’s how.

Three steps to lead metamorphosis

Leaders are oftentimes afraid of changing the culture of their teams. For some reason, it’s thought that culture change is difficult to do. But it isn’t hard. Quite the opposite. Changing the culture might be the only and is definitely the most effective leverage a leader has. The leader is the person responsible for changing the culture. This is the most important and sometimes the only responsibility of the leader.

All leaders have three goals: (1) to change the team culture to reduce information sharing costs, (2) to change the team culture to reduce waste of time and resources, and (3) to change the team culture to reduce frustration and increase opportunities for co-actualization.

I’ve incorporated three culture interventions in my leadership approach at think tank Demos Helsinki, where I’m responsible for the ten-person Democracy and Capabilities team. They are called “Extreme ownership”, “Failure and Rescue”, and “Directing roles”.

The ultimate goal for leading metamorphosis is to enable team members’ participation to a (work) community without guilt. Currently, most work is done because leaving it undone would cause feelings of guilt and shame. These are not beautiful motivations, nor do I believe that they are sustainable. There are stronger forces in play when a team comes together to explore curiously a commonly chosen topic, and to co-actualize someone’s vision from that exploration.

Extreme Ownership

A friend of mine once mentioned that there’s a fairly strange interview with two Navy Seal soldiers in a news website (it was probably Business Insider), where they tell the listener to take responsibility for everything, saying: “In the morning when the alarm hits of like a baseball bat in your face, you can either stay in the bed, or you can take responsibility and attack the morning. Choose responsibility: attack the morning!”

If you, like me, are amused by a testosterone overload, you can watch a short TEDx talk of one of the authors.

Even if the everyday of a Navy Seal is probably quite different from what happens in the Demos Helsinki office, I found their approach useful although not quite in the way they intended.

The point with Extreme ownership is that the leader takes responsibility of everything. If a mistake is made, the leader should avoid hiding behind his ego and admit the mistake, no matter who made it. If the mistake is made in your team, it’s your responsibility.

I find the story is more useful in leaderless teams than in hierarchical teams. If no one takes responsibility, problems arise. In agile development teams the one who is most frustrated by the broken toilet seat if responsible for fixing it. In some teams this might cause ad hoc leadership, but in others just frustration and avoidance of responsibility.

When the team adopts the team-wide doctrine of the extreme ownership, everyone knows that everything that happens in this team is in his or her responsibility. Everyone is responsible for everything. There are no leadership gaps, because leadership is everywhere.

This might sound like a lot of duplicate work, but that is not the case. The absolutely crucial aspect of the extreme ownership is “leading the ego”. This means that you let others to lead you, when they take responsibility. Their mistakes are your mistakes, so you help them, but you shouldn’t override them. Instead, everyone is responsible for giving people the opportunities to lead.

The first important act in leading Metamorphosis is both taking responsibility and letting others to take responsibility. So, it’s all about controlling one’s ego and understanding when to take charge, and this applies to everyone in a team. Just listen to captain Sully of the unfortunate flight that ended up in the Hudson river taking responsibility:

“My aircraft.”

“Your aircraft.”

“We are going to Hudson.”

This is how it should be done.

Failure and Rescue

All work includes complexity and uncertainty. I have even claimed that this is what work is about, all else is just algorithms and instructions (In Finnish: Link). To learn in a complex and uncertain environment, we are told to take risks. And it’s true, progress can happen only if we take risks, test and learn. But no one wants a surgeon ready to take risks, so how do the surgeons learn? And further, most of us work in occupations where we are not allowed and we should not be allowed to fail. Start-ups are a different story, but even in my team in Demos Helsinki we work with limited budgets and important topics and certainly refuse to fail. So how do we take risks and learn?

Most of the following ideas are based on Atul Gawande’s talk on Failure and Rescue. Here is a short text about it. In that text, Gawande links to an article about hospital mortality rates, which points out that hospitals that do the most to minimize risks don’t have any smaller mortality rates than other hospitals. Remarkably, the feature that best predicts a low mortality rate is ability to recover from mistakes, not ability to avoid mistakes.

The great are distinguished from the good not by their ability to fail less, but by their ability to rescue more.

The reasoning behind this is that problems will arise. Just think about captain Sully, a certainly capable pilot, having a bad luck and hitting birds right after ascending. The case Sully also demonstrates that leadership happens only after a crisis emerges. (Read this amazing New Yorker piece on leadership)

The secret in rescuing is to admit mistakes early. This allows the maximum time for recovery and to pivot. As Gawande writes, this applies not only to small teams, but also in policy making. Alarmingly, it might be that we are becoming worse at admitting that we don’t know everything. The first time we collectively had the idea that we don’t know everything, we started to ask what we don’t know. Then the scientific revolution happened. This time there’s always more data to find correlations and more conflicting findings, writings and ideas to confuse us to believe that we know something. We don’t.

Nevertheless, the second act of Metamorphosis approach is to admit constantly that things might go wrong. So again, it’s about ego: not only taking responsibility and avoiding mistakes, but being ready to rescue from those mistakes when they start to emerge.

Directing roles

One of the only leadership books I’ve read was Mamet’s On directing (small clarification: It might be that the book I read was a Finnish collection of texts by Mamet). It was stunning because it’s not written as or intended to be a leadership book. It’s a book particularly about how a theatre director should not confuse playwrights or actors with modifications or creative approaches to the manuscript. Rather, he or she should keep everything as plain as possible, let the actors play and stay away from the play as much as possible. However, the book doesn’t advocate autonomy in teams. Instead, it’s a book about the very small but important role of the director in managing the roles and stage positions of actors in the play.

Mamet writes: the sole responsibility of the director is to fine-tune how actors move on the stage. Who they talk to, and where they stand when talking so that the play looks natural to the audience (but typically not to the actors). Mamet strongly opposes the so-called method acting, where the actor tries to combine self with the role. Instead, the actor should understand the role: the motivations built in to each role in the play. This is the other responsibility of the director: making sure that the actors understand the motivations in the way that each act plays out naturally.

The goal of a leader in leading metamorphosis is similar to that of a director. It’s crucial to understand who should be in the team in the first place, and act accordingly. Furthermore, it’s important to communicate each role and motivations and intentions the role carries to each person. Oftentimes leaders communicate characteristics of the team member to the team member (i.e. “you are strong in negotiations”) but I think this is waste of time and might cause people to take wrong career and development paths. Instead, the leader should communicate the motivations of different roles (“this time you want to be seen strong in negotiations”). When the motivations of the different roles are clearly defined, the collaboration within the team and oscillation between the Wheel and Star approaches are natural.

Thus, the third act of Metamorphosis approach is to direct the roles from outside the team, especially pointing out the significance of listening and empathy in the design phase and ability to follow a single vision in the execution phase, and let the team to manage itself naturally within these role motivations and stage positions.


The steps taken thus far might seem like a fairly nice leadership advice, but what do they have to do with combining the Wheel and the Star approaches to teams? With ability to both carry responsibility together in the design phase and collaborate to elect a leader (“my aircraft”) by putting egos aside in the execution phase the team operates in a dynamic fashion. The leadership can change dynamically especially when the team accepts that it’s more important to focus on rescue than to minimize risks, and that the team is together responsible on the outcome: it’s the vision of the leader-elect, not the director, they are all thriving for (“We are going to Hudson”).

The responsible (outside) leader of such a team can work as one of the team members and let his or her ego also give space to others, or better still, to step aside to let the team self-manage after initial intervention to create the culture of metamorphosis. This is done by directing roles, not people, and letting the action to play out.

The ultimate goal for leading metamorphosis is enabling participation to a work community without guilt. The leader is the person responsible for changing the culture from guilt-based collaboration to a co-actualization. This culture change reduces information sharing costs, waste of time and resources, and frustration and increases opportunities for co-actualization.