Ask a consultant which leadership style works best in a transformation, and the honest answer is: it depends which week you’re asking about. A recent literature review out of Lyon Metropole mapped six models of leadership (distributed, transformational, adaptive, authoritarian, benevolent, recognition-based) and tied each one to a specific managerial tension it’s built to hold. What the research doesn’t spell out, because it’s working from the outside looking in, is that a single transformation mission usually moves through several of these registers in sequence. The skill isn’t picking one. It’s noticing when the one you’re using has stopped working and reaching for the next.
Here’s roughly how that sequence tends to play out, drawn from missions I’ve run.
Phase 1: the kickoff, where authority has to be assumed before it’s earned
Every transformation starts with a moment where somebody has to say “we’re doing this” before the room has any reason to trust that it’ll work. On a recent mission opening a new industrial manufacturing site, the early weeks weren’t about consensus. They were about naming a raison d’être and putting rituals in place before the team had any track record together to justify them. That’s the tension the research calls assumed authority versus acknowledged vulnerability, and at kickoff, assumed authority usually has to win. Nobody co-constructs a launch date. Someone sets it, clearly, and holds the frame while the team is still forming.
The mistake at this stage isn’t being directive. It’s staying directive past the point where the team has enough shared footing to want a say.
Phase 2: giving the launch a reason, not just a plan
Almost immediately after the directive kickoff, the register has to shift toward meaning: not just what’s happening and by when, but why it matters enough to commit to. This is where a purely transactional style (targets, deadlines, sign-offs) starts to run out of road. The tension here is individual sense-making versus collective efficiency, and it’s resolved by connecting the mission to something the team can actually recognize as theirs, not just the client’s. Same industrial site mission: the rituals that got put in place at kickoff only stuck because people were given a reason to want them, not just a schedule for performing them.
Phase 3: the murky middle, where nobody has the playbook
This is the phase every transformation hits and most underestimate. On a public-sector reorganization I ran across two sites, the initial brief was “communication issues between the teams.” It wasn’t. Sitting in the room instead of sending a questionnaire surfaced something the org chart couldn’t show: unclear ownership at the boundaries between roles, informal go-to people carrying responsibility without formal authority, a management layer caught between two sets of expectations nobody had reconciled since the original reorg.
None of that has a known technical fix. This is exactly the register the research calls adaptive leadership: distinguishing a problem you can solve by deferring to an expert from one that requires the whole group to sit in the discomfort of not knowing yet, together, until something new gets built. Trying to skip this phase by imposing a technical solution from the first mission is the single most common way transformations quietly fail: the org chart gets redrawn, and nothing about how people actually cooperate changes.
Phase 4: once the team can actually hold the pen
Only once that adaptive work has surfaced the real fault lines does it make sense to open the room up. On the same reorganization, the second phase moved from diagnosis to design, through workshops where staff and management co-created the new structure directly rather than being handed one. This is distributed leadership, and the research is specific about why it works here and not earlier: it requires clear roles, institutional trust, and real regulation, all of which the adaptive phase just spent weeks building. Skip straight to “let’s co-construct this together” before that trust exists, and you get silence instead of participation.
I’ve watched this happen in real time on a different mission: a questionnaire sent to twenty people asking for their input ahead of a September seminar came back with a single response. Not because the invitation was badly worded. Because co-construction assumes a level of team readiness that isn’t automatic; some teams are mature enough to fill that pen in, others are still waiting to be shown it’s safe to pick it up.
Phase 5: after the launch, when the real work is staying visible
The riskiest moment in most transformations isn’t the kickoff. It’s six months later, once the novelty has worn off and the org chart looks “reorganized” on paper, but nothing about daily cooperation has actually shifted. This is where benevolent and recognition-based leadership earn their keep, and where they’re most often skipped, because they don’t produce a visible milestone the way a kickoff or a workshop does. Continued presence to monitor how a new structure is actually being taken up, catching friction early instead of waiting for it to surface as a formal complaint, and recognizing effort and contribution rather than only outcomes: none of this is glamorous, and all of it is what determines whether people end up genuinely committed to the new way of working or merely compliant with it.
The register is the diagnosis
Laid out this way, the five phases look almost too clean, and real missions rarely move through them in a straight line; they loop back, stall, and sometimes skip a phase that later needs revisiting. But the underlying discipline holds regardless of the order: the leadership register a mission needs at any given moment is itself a diagnosis of where that mission actually is. A team that goes quiet when you open the floor isn’t ready for distributed leadership yet, whatever the workshop agenda says. A team that’s still being handed directives six months after launch is being under-led, not over-managed. Reading which phase you’re in, honestly, is most of the job. Picking the “right” leadership style in the abstract is the part that barely matters.
Based on “Leadership : 6 modèles qui montrent que rien n’est figé,” Direction de la Prospective et du Dialogue Public, Métropole de Lyon, January 2026, and missions run through The Change Office.
If you wish to be supported during your transformation projects, we invite you to schedule a meeting with The Change Society Office.
