John Kotter’s early research on organizational change found roughly seventy percent of transformations fail to reach their stated goals. Nearly three decades and hundreds of case studies later, a University of Oxford Saïd Business School study, run with EY and drawing on more than 1,600 global technology leaders, found the failure rate essentially unchanged. The Oxford research points less at strategy quality and more at a specific set of human factors: purposeful vision, adaptive leadership, psychological safety, and genuine collaboration. Transformations that had those in place succeeded at meaningfully higher rates than transformations that didn’t, even when the underlying strategy looked identical.

One pattern shows up continuously in the transformations that stall: a leader who moves fast, decides alone, and treats speed as proof the transformation is working. It often is working, for a while. The leader is smart, the direction is usually right, and early momentum looks like evidence the approach is sound. Then the transformation hits its first real obstacle — a mid-level manager who doesn’t buy in, a market shift the original plan didn’t anticipate, a cross-functional dependency that wasn’t flagged — and the team built around that leader has nothing to offer except waiting for the next directive. They were never asked to own a piece of it. They were asked to follow.

Why the Hero-Leader Move Looks like Progress Right Up Until it Isn’t

This is where the adaptive-leadership half of the Oxford findings gives more direction. A leader who centralizes every decision is optimizing for a specific kind of visible progress: fast decision-making, clear direction, momentum a board can see in a quarterly update. What that leader isn’t building is the collective capacity to keep making good decisions once the environment shifts in a way the original plan didn’t predict. Speed and adaptability aren’t the same thing, and a transformation that only has speed runs out of runway the moment conditions change.

Psychological safety research offers the other half of the explanation. Amy Edmondson’s decades of work on the topic show that safety alone doesn’t produce good outcomes — it must be paired with real accountability, clear standards for what good work looks like once someone speaks up. A hero-leader culture usually fails this pairing in one specific direction: it isn’t unsafe to speak up in a hard-line, punitive sense. It just doesn’t change anything, because the leader already decided. After enough of that, people stop bothering, and a transformation that depended on distributed judgment is left running on one person’s bandwidth.

Two Ways this Fails, and They Don’t Look the Same from the Outside

A hero-leader transformation can fail loudly — the leader burns out, makes a visible bad decision under too much decision load, or leaves, and the whole effort stalls because nothing was built to survive their departure. That version is easy to spot after the fact and easy to misdiagnose as a personnel problem rather than a structural one.

The quieter failure is more common and more expensive. The transformation technically executes — the new system goes live, the reorg happens, the initiative gets marked complete in the tracker — while almost nothing about how people work changes underneath it. Leaders check the box. Teams comply on the surface and revert to the old way the moment attention moves elsewhere. This is the failure mode that looks like success in the meeting where it gets announced and looks like nothing happened at all eighteen months later.

Where PRESSURE is Built to Interrupt This

DLP’s PRESSURE methodology — Problem, Reflect, Evaluate, Strategize, Sacrifices, Undertake, Reframe, Engage — is sequenced specifically against the hero-leader failure pattern, and two steps do most of that work.

Reflect exists because a hero-leader transformation typically skips it. The step asks a team to examine assumptions, history, and what’s already been tried before committing to a path — which is exactly the step a fast-moving leader is most tempted to shortcut, because it slows down the visible momentum that speed-as-progress depends on. A team that reflects together before committing has already built the shared understanding a hero-leader transformation never gets around to.

Engage is the step that most directly targets the quiet failure mode. It asks who needs to be brought in, and when — not once, at the launch announcement, but continuously, as the transformation moves. A hero-leader approach treats stakeholder engagement as a communication task: tell people what’s happening. PRESSURE treats it as a design question: who needs real decision input, not just notice, for this to survive contact with the parts of the organization the leader doesn’t personally control.

Sacrifices is the step most hero-leader transformations never name out loud. Every real transformation costs something — a process someone liked, a decision right someone used to hold, a habit that was comfortable. A leader who decides alone can absorb that cost invisibly, in their own head, and never force the organization to name what it’s giving up. That avoidance feels efficient. It also means nobody outside the leader ever processed the trade-off, which is precisely the group that must live with it once the transformation is underway.

None of this argues against strong, decisive leadership. It argues against mistaking decisiveness for the entire job. A transformation built to survive its first real obstacle needs a leader willing to move fast and a structure that doesn’t require speed to be the only thing holding it together.