I have sat on both sides of this appointment more times than I can count: as the executive being brought in to fix a stalled programme, and as the person briefing the board on why the previous appointment did not work. The pattern is consistent enough that I no longer think of it as bad luck or a bad hire. I think of it as a structural failure in how organisations define the role before they fill it.
Most Chief Transformation Officers and transformation-mandated CTOs do not fail because they lack capability. They fail because the role they accept on day one is not the role the organisation actually needs them to perform — and nobody discovers the gap until month nine or ten, by which point the credibility required to close it has already been spent.
01The role is defined twice, and the two definitions do not match
When a board or CEO creates a transformation leadership role, the job description is written by HR or by the sponsoring executive, usually drawing on a generic template: strategic vision, stakeholder alignment, change leadership, technology modernisation. It reads well. It is also almost entirely silent on the thing that actually determines success — where the authority to make binding operational decisions sits, and whether the person in the room with the P&L owners can direct them or can only advise them.
The second definition of the role is the one that exists inside the organisation's actual decision rights: who can stop a project, who can reallocate budget mid-year, who can override a divisional head on a sequencing call. That definition is never written down. It is discovered through experience, usually the hard way, in the first contested decision the new leader tries to make.
When those two definitions diverge — and in my experience they diverge in the large majority of appointments — the transformation leader spends the first two quarters operating on the assumption of authority the job description implied, while the organisation operates on the assumption of authority its actual governance structure permits. Nobody notices until the first major decision needs to be forced through, and the new leader discovers they have influence rather than power.
02Three patterns explain almost every failure I have seen
After more than 200 transformation programmes across 65 markets, and after having both succeeded and personally failed at points in this role, I can group nearly every leadership failure I have observed into three categories. They are not mutually exclusive, and the worst appointments suffer from all three simultaneously.
The advisor masquerading as an operator. This is the appointment of someone — often from a Tier-1 consulting background — who is exceptional at diagnosing the problem and structuring the plan, but who has never personally owned a P&L, run an operating unit, or been accountable for a missed quarterly number with their own name attached to it. They produce excellent strategy documents. They struggle the moment the work requires telling a divisional president that their pet initiative is being deprioritised, because they have never had to live with the consequences of that conversation themselves. The organisation senses this within a few months, and authority drains away accordingly.
The operator without strategic range. The inverse failure is equally common: a credible internal operator is promoted into the transformation role because they know the business, but they have never built or led an enterprise-wide change programme before, and they default to running transformation the way they would run a single function — sequentially, conservatively, and without the cross-functional reach the mandate requires. They are trusted, but they are not equipped, and the programme quietly narrows in scope until it stops resembling a transformation at all.
The leader who is never given the mandate to fail forward. This is the pattern boards underestimate most. A genuinely capable leader is appointed, with real authority and real range — but is held to a delivery timeline and a risk tolerance calibrated for steady-state operations rather than transformation. The first setback, which is normal and expected in any programme of this scale, is treated as evidence the appointment was wrong rather than as the predictable cost of doing difficult work. The leader spends their remaining tenure managing perception rather than the programme, and the transformation never recovers the momentum it lost defending its own legitimacy.
03What the first 90 days should actually establish
The leaders I have seen succeed — and the periods in my own career when I have done this well — share a common discipline in the opening months that has nothing to do with the transformation roadmap itself.
They force an explicit, written conversation with the CEO and board about decision rights before a single workstream is mobilised: which decisions the transformation leader can make unilaterally, which require sign-off, and who adjudicates when the transformation leader and a P&L owner disagree. This conversation is uncomfortable, which is exactly why most organisations skip it. Skipping it does not remove the ambiguity — it simply defers the conflict to a moment when it is far more expensive to resolve.
They establish, publicly, what failure tolerance looks like for the programme. Not as a vague gesture toward “psychological safety,” but as a specific, pre-agreed answer to the question: what magnitude of setback in month four is expected and absorbed without triggering a confidence crisis at board level? Without that answer agreed in advance, the first real setback becomes a referendum on the leader rather than a normal feature of transformation work.
They identify, within the first quarter, which existing senior leaders will operate the transformed state once the programme concludes — and they co-design with those people from day one rather than handing them a finished blueprint at the end. The leaders who fail almost always design in isolation and socialise later. The leaders who succeed treat the eventual operators of the new model as co-architects of it, because those are the people whose support determines whether anything survives past the engagement's formal end date.
Get the mandate right before the appointment is made, and the first twelve months become the period where trust is built. Get it wrong, and the first twelve months become the period where it is spent.Dan Collins
04The accountability question boards rarely ask
There is one further structural problem worth naming directly, because it sits underneath everything above: most organisations measure a transformation leader's success against the plan delivered at programme close, rather than what remains active in the business eighteen to twenty-four months later. This is the wrong measure, and it produces the wrong incentives.
A transformation leader optimising for a successful close-out presentation will, entirely rationally, prioritise the initiatives that demo well over the initiatives that require sustained operational discipline long after the leader has moved on. A transformation leader who knows they will be judged on what survives two years out behaves completely differently — they spend disproportionate energy on capability transfer, on embedding ownership with the people who will run the new model permanently, and on resisting the temptation to claim victory the moment the steering committee stops meeting.
Boards that want this role to succeed should change the measurement before they change the person in the chair. Asking a new appointment to succeed against a short-term, presentation-oriented mandate, while privately hoping for durable, multi-year change, sets up the next failure before the new leader has even started.
05Why this gap is widening, not narrowing
It would be easier to treat this as a static problem with a known solution, but two recent shifts are making the leadership gap harder to close, not easier.
The first is the speed at which AI and automation are compressing transformation timelines. Boards that once accepted a multi-year horizon for enterprise change now expect visible progress within a couple of quarters, because the technology itself moves that fast and competitors are demonstrating early wins publicly. That compression rewards exactly the wrong behaviour described above — it pushes transformation leaders toward initiatives that demo well quickly, at the expense of the slower, less visible work of embedding capability. A leader without a clear, board-agreed mandate on time horizon is now under more pressure than ever to trade durability for optics, and fewer boards are pausing to ask whether that is the trade they actually want.
The second is the changing shape of the talent pool itself. A generation of leaders who came up through pure technology delivery now sits alongside a generation who came up through strategy and general management, and organisations are increasingly trying to solve the range problem — operator versus advisor — by hiring two people instead of one: a CTO focused on the technical build and a separate transformation or change leader focused on the organisational side. This can work, but only if the two roles are genuinely integrated under a single accountable owner for the commercial outcome. More often, in my experience, it recreates the exact split described earlier between diagnosis and delivery, except now formalised into two job titles rather than one — which makes the accountability gap structural rather than incidental, and considerably harder to correct once the organisation chart has been built around it.
Neither trend changes the fundamentals. Both raise the cost of getting them wrong, and both make it less likely anyone notices the wrong appointment has been made before real damage is done.
06The appointment decision this should change
None of this argues for a particular professional background. I have seen internal operators and external appointments both succeed and fail, in roughly equal measure, because the determining variable is rarely pedigree. It is whether the organisation did the unglamorous work of defining the actual mandate — authority, failure tolerance, and the time horizon for judging success — before it filled the role, and whether the person appointed has the range to operate at both the strategic and operational level simultaneously rather than excelling at only one.
Get that definition right before the appointment is made, and the first twelve months become the period where trust is built. Get it wrong, and the first twelve months become the period where it is spent — usually beyond the point of recovery, regardless of how capable the individual sitting in the role happens to be.