The "bad execution" myth.

When a transformation falls over, the post-mortem almost always lands on execution. "The team didn't deliver." "The vendor underperformed." "Requirements kept changing." These are rarely wrong — but they're also rarely the real story.

The real story, in my experience, is that the program was set up to fail in its first ninety days, by people who had the best of intentions, a credible-looking plan, and a steering committee that signed off every step.

Four failure modes show up with disheartening regularity.

Failure mode one: scoping to the budget, not the problem.

A number appears early. It's usually a budget envelope or a minister's commitment date. The scope gets shaped to fit it — not because anyone is dishonest, but because the alternative (going back for more money before the work has started) feels politically impossible.

Six months in, scope and budget part ways. But by then the governance model, vendor contracts, and program narrative are all built on the original envelope. The program can no longer describe the shape of the thing it's actually building.

The healthy version of this conversation is: here is the problem, here is what solving it would cost, here is what we can afford to solve now. Three separate statements. Usually they get collapsed into one.

Failure mode two: governance by committee census.

The second failure is a governance model where every stakeholder group gets a seat and every seat gets a voice. This feels fair, and in the first few weeks it even feels efficient — everyone is aligned.

What it actually produces is a forum where no-one can be overruled, which means no trade-off can be made, which means the program carries every constraint until it collapses under the weight of them. By the time leadership intervenes, the team has learnt not to escalate.

Failure mode three: the silent sponsor.

The sponsor was energetic at kickoff, attended the first two steering committees, and then stopped coming. Now their direct report chairs the forum. The direct report makes no decisions without checking, which doubles the cycle time on every one of them.

Silent sponsorship is the single most reliable indicator of a program in trouble. When I'm asked to diagnose an in-flight program, the first question I ask is not about the plan — it's about who ran the last steering committee.

Failure mode four: deferring the hard decision.

Every transformation has a handful of genuinely hard decisions that sit at the centre of everything else. In a good program, those decisions are made in the first ninety days — not because the team has perfect information, but because the cost of deferring them is too high to bear.

In a struggling program, the hard decisions are sequenced to the end. The plan fills with preparatory work, discovery phases, and design sprints that exist primarily to avoid the moment someone has to say "we are choosing A over B, and here is what B will cost us".

What to do instead.

None of these failure modes are complicated. The fixes are equally unglamorous:

  • Scope the problem before the budget. Even if the budget is non-negotiable, describe what's being traded off. Put it in writing.
  • Collapse governance early. Two forums with named decision rights beat five with consensus-seeking cultures. Every time.
  • Protect sponsor time. If your sponsor can't attend a monthly committee, the program is already in drift. Re-negotiate their involvement, or find a different sponsor.
  • Make the hard decisions early. A wrong decision made in week ten is recoverable. A deferred decision in month nine usually isn't.

Most transformations don't need a new methodology. They need a senior, accountable person in the first ninety days whose job is to keep the program honest about what it's really signing up to. That person is often the difference between a two-year delivery and a two-year apology.