Most transformation doesn't fail dramatically. It advances — budgets, platforms, reorgs, dashboards — while the enterprise stays put. Only 12% reach their original ambition. This is about the other 88%, and where transformation stops converting.
No email, no gate. The full argument is below; the paper is the formatted edition.
Most transformation doesn't fail dramatically. It advances — while the enterprise stays put.
Budgets clear. Platforms go live. Reorgs land. Dashboards turn green. On every visible axis — activity, spend, initiatives — the program is in motion. And yet the operating model — how work enters, gets chosen, runs, is proven, and corrects — ends the program roughly where it began.
That is the gap between motion and change. Only about 12% of transformations reach their original ambition. The other 88% don't stall — they keep moving. Motion simply never converts into operating-model change.
Moved Fast. Stayed Put. is a field guide to where that conversion breaks: ten signals across five conversion points, a self-diagnostic to find your earliest fail point, and the operating standard a sponsor can hold — no email, no form, no gate.
of transformations fall short of their original ambition — motion never converted to operating-model change.
reach the ambition — the operating model actually changed.
Moved fast is activity. Stayed put is the operating model.
A program has to convert motion into operating-model change at each point a unit of work passes through. Where it doesn't convert, the enterprise stays put — and a pair of signals tells you exactly where.
A transformation can recite the strategy; the test is whether it changed how work is admitted. Two tells: the strategy arrived but the operating requirements didn't — objectives are funded but no one owns the condition each one needs — and urgent work still enters through side doors the model never governs.
A portfolio can carry a rank order without a single real choice. Two tells: everything is still a priority — nothing was actually stopped, no capacity freed — and when the terms change (a discount, a compressed deadline) staffing, scope, and price don't move with them, so margin quietly absorbs the exception.
The target operating model can be visible while the work stays exactly the same. Two tells: a front-line team describes the org chart instead of a workflow that runs differently, and technology scaled ahead of the operating logic — no one owns the exceptions by rule.
Governance can report a program without being able to change it. Two tells: the steering review only reports status — no priority, funding, or decision right ever moves — and a booked benefit reconciles as a number but can't be traced to the work that produced it.
An initiative can close without the enterprise learning anything. Two tells: correction still depends on the same few heroes pulled in every time it breaks at the same seam, and last year's fix resurfaces this year under a new name.
Get, Sort, Do, Prove, Improve — GSDPI — the chain a unit of work travels from governed entry to provable value. Motion has to convert to operating-model change at each stage. Where it doesn't, the enterprise stays put.
Answer them in order. The first one you can't answer cleanly is where transformation stopped converting — and where the correction begins.
Silence means the work is funded but unowned. That's a conversion gap at Entry, not a strategy gap.
No answer means you have a ranking, not a decision — nothing converted at Choice.
If they describe the org chart, you changed the picture — not the work.
If finance can reconcile the number but not the work, it's counted — not proven.
Recurrence means you closed a deliverable, not the operating condition behind it.
Symptoms are loudest downstream, at Execution or Evidence. The fix begins at the earliest point that never converted — not at the loudest symptom.
The full ten signals and the sponsor standard are laid out in the paper.
You just read the argument. The formatted edition lays out all ten signals, the self-diagnostic, the worked margin-protection read, and the full sponsor standard — read it inline below, or take the PDF for sending or printing. No email, no form, no gate.
Four ways to engage with the argument — at the altitude that fits where you sit.
Put a live transformation in front of the five conversion points. A scoped read locates the earliest point that never converted and returns the one move that scopes the correction — not another initiative.
A working session for a board, executive team, or PE portfolio on the 88/12 gap — and the operating standard that separates a program that moved fast from an enterprise that actually changed.
Bring the five sponsor questions to a program in flight — before the next board read, results call, or diligence cycle runs them for you and the narrative can't answer.
Counted, Not Proven makes the proof argument beneath this one: a number you can report is rarely a number you can prove. The proof companion to this conversion argument.
Read Counted, Not Proven →