The Strategy on the Shelf: Why Execution Drifts From Intent — and Better Documents Won’t Fix It
A strategy is not what an organisation writes down; it is what its programmes actually choose when the two are in conflict.
The Binder Nobody Opened
Somewhere in every large organisation there is a binder. It is handsomely produced — a strategy refreshed at an offsite, argued over for two days, signed off with genuine conviction, and circulated to the leadership team with a covering note about the journey ahead. Eighteen months later the transformation it was meant to drive is well underway. The programme has a director, a plan on a wall, a steering committee that meets fortnightly, and a status report that is, on balance, green. And the binder sits on a shelf, its spine uncracked, describing a set of intentions that no one in the delivery machine could now recite from memory.
This is not a story about a bad strategy or a weak team. I have watched it happen to good strategies run by capable people, which is precisely what makes it worth understanding. The document did not fail because it was wrong. It failed because a document is not a strategy — it is a photograph of one, taken at a moment, and the programme it launched is a moving system that began ageing the photograph the day it was framed.
The gap between what a transformation sets out to do and what its programme actually does has a name in the literature — the strategy-to-performance gap — and a familiar set of villains: poor communication, weak sponsorship, unrealistic plans. Those are real. But they are symptoms. The deeper mechanism is structural, and it operates quietly, from the inside, in organisations that are doing everything the textbook told them to.
Why the Document Dies
Consider what actually happens to a strategy once the applause fades. It is handed to a programme, and in that handover it passes between two populations who rarely share a room. The people who wrote it — the strategy function, a few executives, perhaps an advisory firm — speak in outcomes: become the lowest-cost operator in the segment; shift the revenue mix toward services; consolidate to one operating model. The people who now own its delivery speak in deliverables: migrate the billing platform; stand up the shared service centre; retire fourteen legacy applications. Between the outcome and the deliverable sits a translation, and in most programmes no single person owns it.
Watch where the translation loss shows up. A programme I would recognise anywhere sets out with five strategic priorities and, by its second year, tracks twenty-two workstreams through its stage gates. Ask which workstream serves which priority and you will get confident answers for perhaps half of them, plausible ones for a few more, and silence for the rest. Nobody decided to fund initiatives that no longer traced to the strategy. It happened one reasonable exception at a time — a regulatory must-do here, a sponsor’s pet project there, an integration that turned out to be harder than scoped — until a material share of the budget was being spent on work the original strategy would not have chosen.
The governance machinery cannot catch this, because it was never built to. A stage gate asks whether a workstream is on time, on budget, and on scope. It is exquisitely designed to answer those three questions and structurally blind to a fourth: does this still serve the bet we made? The traffic-light board is the tell. It can shine green across the top while the single initiative that carried the strategy’s central wager — the one genuinely hard thing the whole exercise existed to achieve — is quietly deferred to a later phase, because it was the hardest to schedule and the easiest to justify moving. The board rewards what is measured. Strategic intent is not on the board.
A programme optimises for what its governance measures. If governance measures milestones, spend, and RAG status — and almost all of it does — then under pressure the programme will protect the milestone and sacrifice the intent, every time, and its reporting will look healthy while it does so.
And pressure always comes. Somewhere in year two a hard trade-off arrives: the platform migration is slipping, and something must give. In that meeting, the strategy is not in the room. What is in the room is a delivery plan, a burn rate, and a committee that will be asked next month why a milestone was missed. Of course the milestone wins. It wins not because anyone forgot the strategy but because the strategy had no representative at the table and no claim on the decision. That is the moment the binder truly goes to the shelf — not when it is filed, but when the first serious trade-off is resolved without reference to it.
The Case for Discipline — and Its Limits
The confident answer to all of this is discipline, and it is not wrong. The strategy-focused organisation, we are told, cascades intent through a balanced scorecard; a strategy map draws the causal line from the boardroom objective down to the operational measure; a dedicated office of strategy management keeps the translation honest. Do the cascade properly — give every workstream a line of sight to a strategic objective, and every objective a measure — and the gap should close. This is sound advice, and any programme that ignores it deserves what it gets.
But I have come to believe the cascade treats the symptom and misreads the disease. Every method in that family shares a hidden assumption: that strategy is a thing you get right once and then propagate — a signal broadcast outward and downward, needing only a cleaner channel. The scorecard cascades a snapshot. It makes the translation from outcome to deliverable explicit and visible, which is a real gain, but it freezes that translation at the moment of cascade. When the world moves — when the platform proves harder, when a competitor forces the hand, when the regulatory must-do consumes the capacity that the strategic bet needed — the scorecard keeps faithfully reporting against a map of a world that no longer exists. Rigour applied to the wrong object produces confident measurement of the wrong things. We have all seen a beautifully cascaded scorecard, green throughout, sitting on top of a programme that had long since stopped pursuing the strategy it was cascading.
“Rigour applied to the wrong object produces confident measurement of the wrong things.”
The honest objection to my own argument is that the alternative sounds like an excuse to abandon planning — that if the document is disposable, why write it at all, and what stops “continuous reinterpretation” from becoming a licence to drift wherever the pressure of the week pushes? It is a fair challenge, and it points at the real work. The document is not disposable; it is indispensable as the record of the bet. What is disposable is the belief that writing it down was the act of strategy, rather than the beginning of it.
Where the Strategy Actually Lives
Here is the reframe that a decade of watching this has forced on me. A strategy is not what an organisation writes down; it is what its programmes actually choose when the two are in conflict. The document is a hypothesis about how to win. The programme is the experiment. And an experiment that never revisits its hypothesis under new evidence is not disciplined — it is merely stubborn.
If that is true, then the fix is not a better document or a cleaner cascade. It is to give the strategy standing inside the machinery where decisions are actually made. In practice that means a few unglamorous things.
- Compress the bet into the programme’s own language. Not the twelve-page strategy, but the one or two sentences naming what this transformation must make true for it to have been worth doing — expressed as the delivery team would say it, not as the boardroom said it. If it cannot survive that compression, it was never operational to begin with.
- Give the intent a seat and a voice. Someone in the trade-off meeting must hold the standing to say that move protects the milestone and breaks the bet — and be owed an answer. This is a role, not a document: the person licensed to invoke the strategy against the plan.
- Make the gate ask the strategic question. Alongside on-time, on-budget, on-scope, every gate should force a fourth: does this still serve the bet, and if not, what changed? A “no” is not a failure to be hidden; it is the most valuable signal the programme can produce, because it is the early warning the traffic-light board will never give.
- Treat the document as a living hypothesis. Revisit it not annually as ceremony but whenever the programme makes a decision the original strategy would not have predicted. Those decisions are the data. Either they are drift to be corrected, or they are the strategy learning something — and telling the two apart is the actual work of governing a transformation.
None of this is exotic, and none of it requires a new methodology. It requires accepting that the strategy-execution gap is not closed by pushing intent harder down the pipe. It is closed by pulling the intent into the room where the choices are made, and keeping it there.
The Test
There is a simple test for whether a transformation is still serving its strategy, and it has nothing to do with the quality of the binder. Walk into the programme, find the person resolving this week’s hardest trade-off, and ask them what the strategy requires of this decision. If they can answer — in a sentence, in their own words, without reaching for the shelf — the strategy is alive. If they reach for the shelf, it is already too late, and no amount of green on the board will tell you so.