Thought starter
How strategy processes fail
Most strategy processes fail in one of five ways. The failure is usually visible from the agenda before the first workshop.
The process that flatters the current strategy. The process that prizes consensus over dissent. The process that drowns in frameworks. The process that arrives with the answer. And the process that ends in a binder with no decision.
None of them is a failure of intelligence. Each is a failure of design.
In our experience across engagements, choices made about the process at the start set the quality of the strategy at the end. Those choices are rarely examined.
The five ways
The process flatters the current strategy. The work is framed as a refresh, starting from the existing document. Every analysis tests whether the current course still holds. Nobody is asked to argue for a different one.
The output confirms what the executive already believed, with better charts. The warning sign is an agenda with no session on options.
The process prizes consensus over dissent. The stated aim is agreement. Objections are heard and softened. Options that would divide the room are merged until they divide nobody.
The strategy that emerges is the one nobody opposed, and so the one that commits to nothing. The warning sign is a process with no way to bring disagreement out on purpose.
The process drowns in frameworks. The advisers apply every tool they know. The team fills in matrices and the executive reviews them. The two or three uncertainties that decide the strategy get the same attention as the forty that do not. The warning sign is a workplan organised by framework, with no list of questions.
The process arrives with the answer. The recommendation was formed in the first week, often imported from another business or another era. The analysis is then assembled to support it. The answer is fast and confident, and it fits a business that is not this one. The warning sign is a hypothesis that never changed.
The process ends in a binder. The strategy is developed, documented, presented and filed. Nothing is reallocated and no initiative starts within a quarter.
The organisation never hears about it in plain language. A year later it is a record of a meeting. The warning sign is a project plan that ends at the Board presentation.
The four ways choices fail
Underneath the five is a simpler pattern. Strategy is choice, and choices fail to get made in four ways.
The choice is never made. An issue is studied, restudied and referred to the next cycle, because deciding would cost someone something.
The choice appears to be made and falls apart. The room agreed, nobody committed, and execution is patchy from the first month.
The choice is made on bad ground. The data was wrong, or nobody tested the inference from it. The strategy is right until the first contact with the market.
Or the choice is made too late. The process was so long, and the buy-in so exhaustive, that the window it aimed at has closed.
The last one deserves more attention than it gets. A decision that took eighteen months was probably made by default in month six, when the business kept doing what it was doing. A fast decision can still be a careful one. Set a decision date at the start and hold it. That produces a better decision than waiting for certainty, because certainty does not arrive.
What fixes them
Focus the process early on the two or three choices it exists to make, and say what they are in the first week. Bring the tensions out deliberately. Find where leaders disagree, where the stated strategy and the enacted one diverge, and where the incentives pay for something else.
Structure each choice before analysing it. State the options and name what would have to be true for each to be right. Then analyse only the conditions in doubt.
Wind-tunnel the options against a few scenarios. And set the decision date, the reallocation that follows it, and the first three initiatives, before the process begins.
A good strategy process produces a set of choices the leadership has committed to. With them come a budget that has moved and a handful of things that started this quarter. A process designed to produce those looks different from the first day.