The numbers are down, and the pressure to act is real. So the organisation moves – and watch what it reaches for.
Not, usually, a better answer. Often, the certainty of making itself visible.
Benchmark the performance. Define the methodology. Build the dashboard. Codify what the best people do and roll it out. Each of these feels like rigour, like finally getting a grip on the problem. And each produces something you can show a board on Monday – which is a large part of why the system reaches for it.
But notice what the move assumes. It assumes the problem is that the organisation does not yet know what works – that the answer exists and simply has not been captured. So it sets about capturing, defining, codifying. The trouble is that you cannot codify judgement that has not yet formed. And in a system still learning how it decides, that is exactly the situation.
There is nothing settled underneath to make visible yet.
This is the misdiagnosis at the centre of it. A formation problem gets read as an optimisation problem. A formation problem means the system has not yet learned how to decide well under pressure. An optimisation problem means the system knows how, and now needs tightening, measuring, scaling. They look similar from the outside – underperformance, in both cases. But the responses are opposites. Optimisation makes a working system sharper. Applied to a system that has not formed, it does the reverse: it freezes an unstable pattern in place and calls it a standard.
And there is usually a real signal that gets misread to justify it. Something is working – a new approach lifts win rates, a quarter turns. The instinct is to read that as proof the method works: capture it, mandate it, scale it. But early uplift in a forming system is rarely a transferable method. It is judgement starting to emerge. Mandate it too soon and you codify the behaviour while losing the judgement that produced it – consistency that looks like progress for a quarter or two, then strains the moment conditions shift, because what you scaled was the surface, not the reasoning.
This is the part that inverts the usual instinct. When a system is weak, everything says move faster – more definition, more measurement, more visible control. But the weaker the system, the more premature visibility costs, because there is less underneath to survive being frozen.
The sequence holds most strongly exactly where capability is lowest – which is precisely where the pressure to skip it is highest.
And this is not an argument against rigour, measurement, or codification. It is an argument about timing. The same methodology that breaks a forming system stabilises one that has learned to decide. Visible too early, it substitutes for judgement. Introduced once judgement holds, it protects it.
The act is identical. The timing is everything.
So when performance dips, the sharper question is not “how fast can we make this visible and measurable?” It is “has this system actually shown how it decides under pressure yet – and if not, is making it visible going to build that, or just freeze its absence?” Reaching for certainty feels like leadership. Sometimes it is the precise thing that stops the system from ever developing the judgement the certainty was supposed to capture. When your organisation’s commercial system is underperforming, does it reach first for better answers – or for the certainty that comes from making the system visible before it has demonstrated how it actually decides?
This question closes a chapter of The Architecture of Commercial Performance, where the thinking behind it is developed in full.
