The moment a forecast stops being an estimate

You can always tell when a forecast has stopped being an estimate. The questions change. Before, people ask what it is built on. After, they ask why you are behind.

Nobody announces the switch. There is no meeting for it.

A forecast starts life as an estimate. Somebody builds it out of assumptions, and every one of them is arguable. Then it travels. It goes up, it gets discussed, it gets signed off, and somewhere on that trip it becomes a commitment.

I have been on both sides of this. I have carried a number I knew was one of three plausible futures, and I did not always keep saying so once it was agreed. Partly because caveats fall off things as they move upward. Partly because a qualified number loses to a confident one in the room where it is being defended.

Certainty travels better than accuracy does.

So it firms up with every level it passes, and the assumptions stay behind in a file no one opens again.

This came out of a reply to Monday's post. Somebody asked how we landed on the numbers, and I realised that is not the part I find interesting. What the number is built on matters far less than what happens to it afterwards.

If you are on the receiving end of one of these, the useful question is not accuracy. It is what it rests on, and whether the person who put it there still believes it today.

I have never found a clean way to keep the caveat attached as it moves. Curious whether anyone has.