Threshold.

Insight

Forecast revisions are a habit, not an event

If you only revise forecast at quarter-end, you are revising it under maximum pressure and minimum information. The fix is upstream.

14 March 2026 · 7 min · Alice Thornton

Most forecast accuracy problems are not commitment problems. They are timing problems. The seller is right about the deal — they are wrong about when.

We measure forecast accuracy at two checkpoints in every retainer: forecast committed in week 1 of the quarter, and actual closed at week 13. The gap between those two numbers, expressed as a percentage of the original forecast, is the only metric that matters.

When we run that calculation across the prior eight quarters of a typical engagement we find a particular pattern: the forecast in week 1 is broadly defensible, but the seller does not revise it down at week 6 even though they know by then that two deals will slip. They wait until week 11 and revise then under board pressure. That last-minute revision is what kills accuracy.

The fix is to make week-6 revision a ritual, not a confession. The Monday review meeting at week 6 should ask one question of every late-stage deal: 'is this still the same close date and the same value as last week?' If the answer is no, the rep revises now. The board sees a smaller, earlier, more defensible revision. The seller is not punished. The team learns the cadence.