The forecast got worse when it got bigger
One CFO's model grew from one tab to eleven and the variance to plan got wider. A forecast you can't rebuild from memory has stopped steering.
A bigger forecast is supposed to be a better one. More lines, more accuracy. It rarely works that way.
One CFO ran a model that started as 1 tab and 4 assumptions. Three years later it held 11 tabs and 57 assumptions, and the variance to plan was the widest it had ever been. The model did not get smarter. It got more confident.
Every new line, vendor cohort, headcount step-up, churn by segment, was a guess wearing the costume of a calculation. Guesses compound. Eleven tabs of compounded guesses give you a number that is precisely wrong instead of roughly right. That is the same trap as four green metrics sitting on top of a red quarter: more instrumentation, less truth.
The one-tab version was honest. Four numbers, carried in her head. When something moved she felt it inside a week. The eleven-tab version was not felt. It was operated. The board got a deck instead of a person who could answer in the room.
A forecast is not there to be comprehensive. It is there to be rebuilt from memory in five minutes. Once you cannot defend it without opening the file, it has stopped steering the business and started standing in for a promise you never made.
She went back to one tab. Same business, fewer lines. Variance to plan dropped by half the next quarter. Not a smarter model. She had stopped fooling herself.
The test: sketch it on a napkin in front of an investor. If it survives without laughter, keep it. If it needs a second monitor to read, it stopped being a forecast. Call it bookkeeping with ambition.