Ask a finance team what next quarter's revenue will be and you will often get one number. Ask where it came from and it turns out to be three different things wearing the same figure: what the business committed to at the start of the year, what it now expects to happen, and what leadership would like to happen. Those are different questions. A planning process that answers all of them with a single number ends up answering none of them well.
Three numbers, three jobs
The budget is a commitment. It allocates resources and sets the baseline against which managers are held to account. Its value comes from being stable: once agreed it should not move every time conditions change, because a baseline that moves cannot measure anything.
The forecast is a best estimate. Its only job is to be as accurate as possible about what will actually happen, given everything known today. Its value comes from being unstable: it should move whenever the evidence moves. A forecast that has not changed in six months is either very lucky or not being maintained.
The target is an ambition. It exists to stretch performance and is usually set above what is likely. There is nothing wrong with that, as long as nobody mistakes it for a prediction.
What goes wrong when they merge
When the forecast is expected to equal the budget, two things happen. Managers stop reporting bad news early, because a forecast below budget reads as a failure rather than as information. And the forecast loses its purpose, because it can no longer warn anyone about anything. The first sign of trouble then arrives in the actuals, when it is too late to act on.
When the target is treated as the forecast, the organisation plans cash, hiring and inventory around an outcome it did not expect to achieve. Working capital is tied up against sales that were never likely, and the shortfall is discovered as a funding problem rather than a planning one.
Keeping them apart in practice
- Label every number. A figure in a board pack should say whether it is budget, forecast or target. It sounds trivial and is the most effective control on this list.
- Report variance to budget and change in forecast separately. Variance to budget shows how performance compares with the commitment. Movement in the forecast shows what has been learned since last month. They answer different questions and belong in different columns.
- Build the forecast from drivers, not from the budget. A forecast that starts from the budget and adjusts it inherits every assumption in it. One built from volumes, prices, headcount and conversion rates can be challenged line by line, and it shows exactly which assumption moved when the number does.
- Separate the forecaster from the accountable manager where you can. People forecast differently when the forecast becomes their performance review. Finance owning the forecast, with operational input, reduces the pull towards the budget.
- Roll the horizon. A forecast that always ends at the financial year-end gets shorter and less useful as the year goes on. A rolling horizon of four or six quarters keeps the view the same length and makes the organisation look past December.
A useful test
Take the last three forecasts for a single line — revenue, say — and compare each with what eventually happened. If the errors are consistently in one direction, the forecast is biased, and the bias usually points towards the budget. If the errors are large but balanced, the drivers need work. If the forecast barely moved between versions while the actuals did, it was never a forecast.
Separating the three numbers costs very little. What it buys is a planning process in which each figure can be trusted to mean what it says.