The Forecast Accuracy Formula
Forecast accuracy tells you how close a prediction was to reality, expressed as a percentage where 100% is a perfect forecast. The standard formula is:``` Forecast Accuracy = 1 - (|Forecast - Actual| ÷ Actual) ```
If the forecast was $4.2M and actual closed revenue was $4.5M, the calculation is:
``` |4.2 - 4.5| ÷ 4.5 = 0.067 1 - 0.067 = 0.933 = 93.3% accuracy ```
The absolute value in the numerator means overforecasting and underforecasting produce the same accuracy score.
Why Most Teams Measure This Wrong
The most common error is measuring forecast accuracy only at the end of the quarter using the final submitted number. This produces a single data point with no diagnostic value. Accurate measurement requires:
| What to Track | Why |
|---|---|
| Forecast submitted at a fixed point (e.g., day 1 of the quarter) | Consistent measurement date across quarters |
| Forecast submitted at end of month two | Tests late-stage conviction |
| Accuracy by rep, separate from team rollup | Manager haircuts can mask rep-level drift |
| Accuracy by deal segment | SMB and enterprise errors are structurally different |
What Accuracy Thresholds Mean
There is no universal threshold. What counts as acceptable depends on your deal volume, cycle length, and the decisions your forecast is driving. A small team closing large enterprise deals in long cycles will naturally forecast with more error than a high-volume SMB team closing in days.
The practical question is fitness for purpose: can a finance team make capital allocation decisions from your forecast, or does the error range make it unusable? If forecast error is large enough to change whether a hire gets approved or a campaign gets funded, that is the threshold that matters for your business. Establish your own baseline from historical data and set targets relative to it.
Accuracy, Bias, and Variance Together
Pair forecast accuracy with forecast bias to understand whether errors are directional, and with forecast variance to understand spread across reps or segments. A team with consistently high accuracy at the aggregate level may still have high variance at the rep level, which creates fragility when key reps miss.
For teams building systematic forecasting processes, forecast accuracy benchmarking across rolling quarters is the foundation. Set a target, track it weekly, and trace anomalies before the quarter ends.
Frequently Asked Questions
What is the standard formula for forecast accuracy?
The most widely used formula is: Forecast Accuracy = 1 - (|Forecast - Actual| / Actual). Multiply by 100 to express as a percentage. This measures how close the forecast was to reality without caring about the direction of the error.
What is the difference between forecast accuracy and forecast bias?
Accuracy measures the magnitude of the error regardless of direction. Bias measures the directional tendency: whether a team consistently forecasts too high or too low. A team with large but randomly distributed errors has low accuracy and low bias. A team with systematic but small errors has moderate accuracy and high bias.
How do I track forecast accuracy over rolling quarters?
Record the submitted forecast at a fixed point each quarter, typically at the start or at the end of month two, then compare to the actual close at quarter end. Track both the raw accuracy score and the directional error over time. A rolling four-quarter view smooths seasonal anomalies and reveals structural drift.
Put these metrics to work
ORM builds custom revenue forecast models that turn concepts like forecast accuracy formula into prescriptive action for your team.
Schedule a Demo