How to forecast sales
A sales forecast predicts how much you will close in a period. Its value is not precision to the dollar, it is reliability: a forecast leadership can plan around. That comes from clean data, a consistent method, and a discipline of measuring accuracy over time.
Short answer
Forecast sales by starting with a clean pipeline, applying a method that fits your data (stage-weighted, historical, or rep commit), categorizing deals into commit, best case, and pipeline, then reconciling bottom-up rep forecasts against top-down history. Review weekly, snapshot the number, and measure accuracy so the forecast earns trust.
Step by step
Start with a clean pipeline
Audit out zombies, fix stages, and correct fake close dates. A forecast is only as good as the pipeline underneath it.
Pick a forecasting method
Choose the approach that fits your data maturity, and often blend more than one.
- Stage-weighted: value times stage probability
- Historical: apply past conversion to current pipeline
- Rep commit: judgment-based commit and best case
Categorize every deal
Sort deals into commit, best case, and pipeline so the forecast has ranges, not a single fragile number.
Reconcile top-down and bottom-up
Compare the sum of rep commits against what history says the pipeline should produce. Investigate the gap.
Snapshot and measure accuracy
Record the forecast at a fixed point and compare to actuals each period. Accuracy tracking is what turns a guess into a trusted number.
How Ardovo helps
Ardovo forecasts from clean pipeline using stage-weighting, historical conversion, and rep commits together, snapshots each forecast, and scores accuracy by rep. Rook reconciles bottom-up and top-down and flags the deals most likely to slip.
Frequently asked questions
What is the best sales forecasting method?
There is no single best method; mature teams blend stage-weighted, historical, and rep-commit approaches and reconcile them. Judgment-only forecasts are biased, and pure formula forecasts miss deal-specific context. Combining them and measuring accuracy produces the most reliable result.
How often should I forecast?
Update the forecast weekly in most teams, with a formal commit at a consistent cadence such as each week or the start of the period. Frequent updates catch slippage early; a consistent snapshot point makes accuracy measurable.
Why is my forecast always wrong?
Usually because the pipeline is dirty, close dates are guesses, and no one measures accuracy to correct bias. Clean the pipeline, tie close dates to buyer-confirmed events, snapshot the forecast, and coach the reps whose numbers consistently miss.