· 2 min read
Sales forecasting for small businesses: a practical guide
Build a sales forecast you can explain: clean pipeline data, stage probabilities, close-date windows, targets and a monthly accuracy check.
A sales forecast answers one question: how much do we expect to sell, and when? For a small business that drives hiring, cash planning and how hard to push on new leads. You don't need a data team; you need clean pipeline data and a repeatable method.
Step 1: Make the pipeline trustworthy
A forecast is only as good as the deals behind it. For every open deal check:
- a realistic value,
- the correct stage,
- an expected close date in the future,
- an owner and a next action.
Close deals that have gone cold. This step usually changes the forecast more than any formula.
Step 2: Use stage probabilities consistently
Give each stage a default win probability and apply it to every deal in that stage. Multiply value by probability to get the weighted value (see our guide to the weighted pipeline).
Step 3: Group by expected close date
Sum gross and weighted value for windows such as this month, next 30, 60 and 90 days. Present both numbers: the gross figure shows what's possible, the weighted figure what's likely.
Step 4: Compare with targets
Set monthly targets and compare them with won bookings (won deal value by actual close date). The gap between target and bookings plus weighted pipeline tells you whether you need more leads now.
Step 5: Check accuracy every month
At month end, compare what the forecast said with what actually closed. If you consistently over-forecast, lower stage probabilities or tighten your definition of each stage. Accuracy improves through this loop, not through a cleverer formula.
Keep it honest
- Bookings are not accounting revenue; talk to your accountant about recognition and cash.
- Don't add overlapping windows together.
- A forecast is an estimate, not a promise; show the deals behind it.
Doing it in Zyrft
Zyrft (Pro) shows gross and weighted forecasts by close date, monthly targets against won bookings, and a forecast-accuracy report comparing forecasts with results. How the forecast works · Targets and reports · Start free
Written by the Zyrft team. Examples use fictional companies; this article uses no external statistics.