Evrenus Fırat Polat · ·
A revenue forecast does not need to be sophisticated to be useful. Its real job is to give you enough warning to act (hire, cut back, chase receivables harder, or build a cash cushion) before the month you are forecasting actually arrives.
Start with what's already committed
The most reliable part of any forecast is revenue that is already agreed: signed contracts, recurring invoices, retainers, and accepted quotes for work already scheduled. List these first, by the month you expect to invoice them.
Add likely work, discounted honestly
Next, add work that is probable but not confirmed: proposals out, and repeat clients who usually order around this time. Do not count these at full value. Weight them by how likely they genuinely are. A proposal you would put at 50% odds counts for half its value in the forecast.
Separate invoiced from collected
Revenue invoiced in March is not cash in March. If your clients typically pay in 30 to 45 days, shift the expected cash into the month it will actually arrive. Your DSO is a good guide to how far to shift it. Forecasting both, revenue and expected cash, shows you timing gaps that a revenue-only forecast would hide. This connects directly to cash flow and working capital.
Use three scenarios, not one number
A single forecast figure implies more certainty than exists. A cautious case (committed revenue only), an expected case (committed plus weighted likely work), and an optimistic case give you a range to plan against. The cautious case is the one to make sure your fixed costs can survive.
Worked example: a 3-month forecast
A small consultancy has fixed monthly costs of $9,000. For the next three months:
Committed
| Month 1 | Month 2 | Month 3 | |
|---|---|---|---|
| Retainers | $5,000 | $5,000 | $5,000 |
| Signed projects | $3,000 | $2,000 | $0 |
| Committed total | $8,000 | $7,000 | $5,000 |
Likely work (value × probability)
- Proposal A: $6,000 at 50% = $3,000, expected in month 2.
- Proposal B: $4,000 at 25% = $1,000, expected in month 3.
- Repeat client usual order: $2,500 at 70% = $1,750, expected in month 3.
Scenarios (revenue)
| Month 1 | Month 2 | Month 3 | |
|---|---|---|---|
| Cautious | $8,000 | $7,000 | $5,000 |
| Expected | $8,000 | $10,000 | $7,750 |
| Optimistic (all won at full value) | $8,000 | $13,000 | $11,500 |
Expected month 3: $5,000 + $1,000 + $1,750 = $7,750. Optimistic month 2: $7,000 + $6,000 = $13,000. Optimistic month 3: $5,000 + $4,000 + $2,500 = $11,500.
Fixed costs of $9,000 a month against the cautious case: month 1 −$1,000, month 2 −$2,000, month 3 −$4,000. Total shortfall over three months: $7,000. That tells you now that if no new work lands, you need about $7,000 in reserves, or you need to win work or cut costs by month 3.
Cash timing. Clients pay in about 30 days. Month 1 invoices of $8,000 arrive in month 2, and so on. Expected cash in month 1 is therefore last quarter's tail, which you also need to count. Add it to the model. A rise in DSO from 30 to 45 days could push another half month of revenue out. Test it in the DSO calculator.
Step-by-step: build the forecast
- Pull a list of all committed revenue by month.
- Add open proposals, with a probability for each. Use your past win rate, not hope.
- Add expected repeat work.
- Shift each item to the month cash will arrive.
- Subtract fixed costs and known variable costs.
- Create the three scenarios.
- Write down one decision each scenario would trigger. For example: "If cautious cash falls below one month of costs, pause hiring."
- Save the file with the date, so you can compare later.
Compare against reality every month
The most valuable part of forecasting is checking it afterward. Where you were consistently too optimistic or too cautious tells you how to adjust next time. After a few months, your forecasts get noticeably more accurate simply because you have learned where your own guesses tend to be off. Calculate your win rate from past proposals, and your real collection time from invoice aging.
Common mistakes
- Counting hopes as revenue. Weight by probability.
- Ignoring seasonality. Look at the same months last year.
- Forecasting revenue but not cash. Late payments can cause a shortfall even when sales are strong.
- Too much detail. A simple monthly table beats a complex model you never update.
- Forgetting tax and fixed costs. Revenue is not profit. See gross profit vs. net profit.
- Never revisiting. A forecast is a living document.
FAQ
How far ahead should I forecast?
Three to six months is practical for most small businesses. Go further only if you have long contracts.
What if I have no history?
Start with committed work only and add a conservative estimate for new work. Update monthly as you gather real data.
How do I estimate probability?
Use your past win rates by type of lead. If you win 2 of 10 cold proposals, use 20%, not 50%.
Do I need software?
A spreadsheet is enough. Move to specialized tools only when the spreadsheet becomes hard to maintain.