Sales forecast template: building one you can defend in a review

A sales forecast template is the structure you use to say how much you expect to sell and when. The structure matters less than one property: every figure should trace back either to a specific deal or to a stated driver. Forecasts built by typing a plausible growth rate into a row cannot be challenged, corrected or learned from — they can only be believed or disbelieved.

There are two honest ways to build one, and which you use depends on how you sell. Few, large, named deals call for a bottom-up pipeline forecast. Many small transactions call for a driver model: traffic, conversion, price. Mixing the two in one sheet without saying which is which is where most forecasts become unauditable.

What the template needs

• One row per month, across the horizon you actually plan against — twelve months is plenty.
• For a driver model: units and price as separate inputs, with revenue as a formula. Never a typed revenue figure.
• For a pipeline model: deal, stage, value, probability and a weighted column that multiplies the last two.
• An assumptions block, visually separate, holding every input anyone might question.
• Colour or formatting that distinguishes inputs from formulas, so a reviewer knows what is safe to change.
• A cumulative column, because the running total is what tells you whether the year is on track.
• Last year's actuals alongside, if you have them — the single most useful sanity check available.
• An actuals column filled in monthly, so forecast and outcome sit side by side rather than in different files.

Stage probabilities are the part most quietly abused. If every deal at "proposal" is scored at 60% because that is what the template said, the weighted total is arithmetic rather than a forecast. Derive the percentages from your own history — what share of proposals actually closed last year — and revisit them annually. If you have no history, say so and use round numbers you can defend as guesses.

Building it

1. Decide which model fits: pipeline for a handful of named deals, drivers for volume. Write which one at the top of the sheet.
2. Put every input in one assumptions block. If a number lives inside a formula, nobody will find it when it needs changing.
3. Build revenue from units × price, or value × probability. Typed revenue is the single clearest sign a forecast cannot be trusted.
4. Add last year's actuals in a parallel column and look at the implied growth. If it is unlike anything you have ever achieved, revise before the meeting rather than during it.
5. Produce three cases — base, low, high — by changing only the assumptions block.
6. Fill in actuals monthly and keep the original forecast visible. Overwriting it destroys the only feedback loop you have.
7. At quarter end, compare forecast to actual by month and write down the bias. Most teams are consistently wrong in one direction, and knowing the size of that bias is worth more than a better model.
8. Prune stale deals before every review — a pipeline full of deals that stopped moving inflates every weighted total.

Forecast, target and plan are three different numbers

A target is what you have committed to. A plan is what the business budgeted against. A forecast is your honest estimate of what will happen. Teams that collapse these into one number end up forecasting the target, which means the forecast carries no information — and the first time it matters, finance discovers the gap too late to react. Keep the three in adjacent columns and let the difference be visible; that difference is the actual subject of the review.

Ettex Sheets is a straightforward home for this: import an existing XLSX with formulas intact, 150+ functions with instant recalculation, conditional formatting to flag a month falling behind plan, charts from any range for the cumulative curve, pivots when you want the forecast by rep or by product, cell comments with threads so a manager can question a probability against the cell it sits in, and version history so the forecast as submitted in March is still recoverable in June. The deals behind a pipeline forecast live in Ettex CRM with a visual pipeline and change history, closed revenue is invoiced through Ettex Invoices, and the commentary that accompanies the numbers belongs in Ettex Docs.

Plainly: Ettex does not forecast for you. There is no forecasting engine, no weighted-pipeline roll-up from CRM into a sheet, no scenario modelling and no AI prediction. You export the deals and maintain the model yourself. For a small sales team that is a ten-minute monthly job; a large team with many reps wants a sales platform that does the roll-up automatically.

A forecast needs something underneath it. How to build a simple sales pipeline covers the stages, what moves a deal from one to the next, and how few stages you can get away with.

A forecast is only as honest as the records under it. CRM data hygiene puts numbers on that — duplicate rate, completeness of the decision fields, and the share of contacts nobody has spoken to in a year.

Why forecasts lose credibility

• Revenue typed rather than driven, so no question about volume or price can be answered.
• Probabilities inherited from a template rather than from your own close rates.
• Deals left in the pipeline months after they stopped moving.
• The forecast quietly equal to the target, which makes it uninformative by construction.
• Actuals overwriting the forecast, destroying the record of what you predicted.
• One case only, so the discussion becomes whether the number is right rather than which assumption drives it.
• No comparison of forecast to actual over time, so the same directional bias repeats every quarter.

Keep the forecast, the plan and the report as three separate documents that are compared, not merged. A sales plan template states the intent for the quarter; a sales report template states what closed. When the three are combined into one spreadsheet, the forecast quietly drifts toward the target and stops being a forecast.

A sales forecast template earns trust by being checkable: inputs in one block, revenue from formulas, probabilities from your own history, and last month's forecast still visible next to what actually happened.