Build a business forecast

Forecast the mechanism, not the line

Hollis Field Services maintains heating systems for landlords. Last year it billed $4.8m. The owner wants next year's plan to say $6m, and the first draft is a spreadsheet with 25% typed into one cell and every other line dragged to the right.

A forecast turns beliefs about operations into financial consequences. Start with drivers: customers, units, price, capacity, staffing, collection terms. A growth percentage can be a scenario input, but it is not an explanation of why growth will happen.

Build revenue from the activity that produces it. Hollis has 40 technicians with 1,600 available hours each. At 75% billable utilization and $100 of realized revenue per billable hour, revenue is 40 × 1,600 × 0.75 × 100 = $4.8m. To reach $6m, something physical has to change: more technicians, higher utilization, more hours or a higher rate. Fifty technicians at the same rates give 50 × 1,600 × 0.75 × 100 = $6.0m, so the plan is really "hire ten technicians", with hiring costs and a ramp. Utilization cannot climb forever; nobody bills 110% of the day.

Costs have drivers too, and they are not the revenue drivers. Technician pay follows headcount, not billed hours. Parts follow jobs. Rent stays flat until the depot is full, then steps. Collection terms are a driver as well: landlords who pay in 60 days turn the same revenue into much later cash. Label every assumption as an estimate, a contract term or an observed relationship, so the reader knows which numbers are evidence and which are hope.

Changing one number is a sensitivity; changing a linked set is a scenario. A real downside for Hollis combines fewer jobs, a worse mix of small repairs, landlords paying later, and technician pay that does not fall with revenue. Build a base case, then alternatives that hang together as a story. Weak sales and slow payers tend to arrive together, which is why a downside is a story and not a haircut.

A target is a wish; a forecast is a consequence. The owner's $6m is a target: a desired outcome. The forecast is what follows from the stated beliefs about headcount, utilization and rate. Management guidance to investors is a third thing, often a target dressed as a forecast. Treating the target as an unbiased estimate hides the resources and risks needed to reach it.

This unit runs the bridge question forward instead of backward: drivers to revenue on the Scorecard (income statement), then profit to cash. A forecast you cannot trace to a driver is a line, not a forecast.