The forecasting approach you need depends on the questions you want to answer. I keep two distinct forecasts.

Forecast 1: the next three months

Use actual numbers to estimate profit. Focus on real revenue sources — existing clients or pipeline — and predictable costs. Team composition rarely changes much in three months, so staffing costs are easy to predict.

You’ll start to understand current performance, learn the gap between predictions and reality, and identify where to improve. You start to learn the difference between what you think is going to happen and what actually happens.

Forecast 2: the financial year

Use budgets rather than exact figures. This longer-term model takes in:

  • Expected monthly revenue at the start of the year
  • Anticipated growth during the year
  • The team size you’d need to hit those targets
  • The associated salary and overhead costs

It tells you whether your growth plans actually increase profit without taking on too much risk.

Building blocks for both forecasts

Known revenue

Track ongoing and nearly-certain projects from proposals and client agreements. The main variable is timing — money typically arrives later than expected. Account for your earning model, whether time-and-materials or value-based, recognising that capacity constraints exist regardless of business model.

Pipeline revenue

Active pipeline is current inquiries at various stages. Weight deals by win probability, using either instinct or historical conversion data:

  • Early-stage leads: ~5% probability
  • Qualified prospects: ~20%
  • Proposal stage: ~50%

Future pipeline is revenue from work not yet in the market. This involves either projecting growth rates or planning specific investments and expected returns.

Cost of delivery

Include spending on freelancers, third parties, and project-specific expenses like advertising. If you forecast revenue, forecast the corresponding delivery costs.

Salaries and staff costs

Account for salaries divided by twelve, plus taxes, pensions, bonuses, and overhead increases. Include your own minimum monthly compensation.

General overheads

These typically remain stable unless planned changes occur. Set budgets for variable investments like marketing and track actual spending against them.

Make it a ritual

Make forecasting a monthly ritual. Review actual results against predictions to identify execution gaps and inform tactical adjustments. Run scenarios to stress-test assumptions: What happens if that client leaves? Can you afford 2 new hires?

Forecasting develops as a skill unique to each agency, ultimately revealing how good you are at keeping revenue flowing.