How to build a driver-based revenue forecast for a startup
A driver-based forecast builds revenue from the things you can actually influence: how many customers you win, what they pay and how long they stay. It is the single biggest difference between a model that convinces investors and one that does not. Here is a practical method you can follow in a spreadsheet.
Step 1: Pick the unit of revenue
Decide what one unit of revenue is for your business. For a subscription product it is an active customer paying a monthly fee. For an e-commerce business it is an order. For a services company it is a project. Everything else in the forecast hangs from this unit.
Step 2: List the drivers
- Volume: leads, conversion rate, new customers per month.
- Price: average revenue per customer or per order, with any planned price changes.
- Retention: churn rate or repeat purchase rate.
- Capacity: sales headcount, delivery capacity or inventory limits that cap volume.
Step 3: Build a customer roll-forward
For each month calculate opening customers, plus new customers, minus churned customers, equals closing customers. Revenue is then average customers multiplied by average revenue per customer.
This single roll-forward makes churn visible. A small increase in churn compounds over time and is often the most important sensitivity in the model.
Step 4: Tie acquisition to cost
New customers do not appear for free. Link marketing and sales spend to the number of customers acquired through a cost per acquisition, so the forecast cannot grow without the spend that should accompany it.
Where you have no history, use a range and label it as an assumption until you have pilot data.
Step 5: Add capacity limits
Growth that needs ten salespeople should show ten salespeople in the hiring plan. Check that the volume in your forecast is possible with the people, stock or delivery capacity you have budgeted.
Step 6: Test the forecast
- Change conversion by plus and minus 20% and see what happens to revenue and cash.
- Raise churn by a few points and check the effect on year-three revenue.
- Delay the first sales month by a quarter and check runway.
- Compare unit economics against your own early data.
Common mistakes
- Growing revenue by a fixed percentage each month.
- Ignoring churn in the first year.
- Using a market-size percentage as a revenue forecast.
- Letting revenue grow without matching headcount or spend.
Common questions
What is a driver-based forecast?
A forecast in which revenue is calculated from operational inputs such as customers, price, conversion and retention, rather than from a growth rate.
How far ahead should a startup forecast revenue?
Monthly for the first one to two years, then quarterly or annual out to three to five years.
What if I have no historical data?
Use ranges, source them from pilots or comparable businesses, and label them as assumptions. Update them as real data arrives.