How to Forecast Revenue When You Have No Sales History

Every early-stage founder eventually has to put a revenue number in front of a board or an investor with almost no data to back it up. Most reach for the same thing: a top-down wish. "The market is 10 billion, we only need 1%, so that is 100 million." That is not a forecast. That is a fantasy with a decimal point.

When you have no sales history, you forecast from the bottom up, from activity and conversion, not from the size of the ocean.

Build the model from activity, not the market

A bottoms-up model starts with what you can actually do, then applies honest conversion assumptions at each step. Something like:

• Meetings you can realistically book per month.
• The percent that become qualified opportunities.
• The percent of those that close.
• Average deal size and sales cycle length.

Multiply it through and you get a revenue number that is tied to actions, not hope. The magic is not the output number. It is that every assumption is now visible and arguable. When the board pushes back, you are debating "can we really book 40 meetings a month," which is a real conversation, instead of defending an imaginary market share.

Track leading indicators and tighten as data arrives

With no history, your first assumptions are educated guesses. That is fine, as long as you treat them as guesses and hunt for the real numbers fast. Track the leading indicators, meetings booked, meeting-to-opportunity rate, opportunity-to-close rate, cycle length, from day one. Each closed deal and each lost one is data that replaces a guess.

Within a quarter or two you stop forecasting from assumptions and start forecasting from your own thin but real conversion rates. This is also where a lot of founders fool themselves, because a pipeline built on inflated stage assumptions lies to you. I wrote about that in your pipeline is lying to you: a forecast is only as honest as the conversion rates underneath it.

Separate commit from best case

The last discipline is refusing to give a single number. Give a range with named layers. Commit is what you would genuinely bet the quarter on, built on conservative conversion. Best case is the upside if things break your way. Never blend them into one hero number, because that number becomes the plan, and then you hire and spend against a fantasy.

Do this and your forecast becomes a tool instead of theater: something you steer the business with, correct monthly, and can actually defend. If you want a model built right and maintained as the real numbers come in, that is core to what a fractional CRO does in the first ninety days.