How to forecast sales with no historical data
New teams and new products face a chicken-and-egg problem: accurate forecasting wants history, and history only accrues by forecasting and measuring. You have to start somewhere honest.
The answer is to lean on bottom-up rep judgment and reasonable benchmarks, present ranges instead of false precision, and treat the first few quarters as data collection.
Short answer
Forecast with no historical data by combining a bottom-up build (each rep commits deal by deal) with stage-based probabilities borrowed from industry benchmarks, then present a conservative range rather than a single number. Track every actual against the forecast from day one so that within a quarter or two you can replace assumptions with your own data.
Step by step
Build bottom-up from real deals
Without history, aggregate rep-level commits deal by deal rather than trusting a top-down model. Reps closest to the deals have the best available signal when there is no data to lean on.
Borrow benchmark probabilities
Use industry or analog benchmarks for stage win rates and cycle length as a starting point, clearly labeled as assumptions. A borrowed 20 percent stage probability beats a number pulled from nowhere.
Present a range, not a point
With thin data, false precision misleads. Forecast a conservative-to-optimistic range so leadership plans around uncertainty instead of a single fragile number.
- Conservative: only high-confidence commits
- Likely: commits plus a share of best case
- Optimistic: full weighted pipeline
Instrument everything from day one
Log every deal's stage, date, and outcome from the first week. This is the history that makes your next forecast real, so capture it deliberately.
Replace assumptions with data fast
After one or two quarters, swap borrowed benchmarks for your own measured conversion and cycle length. Your forecast gets sharply better the moment it runs on your own numbers.
Honesty beats precision early
The worst thing a new team can do is present a confident single number built on assumptions. Leadership makes hiring and spending bets on it, and when reality diverges, trust is lost. A clearly labeled range preserves credibility.
Treat the first quarters as an investment in data. Every logged deal, even the lost ones, is raw material for a forecast that will soon stand on its own.
How Ardovo helps
Ardovo is alive with data from day one, so even a new team forecasts against real deals, and Rook starts learning your conversion and cycle patterns immediately, replacing benchmark assumptions with your own data as quickly as deals close.
Frequently asked questions
How do you forecast sales for a brand new team?
Build bottom-up from rep-level deal commits, borrow stage probabilities and cycle length from benchmarks as clearly labeled assumptions, and present a conservative-to-optimistic range. Log every deal from day one so you can replace assumptions with your own data within a quarter or two.
Can you forecast without any past data?
Yes, but with wider uncertainty. Use rep judgment and benchmark probabilities, and communicate a range rather than a precise number. The key is instrumenting every deal now so future forecasts run on real history instead of borrowed assumptions.
How long until a new team can forecast accurately?
Usually one to two full sales cycles. Once you have a few hundred closed deals, you can measure your real stage conversion and cycle length and replace benchmark assumptions, which sharply improves accuracy.