You can build a real sales forecast with zero years of history. The trick is to stop forecasting off gut feel and start forecasting off your open pipeline, weighted by how deals at each stage actually convert.
Most early-stage founders skip forecasting entirely. They look at their pipeline, feel optimistic, and tell the board or themselves "we'll probably close $80K this quarter." That number is usually wrong in one direction: too high. Not because founders are bad at math, but because a flat pipeline total ignores the single biggest variable in any deal: how far along it actually is.
Why gut-feel forecasts fail
A gut-feel forecast treats every open deal as equally likely to close. A deal that just had a first call and a deal with a signed verbal commitment get counted the same way. That's the core error.
The fix is weighting. Instead of summing total pipeline value, you multiply the value at each stage by the historical win rate for that stage, then add the results together. A $50,000 deal sitting in "discovery" is not worth $50,000 to your forecast. If deals at that stage close 10% of the time, it's worth $5,000. A $50,000 deal in "verbal commit," closing 70% of the time, is worth $35,000. Same pipeline total, very different forecast.
The weighted-pipeline method, step by step
This is the same method revenue teams at much larger companies use, scaled down to work with a spreadsheet and no dedicated sales ops person.
- List your stages. Keep it to 4-5: something like discovery, qualified, proposal sent, verbal commit, closed.
- Pull your win rate per stage. Use pipeline win rate, not overall win rate: deals that closed won from that stage divided by all deals that ever passed through that stage. If you don't have enough closed deals yet to calculate this per stage, borrow a public benchmark to start (10-15% for early discovery, 40-60% for proposal sent, 65-80% for verbal commit are reasonable seed-stage SaaS ranges) and replace it with your own data the moment you have 15-20 closed deals.
- Multiply and sum. Value at each stage × win rate for that stage, added across all stages, gives you the forecast for the period.
- Re-run it weekly. Pipeline shifts stage constantly. A forecast from three weeks ago is a historical artifact, not a live number.
This method only forecasts near-term, open pipeline. It won't tell you about deals that haven't entered your funnel yet, so pair it with a separate top-of-funnel target for anything beyond 60-90 days out.
A second number worth tracking: sales velocity
Sales velocity gives you a sanity check on the weighted forecast and doubles as a single metric to report to a board or co-founder.
The formula: (number of open opportunities × win rate × average deal size) ÷ average sales cycle length in days.
Worked example: 40 open opportunities, a 25% win rate, a $2,000 average deal size, and a 30-day average sales cycle gives you (40 × 0.25 × $2,000) ÷ 30 = $667 in expected revenue per day, or roughly $20,000 for the month. If your weighted-pipeline forecast lands wildly above or below this number, one of your inputs is wrong, usually the win rate or the deal size assumption.
The three inputs to start tracking today, even manually
You don't need a CRM with automation to do this. A spreadsheet with these three columns, updated weekly, is enough:
- Stage per deal, updated the moment it changes, not retroactively at month end.
- Deal value and close date estimate, re-estimated honestly each week rather than left at the original guess.
- Stage-to-stage conversion, tracked as a running count so your win rate per stage gets more accurate every month instead of staying a guess forever.
The mistake that wastes the most time here is treating this as a one-time setup. A forecast built once and never updated is worse than no forecast, because it creates false confidence.
What to do this week
Pick your 4-5 stages, pull every open deal into them honestly (not where you wish they were), and apply seed-stage benchmark win rates if you don't have your own yet. Run the weighted-pipeline math once. Then commit to a 15-minute Monday pipeline review where you re-stage every deal and re-run the number. That single weekly habit will do more for forecast accuracy in the first 90 days than any tool you could buy.
Frequently asked questions
How accurate can a sales forecast be with no historical data?
Expect wide error bars at first, often 30-50% off in either direction. Accuracy improves fast once you have 15-20 closed deals to calculate your own stage-level win rates instead of relying on benchmarks.
What's the difference between pipeline win rate and overall win rate?
Pipeline win rate divides closed-won deals by everything that ever entered that stage, including deals still open or lost. Overall win rate only compares closed-won to closed-lost. Pipeline win rate is the one to use for forecasting, since it doesn't ignore deals that are still moving through the funnel.
Do I need a CRM to forecast sales accurately?
No. A spreadsheet with stage, deal value, and close date per opportunity is enough to run the weighted-pipeline method. A CRM helps once deal volume makes manual tracking error-prone, usually somewhere past 25-30 open deals at once.
How often should I update my sales forecast?
Weekly, at minimum. Pipeline stage shifts fast enough at early-stage velocity that a forecast older than a week is usually already wrong.
Why is my pipeline total so much higher than my actual forecast?
Because pipeline total assumes every open deal closes, and most won't. The gap between raw pipeline and weighted forecast is the clearest sign of how much cushion (or risk) is really in your number.
Should I use my quota as my forecast?
No. Quota is a target, not a prediction. A forecast built from actual weighted pipeline will almost always be lower than quota, and that gap is exactly the information a founder needs to see early enough to act on it.
Building this once, badly, and running it every week will teach you more about your sales motion than a perfect model you never update.