Forecast Only What You Have Proven (Investors Delete the Rest)
When you raise a seed or a Series A, an investor mentally erases every unproven line in your forecast and judges you on what survives. Lean on revenue you have not earned yet and the model scares them off.
Founders build forecasts upward. You start with what you have, then add a line for the new product, another for the new segment, another for the channel you are about to try, and you stack them until the number at the bottom looks like a venture outcome. Investors do the exact opposite. They read a forecast downward, deleting every line that depends on something you have not actually done yet, and they judge you on whatever is left standing. You are both looking at the same slide and seeing completely different numbers.
This matters most at seed and Series A, because at those stages you are not funded on vision alone. You are funded on evidence that something repeats. The forecast is the exact place where vision and evidence collide, and for the founder who padded the model with hope, the collision goes badly.
Why a bigger forecast can scare them off
Founders assume a bigger number signals ambition. To an experienced investor it signals one of two things, and both are bad. Either you cannot tell the difference between revenue you have proven and revenue you are hoping for, which means you do not understand your own business. Or you can tell the difference and you are betting they cannot, which is worse. Neither reading makes them want to wire money. The speculative stack you added to look impressive is the precise thing that makes you look unproven.
What actually counts as proven
Proven is a narrow word, narrower than founders want it to be. It means revenue from a motion you have actually run, at a conversion rate you have actually observed, that has actually repeated. Everything short of that is an assumption wearing a revenue line's clothing, and assumptions are exactly what the investor strips out.
- Revenue from a channel you have run, at a conversion rate you have measured
- A cohort that has renewed, so the retention is observed, not assumed
- Pricing real customers have actually paid, more than once
- A new product still on the roadmap, booked as if it ships and sells on schedule
- A second segment you have closed exactly one logo in, scaled like a proven motion
- A signed LOI, a pilot, or a pipeline number treated as committed revenue
The lines investors delete on sight
- A new product that has not shipped, modeled as if it launches on time and sells immediately.
- A second segment or vertical you have one reference customer in, projected like a repeatable engine.
- A new channel with no payback data, assumed to scale at the economics of your best existing one.
- International expansion into a market you have not entered.
- A price increase you have never actually tested on a paying customer.
- Partnerships that are still in conversations, booked as distribution.
Any of these might be real someday. That is not the point. None of them belong in the number you are raising on, because the investor has watched a hundred forecasts that depended on exactly these lines, and almost none of them happened on schedule.
This is the same disease as the top-down market slide
The padded forecast and the “one percent of a ninety billion dollar market” slide come from the same instinct: making a number big by detaching it from evidence. We took apart why the top-down market read as a fantasy in why investors pass, and the cure is the same one we prescribed in the numbers that kill a seed round: build it bottoms-up, from a real customer and a real conversion rate, and never start from the size you wish the number were. A forecast grounded in proven revenue is just the bottoms-up discipline applied to the one slide founders most want to inflate.
How to forecast on proven ground and still show upside
None of this means you have to hide your ambition. It means you have to separate it from your plan, visibly, so the investor never has to do the separation for you. The structure that earns trust is simple.
- Isolate the proven engine: the revenue that comes from a motion you have run and measured.
- Grow it at a rate you have actually sustained, not the rate you wish you could. This is your plan, your base case, the number you are accountable to.
- Put every speculative line in a clearly labeled upside case, on its own, smaller than the base, and never blended into the headline.
- Say out loud which lines are proven and which are bets. Naming the bets is what proves you understand the difference.
- Raise against the proven base plus a sliver, not against the dream. Size the round to a milestone you can actually reach, which is the whole argument in how much to raise.
An investor funds the part of your forecast you could not have faked. Everything stacked on top is decoration on a number they already decided whether to back.
Why this gets stricter at Series A
At seed you can sometimes be funded on a sliver of proof and a compelling story, because there is not much history to point to yet. By Series A the bar has moved to repeatability: the proof is supposed to be the engine, not a hint of one. A Series A forecast that leans on unproven lines reads as a confession that you have not actually found the repeatable motion yet, which is the one thing the round is supposed to demonstrate. It is a stage-fit failure, and it connects directly to whether you are default alive on the revenue you can actually count on, and to whether your traction is the durable kind we described in what actually counts as traction.
So forecast the company you have proven, grow it at a rate you have earned, and quarantine the dreams in a clearly marked upside case. The smaller, honest number is the one that raises. If you want a firm of quantitative analysts to find the speculative lines hiding in your model, separate the proven engine from the wishful stack, and tell you exactly what an investor will delete before they read your total, that is what the forecast review inside Roast My Startup is built to do.
Find the holes before an investor does
Roast My Startup is a firm of AI analysts that tears apart your deck, model, forecast, and data room, then tells you exactly what an investor would use to pass. Brutal first, constructive second.