Why Investors Ask for Five Year Projections at Pre-Seed
Not because anyone believes the numbers. The model is read as evidence of how a founder thinks: whether they know which few variables drive the business, whether the opportunity is large enough under their own assumptions, and whether the amount being raised is consistent with the plan just described.
What is actually being evaluated
Everyone in the meeting knows a five year forecast from a company with no product is fiction. Asking for it anyway is not a failure of imagination on the investor's part.
The model is a thinking artefact. It reveals things a pitch narrative can conceal.
Whether you know what drives the business. Every business comes down to a handful of variables. How many customers you can reach, what fraction convert, what they pay, how long they stay, what it costs to serve them. A founder who has identified the right handful understands their business. A founder whose model has forty inputs and no hierarchy has not yet worked out which ones matter.
Whether the opportunity is large enough on your own numbers. Investors need outcomes of a certain scale to make their model work. If your own assumptions, stated optimistically, produce a business that tops out modestly, that is a decisive answer and it came from you rather than from them.
Whether the ask matches the plan. You said the raise covers eighteen months. The model implies a burn that exhausts it in ten. That inconsistency is read as either not having done the arithmetic or not being straight about the runway, and neither is recoverable in the same meeting.
How you handle uncertainty. A founder who presents fiction with total confidence is signalling something. So is one who cannot commit to any number at all. The useful posture is stating the assumption, saying why it is plausible, and naming what would change it.
How to build one that helps you
Build bottom up, from things you can act on. Start with the activities you control: people hired, accounts contacted, content published, experiments run. Convert those into outcomes through explicit rates. The alternative, taking a market size and claiming a percentage of it, is the single fastest way to signal that you have not thought about how customers actually arrive.
Make the drivers visible and editable. Assumptions belong in labelled cells that feed formulas, never buried inside a calculation. An investor should be able to ask what happens if conversion halves and watch you change one number.
Keep it small. A model with a few dozen meaningful rows that you fully understand beats an elaborate one you inherited. Complexity is not rigour, and at pre-seed nobody is auditing your depreciation schedule.
Model the costs honestly, especially people. Headcount usually dominates early costs, so the hiring plan is most of the model. If revenue grows steeply while headcount stays flat, explain the mechanism or fix the assumption.
Include a downside case. Not a formality. A version where things take twice as long, which is the version that most often happens, showing what you would do about it. Founders who present only the good case are read as either inexperienced or unwilling to look at the risk.
Tie it to the raise explicitly. State what the money buys, what it proves, and what condition the company is in when it runs out. That connection is the part investors most want and most often do not receive.
Templates are a starting point, not an answer
A downloaded financial model gives you structure and a set of assumptions belonging to a different business. The value is in the layout, and the risk is arriving with drivers you never chose and cannot defend. Rebuild the driver section from your own understanding of how customers arrive, even if you keep the template's arithmetic underneath.
The mistakes that end the conversation
Top down market sizing. Claiming a small percentage of a large market with no account of how you reach anyone. It is the most common error and it is treated as a reliable signal of inexperience.
A curve with no mechanism. Revenue that bends sharply upward in year three with nothing in the model producing the bend. If growth accelerates, something must cause it, and that something has to appear as a driver.
Inconsistency with everything else you said. Numbers that contradict the deck, the team plan, or the raise amount. Investors check this, and it is checked more often than founders expect.
A model you cannot navigate. Being unable to find an assumption or explain a formula in a live conversation ends the credibility of the whole document, particularly if it suggests someone else built it.
Precision that is not real. Forecasting a specific figure for the sixtieth month implies a confidence nobody has. Rounding and stating ranges is more credible than false exactness.
Ignoring the shape of the sale. A model assuming immediate revenue from customers who in reality take months to buy, or usage that ramps slowly, will be wrong in a way an experienced investor spots immediately.
The workable summary: treat the model as an argument about how the business works, expressed in numbers, and expect to be questioned on the argument rather than on the total.
This is general information rather than legal or financial advice.
Frequently asked questions
- Why do investors ask for five year projections when everyone knows they are wrong?
- Because the model is read as a thinking artefact rather than a forecast. It shows which variables a founder believes drive the business, whether the opportunity is large enough under their own assumptions, and whether the raise amount is consistent with the burn the plan implies.
- How should a pre-seed financial model be built?
- Bottom up from activities you control, converted into outcomes through explicit rates you can defend. Keep assumptions in labelled cells that feed formulas, keep the model small enough that you understand every row, model headcount honestly since it dominates early costs, and include a downside case.
- What is the most common mistake in startup projections?
- Top down market sizing: claiming a small percentage of a large market with no account of how customers are actually reached. It is treated as a reliable signal of inexperience. The related error is a revenue curve that bends upward with nothing in the model producing the bend.
- Can I use a financial model template?
- As a starting point for structure. The risk is arriving with drivers you never chose and cannot defend in conversation. Rebuild the driver section from your own understanding of how customers arrive, keeping the template's arithmetic underneath if it is sound, since you must be able to navigate it live.