How to build a financial forecast investors will believe

Any SME or startup owner who has ever presented a forecast to a banker or a fund knows that polite silence, followed by a single question: “and why 30% growth?” At that moment the forecast stops being a spreadsheet and becomes a credibility test. The investor is not trying to find out whether you will hit the exact number, they already know you will not. They are trying to find out whether you understand your own business well enough to explain every line. A forecast people believe is not an optimistic forecast, it is a defensible one. Here is how to build one.

Why most forecasts are not believed

The typical SME or startup forecast fails for the same reasons every time. It starts from the desired result and works back to the assumptions, instead of the other way around. It shows a revenue curve that rises smoothly, as if growth were a rate rather than a sequence of decisions. And it presents a single scenario, which tells the investor: “I have not thought about what could go wrong.” An experienced reader spots these three flaws in under two minutes. Often they have stopped reading you before you finish presenting the first slide.

• The hockey stick: revenue flat yesterday, exponential tomorrow, with no visible change in resources. Nobody believes it.
• Backward reasoning: you decided you needed €5 million of revenue, then filled in the cells to get there.
• Costs that do not follow revenue: sales double but headcount and variable costs stay frozen. Mathematically impossible.
• A single scenario, with no margin of error and no breaking point identified.

The principle: start from the drivers, not the final number

A credible forecast is built bottom up, from the few variables that genuinely steer your business. They are called the drivers. Revenue is not an assumption, it is an output: it follows from the number of customers, the average ticket, the retention rate, the sales cycle. Your job is to identify the three or four drivers that explain 80% of your results, make them explicit, and argue only about those.

Take a B2B services business. Rather than writing “revenue: €4 million,” you write the chain: number of salespeople, meetings per salesperson per month, win rate, average contract, renewal rate. If each link is plausible on its own, the total is plausible too. And crucially, when the investor challenges the number, you are not defending a hunch, you are defending a 20% win rate you have already observed over the last six months.

An investor does not believe a number. They believe the reasoning that produces it.

Linking the three statements: the test amateurs miss

This is where technical credibility is won. A serious forecast links three statements: the income statement (am I profitable?), the cash flow plan (do I have enough cash to last?) and the balance sheet (what does my financial structure look like?). Many present only the first. Yet an investor looks at the second first, because a business that is profitable on paper can still die for lack of cash. Consistency across the three is the number one signal of seriousness.

• Profit feeds cash, but with a lag: a sale signed in March is only collected in May if your customers pay on 60 day terms.
• Growth consumes cash through working capital: the faster you sell, the more customers and inventory you finance before being paid.
• Capex and loan repayments leave cash without passing through profit, which is why a positive income statement can coexist with a bank account that keeps draining.

If you were to do only one thing to make your forecast credible, it would be this: build a real cash flow plan that turns every sales assumption into a dated collection. Our 13 week cash flow guide walks through that mechanism week by week; it is exactly the logic to extend across 24 or 36 months for a forecast.

Three scenarios, and only one that truly matters

Presenting a single scenario is an admission of naivety. Present three: a base case (the most likely, the one you defend), a low case (growth does not come, the sales cycle lengthens) and a high case (everything aligns). But the investor only truly cares about the low case. Their real question is not “how much do you make if all goes well?” but “what happens if nothing goes to plan, and how long do you survive?” It is, incidentally, the first thing I look at when an owner hands me their model: I go straight to the low case.

The low case must therefore answer one precise question: what is my breaking point? For a startup, it is the month cash goes below zero in the cautious scenario. For an SME, it is the revenue level below which the bank covenants break. Showing that you know that point, and that you have a plan B to never reach it, is worth more than ten points of growth in the high case.

A worked example, read in one minute

A startup is raising. Starting cash: €800,000. Base case: revenue moving from €1 million to €2.5 million over 18 months, cash burn of €60,000 a month that winds down by month 14. Low case: growth plateaus at €1.6 million, burn stays at €60,000. The honest forecast then shows that in the low case cash hits zero at month 13. The defensible conclusion in front of the investor: “I am raising 18 months of runway in my base case, but only 13 in my low case; here are the two cost levers I pull if by month 9 I am not on the base trajectory.”

That level of precision changes everything. The owner who says this is not selling a dream, they are managing a risk. That is precisely what a banker or a fund will happily finance.

The defensible forecast checklist

• Every revenue line breaks down into explicit drivers (volume × price, or customers × ticket × retention), never a growth rate dropped in from above.
• Variable costs and headcount move with activity; no cost line stays frozen while sales double.
• The three statements are linked and consistent: the cash in the cash flow plan matches the cash line on the balance sheet.
• Customer and supplier payment terms are modeled, not ignored; working capital appears in black and white.
• Three scenarios, with the low case breaking point identified and an action plan attached.
• Every key assumption is tied to real data (history, open quotes, sector benchmark), not to a hunch.

Building this forecast takes method more than spreadsheet hours. If you are preparing a raise, a credit application or simply a 36 month plan for yourself, a one hour diagnostic conversation, the kind a fractional CFO runs, surfaces the assumptions that will not survive an investor and structures the model around your real drivers, before your banker finds them for you.

The useful test: which assumption changes the decision?

Keep a dated forecast before revising it. Next month, separate volume, pricing, contribution and collection timing differences. Updating the forecast without preserving its predecessor removes the comparison needed to learn. Each important assumption needs an owner, a reference and a trigger for review.

For hiring, test a slower commercial start and later collections. Use the downside case to set a launch condition rather than imply certainty.

See how to prepare this decision.

Frequently asked questions

Also worth reading