How to build a financial model if you've never built one before
Most founders avoid building a financial model for the same reason they avoid going to the dentist — they know they should, they're not sure what it will reveal, and they're pretty sure it's going to be uncomfortable. So they put it off until something forces the issue: an investor asks for projections, a bank wants a forecast, or the business hits a cash wall nobody saw coming.
Here's the thing: a financial model isn't a spreadsheet you build for other people. It's a decision-making tool you build for yourself. And you don't need to be a finance person to build one that actually works.
What a financial model actually is (and isn't)
A financial model is not a prediction. It's a structured way of thinking about how your business makes and spends money — and what happens to both when conditions change.
At its core, a model answers three questions:
What does revenue look like, and what drives it?
What does it cost to generate that revenue?
What does cash look like at the end of each month?
That's it. Everything else — the multiple tabs, the color-coded inputs, the waterfall charts — is in service of those three questions. If your model can answer them clearly, it's doing its job.
What a model isn't: a guarantee. The numbers will be wrong. That's not a bug — it's a feature. The value of modeling isn't the output. It's the discipline of thinking through your assumptions explicitly, so you know why you were wrong and what to adjust.
Start with revenue — and work backwards from the unit
The most common mistake founders make when building their first model is starting with a target number. "We want to hit $3M next year" — and then they work backwards to justify it. That's a budget wish list, not a model.
Start instead with the unit of revenue in your business. What's the smallest repeatable transaction?
For a service business: one client engagement or one monthly retainer.
For an e-commerce brand: one order.
For an agency: one active client account.
Once you know the unit, the model becomes: how many units do I have right now, how many do I expect to add per month, how many do I expect to lose, and what does each one generate?
A simple example: you run a marketing agency. You have 12 retainer clients at an average of $6,500/month. You close 2 new clients per month and lose 1. After 6 months, you have 18 clients. Revenue at month 6: $117,000. That took 10 minutes to model — no finance background required.
Revenue drivers worth capturing for most founder-led businesses:
New client/customer acquisition (monthly adds)
Churn or attrition rate
Average contract value or order value
Upsell or expansion revenue (if applicable)
Model your costs in two buckets: fixed and variable
Once you have revenue, costs come next. The most useful distinction in a first model isn't operating vs. non-operating — it's fixed vs. variable.
Fixed costs stay roughly the same regardless of how much you sell. Rent, software subscriptions, base salaries, insurance. These are your floor — the minimum burn even in a flat month.
Variable costs scale with revenue. Cost of goods sold, contractor hours tied to delivery, commission, shipping. These tell you your gross margin — the percentage of each dollar of revenue left after paying to deliver it.
The reason this matters: if your gross margin is 40%, every dollar of new revenue brings 40 cents toward covering your fixed costs and generating profit. If it's 20%, you need twice as much revenue to cover the same fixed base. Knowing your gross margin tells you how hard your business has to run just to break even.
Build the cost side as simply as possible at first. List your fixed monthly expenses line by line. Then calculate variable costs as a percentage of revenue using your actual historical data. Refine from there.
Add a cash flow layer — this is where it gets real
Revenue and cost projections give you a picture of profitability. But profitability and cash are not the same thing, and for most founder-led businesses, cash is the constraint that actually matters.
The cash flow layer of your model captures timing. When does revenue actually hit your bank account? When do you actually have to pay your vendors and team?
For service businesses with net-30 or net-60 payment terms, you might be profitable on paper and still scramble to make payroll. For e-commerce businesses with inventory, you're spending cash 60–90 days before you collect it. None of that shows up in a P&L.
The simplest way to build a cash flow layer for a first model:
Start with your bank balance at the beginning of each month
Add cash collected (not revenue recognized — cash collected)
Subtract cash paid out (not expenses incurred — cash paid)
End balance = beginning balance + collected – paid
Track this monthly for 12 months forward. If the ending balance goes negative in month 7, you now have a specific problem to solve — not a vague sense that things might get tight.
Scenario planning isn't optional
Once your base model is built, run two additional scenarios before you call it done.
A downside scenario: what does the model look like if revenue comes in 20% below your base case? Where does cash go negative? How many months of runway do you have?
An upside scenario: what does the model look like if you close twice as many new clients next quarter? What new costs does that create, and when? Do you need to hire before the revenue hits?
These aren't exercises in pessimism or optimism. They're how you identify which assumptions your business is most sensitive to — and which levers actually move the needle.
Most first-time modelers build one set of numbers and treat them as gospel. The founders who use financial models well treat the base case as one possible future and manage the gap between what they projected and what actually happened.
Vera's Take
The founders who resist building models usually tell me they're not spreadsheet people. That's not the real issue. The real issue is that a model forces you to make your assumptions explicit — and explicit assumptions can be wrong in ways that vague optimism can't. Build the simplest version that answers the three core questions: what drives revenue, what does it cost, and what does cash look like month by month. You can add complexity later. What you can't add later is the time you spent making decisions without it.
If you want this applied to your business, that's what Vera CFO is for.