What Is a Financial Model and Why Does Your Startup Need One Before You Raise?
The investor asked the founder to send the financial model.
The founder opened Excel, pasted in last year's actuals, added 30% to each monthly revenue row for next year, and called it done. Investor response: fourteen questions about assumptions.
Where does the 30% growth come from? What's driving it, new customers, price increases, or expansion from existing accounts? If new customers, what's the sales capacity to support that? How does the cost structure scale with revenue?
The founder didn't have answers. Not because the business was bad, but because a row of revenue projections isn't a financial model.
What a Financial Model Actually Is
A financial model is a structured set of assumptions about how your business works, translated into a financial projection showing what those assumptions imply for profit, cash, and the balance sheet.
The assumptions are the model. The numbers are the output of the assumptions. When an investor stress-tests a model, they're changing the assumptions to see what happens, not just reading the bottom-line number.
A proper model has three interconnected statements:
The P&L: shows revenue, costs, and profitability month by month over a 3-year horizon.
The Cash Flow Statement: translates P&L profit into actual cash movement, accounting for the timing differences between when revenue is recognised and when cash is collected, and between when costs are incurred and when they're paid.
The Balance Sheet: shows what the company owns (assets), what it owes (liabilities), and what belongs to shareholders (equity) at each point in time. Often skipped by early-stage founders, but checked closely in due diligence.
What Goes Into the Model
Revenue model
This is where most founders stop too early. Revenue projections without underlying drivers are just guesses. The revenue model should explain specifically what generates revenue.
For a SaaS business: number of leads entering the pipeline × conversion rate × average contract value. Then, for existing customers: starting MRR + expansion MRR - churned MRR = ending MRR each month.
For a D2C brand: website sessions × conversion rate × average order value, with separate assumptions for paid traffic versus organic, and a repeat purchase rate assumption for existing customers.
For a B2B services firm: number of active engagements × average engagement value, with hiring plan tied to delivery capacity.
The revenue model should show what has to be true for the projections to be achieved.
Cost of goods sold
COGS should be modelled as a function of revenue, not as a fixed number. If COGS is 35% of revenue now, model what happens to that ratio as volume increases. Gross margin expansion (or compression) should be justified by specific assumptions: supplier negotiations, infrastructure cost efficiencies, or product mix shift.
Headcount plan
For most startups, people are the largest cost. The headcount plan should list every planned hire by role, start month, and salary band, with the employer PF (12% of basic) and gratuity provision (4.81% of basic) included in the cost.
Critically, headcount additions should be tied to something specific: a revenue milestone, a product release, a customer commitment, not just "we need more people to grow."
Operating expenses
Break OpEx by function: Sales & Marketing, Engineering/Product, General & Administrative. Within each, distinguish fixed costs (salaries, rent) from variable costs (marketing spend, commissions) so you can model what happens if you scale up or down.
Capital expenditure and working capital
Early-stage software businesses often overlook these. For product companies, inventory financing and receivables cycles can consume significant cash even in a profitable business. For companies with hardware components, capex needs to be modelled and depreciated.
Funding plan
The model should show when you'll run out of money at current burn rate, what the next raise will be, and how the runway changes under different scenarios.
Three Scenarios, Not One
A single-scenario model is a prediction. A three-scenario model is a planning tool.
Conservative scenario: key growth assumptions miss by 20-30%. What does burn look like? How long does the runway extend before the next raise?
Base scenario: things go roughly as planned.
Aggressive scenario: a key bet pays off early. What does that enable? Does the unit economics hold?
Investors don't expect the base case to be right. They want to see that you've thought about what happens when it isn't.
What Investors Look For in the Model
They are checking whether the assumptions are internally consistent. If revenue grows 3x but headcount grows 20%, they'll ask how the delivery capacity supports the growth. If gross margin improves dramatically at scale, they'll ask what drives it.
They're also checking whether you understand your own business. The founder who can walk through the revenue model driver by driver, explain each cost assumption, and articulate what would have to change to hit the aggressive case, that founder is fundable. The founder who says "we assume 30% month-on-month growth" and can't say why, isn't.
Common investor probes in this section: What is the revenue recognition policy (especially for annual contracts with upfront payment)? How does CAC trend as you grow? What happens to burn if revenue is 25% below plan? When do you hit cash flow breakeven, and what does that require?
When to Build the Model
For pre-seed founders raising the first round: a basic model showing use of funds, monthly burn, and runway is sufficient. Revenue projections are guesses, and sophisticated investors know it.
For seed-stage founders with 6-12 months of revenue history: the model should be grounded in actual unit economics. Real CAC, real gross margins, real churn. Projections built from those drivers.
For Series A: a full 3-statement model, 3-year horizon, monthly granularity, three scenarios, and a documented assumptions tab. This is non-negotiable.
Common Mistakes
Revenue projections that are just a number, with no driver explanation. "We'll grow 30% MoM" is not a model.
Not connecting headcount to revenue delivery. If the model shows revenue tripling but the team only grows 20%, investors will ask how.
Ignoring working capital in the cash flow. A profitable business can run out of cash if receivables extend while payables shorten.
Building the model in a format only the founder understands. The model gets shared in due diligence. Anyone should be able to navigate it.
Using annual numbers when monthly granularity is needed. Month-by-month cash flow is how you identify when you'll run out of money.
Key Takeaways
- A financial model is a set of assumptions, not a spreadsheet of projections. Investors evaluate the assumptions.
- A complete model has three statements: P&L, Cash Flow, and Balance Sheet, all connected.
- Revenue projections must be grounded in drivers: leads, conversion rates, churn, expansion, not just a percentage growth assumption.
- Build three scenarios. Conservative, base, and aggressive. Investors will stress-test your model; you should do it first.
- For Series A, a 3-statement model with 3-year monthly projections and a documented assumptions tab is expected.