How to Build a Financial Model for Your Startup (Without Overcomplicating It)

Why most startup financial models are wrong

Let's start with an uncomfortable truth: every early-stage financial model is wrong.

Your revenue projections for Year 3? Wrong. Your hiring plan for Q4? Wrong. Your customer acquisition cost in 18 months? Also wrong.

And that's completely fine.

The point of a financial model is not to predict the future. It's to show that you understand your business deeply enough to think clearly about where it's headed.

When I sat in on investment decisions at Startmate and Blackbird, nobody believed a founder's Year 3 revenue projection. What we looked for was whether the founder understood the fixed vs variable factors in their business. Could they explain which assumptions were temporary? Could they articulate what would change as they scaled?

The founders who impressed us weren't the ones with the most detailed models. They were the ones who could talk through every number as if it was tattooed on their forearm.

The numbers you need to know cold

I've coached hundreds of founders, and I always come back to the same set of numbers. If you know these, you can answer almost any investor question:

1. Revenue (and who's paying you)

Not just the total number - your biggest clients and where your revenue comes from. At pre-seed, you might only have five paying customers. Know all five by name. Know what they pay. Know when they signed up and why.

You should literally know your revenue on a daily basis. Not because the exact number matters that much at this stage, but because if you don't track it, people assume you don't care about the business. And they're probably right.

2. Burn rate

How much cash leaves your account every month. This includes everything - salaries, software, rent, that marketing tool you signed up for and forgot about.

Break it into two buckets: - Fixed costs: Rent, salaries, insurance - things that don't change with revenue - Variable costs: Marketing spend, cloud hosting, transaction fees - things that scale with activity

3. Runway

The simplest and most important calculation in startup finance:

Cash in the bank / monthly burn rate = months of runway

If you have $200K and burn $20K per month, you have 10 months. That's how long you have to either reach profitability, raise more money, or run out of cash.

Every founder should know their runway to the month. Not approximately. Exactly.

4. Unit economics (roughly)

At pre-seed, you don't need a precise customer acquisition cost (CAC) or lifetime value (LTV). But you should be able to talk about them directionally.

"We spend about $50 on ads to get a customer, and each customer pays us $30 per month and stays for about 6 months. So our LTV is roughly $180 against a $50 CAC."

That's it. No need for decimal places. Just show you understand the relationship between what it costs to get a customer and what that customer is worth.

The one-page financial model

Here's what a good early-stage financial model actually looks like. It fits on one page of a Google Sheet.

Row 1: Revenue Monthly revenue, broken out by product line or customer segment if you have more than one. Include the number of customers and the average revenue per customer.

Row 2: Cost of goods sold (COGS) What does it cost to deliver your product? Hosting, third-party APIs, payment processing fees. This gives you your gross margin.

Row 3: Operating expenses Salaries, rent, software, marketing. The biggest line items that make up your burn.

Row 4: Net burn Revenue minus all expenses. This is positive (you're making money) or negative (you're burning cash).

Row 5: Cash balance How much is in the bank at the end of each month. This is the row that tells you when you run out of money.

Row 6: Runway Cash balance divided by net burn. How many months you have left.

That's six rows. Twelve columns for twelve months. One Google Sheet tab.

Is this a complete financial model? No. Is it enough for a pre-seed or seed stage founder? Absolutely. You can add complexity later when you have the data to justify it.

When financial models actually change investment decisions

Here's an honest answer from being on the investor side: at pre-seed, the financial model rarely changes the decision. The decision is made on the team, the problem, and the early traction.

But at Series B and beyond, the model matters enormously.

At that stage, you have real data. Real cohorts. Real unit economics. And the financial model becomes a confirmation tool - it either confirms the investor's thesis or raises red flags.

I've seen models change opinions at the later stage, or more accurately, confirm opinions. When a partner at a VC fund is trying to convince their colleagues to vote yes on a deal, a clean financial model with strong unit economics becomes their best weapon. "Look at the payback period. Look at the retention curves. Look at how the margins improve at scale."

At pre-seed? The model shows you can think clearly. At Series B? The model shows the business actually works.

So don't stress about building a perfect model before your pre-seed raise. But absolutely build one. Because the discipline of putting numbers to your assumptions is how you find the holes in your thinking.

The assumptions page (this is what investors actually read)

Here's the secret most founders miss: investors don't care about the numbers in your model. They care about the assumptions behind the numbers.

Build a separate tab (or section) called "Assumptions." For every number in your model, write down why you chose it.

• Revenue growth rate: 15% month-over-month. Why? "Based on our current growth trajectory - we grew 12% last month and 18% the month before. We're averaging 15%."
• CAC: $50. Why? "We've spent $2,000 on ads and acquired 40 customers. That's $50 per customer."
• Churn: 5% monthly. Why? "We've had 3 customers cancel out of 60 total over the past 3 months."

If you can't explain an assumption, you shouldn't have it in your model. The CEO needs to be across every single assumption and be able to talk through it. This isn't something you outsource to your accountant or your cofounder who "does the numbers."

When a founder can rattle off their assumptions without looking at a screen, that tells me they live inside their business. When they have to flip through tabs to find an answer, that tells me the model was built for show, not for decision-making.

The three-scenario trick

One powerful technique: build three versions of your model.

Bear case: Everything goes wrong. Growth slows, churn increases, that big client leaves. How long is your runway?

Base case: Things go roughly as expected. This is your main model.

Bull case: Things go better than expected. That partnership closes, viral growth kicks in, retention improves.

This does two things. First, it shows investors you think in ranges, not certainties. Nobody believes a single projection - but three scenarios that bracket reality? That shows maturity.

Second, it helps you make better decisions. If your bear case shows you running out of money in four months, you probably need to raise sooner or cut costs now - regardless of what your base case says.

The bear case is the one you should pay the most attention to. As the saying goes, pessimists sound smart but optimists make money. Be optimistic about your business, but pessimistic about your assumptions.

Tools and templates (keep it ridiculously simple)

I'm going to give you the most boring advice possible: use Google Sheets.

Not Notion. Not a fancy financial modelling tool. Not an AI-powered forecasting platform. Google Sheets.

Why? Because: 1. Everyone can read it. Investors, cofounders, advisors - everyone knows how to open a Google Sheet. 2. You can share it instantly. Send a link, set permissions, done. 3. It forces simplicity. You're not going to build a 50-tab monstrosity in Google Sheets because it would be painful to navigate. 4. It's free. You're a startup. Save your money for things that matter.

You can also use Excel - Claude and other AI tools now integrate with spreadsheets directly, so you can literally describe what you want and have AI build the model structure for you. Then you fill in your actual numbers and assumptions.

But whatever tool you use, the principle is the same: if your financial model takes more than five minutes to explain, it's too complicated. Simplify until a smart person with no context could understand it in one pass.

Common financial model mistakes

After reviewing hundreds of models, these are the patterns I see over and over:

Mistake 1: Hockey stick revenue with no explanation. Every startup model shows flat revenue for 6 months then an explosion. Why? What changes in month 7 that causes 10x growth? If you can't answer that specifically, your model isn't a model - it's a wish.

Mistake 2: Forgetting to model hiring. "We'll hire 5 engineers in Q3." Great - that's $75K each including on-costs, so your burn rate jumps by $375K per quarter. Did you account for that? What about the 2-3 month ramp time before they're productive?

Mistake 3: Ignoring seasonality. Many businesses have seasonal patterns. If you're selling to schools, summer is dead. If you're in retail, December is massive. Your model should reflect reality, not a smooth average.

Mistake 4: Building it once and never updating. Your financial model is a living document. Update it monthly with actual numbers. Compare actuals to projections. This is how you get better at forecasting - and how you spot problems early.

Mistake 5: Having someone else build it for you. The worst thing you can do is hand your financial model to an accountant or consultant. They'll build something technically perfect that you can't explain. Build it yourself, even if it's messy. The process of building it is where the learning happens.

The bottom line

Your financial model is not a prediction. It's a thinking tool.

At pre-seed, it shows investors you understand your business. At Series A, it shows your growth trajectory. At Series B and beyond, it proves the unit economics work.

But at every stage, the value isn't in the spreadsheet - it's in the conversation you can have about the numbers. The founder who knows their revenue by heart, who can explain every assumption, who has modelled the bear case and knows exactly how long their runway is - that's the founder investors trust.

Keep it simple. One page. Six rows. Twelve months. Update it every month. Know every number by heart.

And please, for the love of everything - don't hire a Big Four accountant to build it for you. Build it yourself in a Google Sheet. It'll take you an afternoon, and the act of building it will teach you more about your business than any consultant ever could.