How to Build a Financial Model for Your Startup
Most first-time founders avoid financial modeling for as long as possible. It feels like something you do when you're bigger, more serious, when you actually have revenue to model. That's exactly backwards.
A financial model isn't a record of what happened. It's a map of what you think will happen, and why. Build it early, and you'll make better decisions about pricing, hiring, and how long your money lasts. Skip it, and you'll hit a wall you could have seen coming six months earlier.
Here's how to build one that's actually useful, without an MBA or a finance background.
What Is a Startup Financial Model, Really?
A financial model is a set of connected spreadsheets (or a structured planning tool) that projects your revenue, costs, and cash position over time. Most startup models cover 12 to 36 months.
The key word is "connected." Your revenue assumptions feed into your headcount plan. Your headcount plan feeds into your burn rate. Your burn rate tells you how long your money lasts. Change one number, and everything else updates. That's what makes a model useful.
A good startup financial model includes five core components:
- Revenue projections (what you expect to earn and when)
- Cost structure (fixed vs. variable costs)
- Headcount plan (who you're hiring and what they cost)
- Cash flow (month-by-month cash in vs. cash out)
- Key metrics (burn rate, runway, MRR, CAC, LTV)
You don't need all of these on day one. But you do need to understand what connects to what.
Why Your Assumptions Matter More Than Your Numbers
Here's the thing most people get wrong: the numbers in your model are not the point. The assumptions behind the numbers are.
Any investor who looks at your model knows the projections won't be exactly right. What they're evaluating is your thinking. Do you understand your unit economics? Do you know what drives your growth? Are your assumptions grounded in reality, or did you just multiply "1% of a $10 billion market"?
Common assumption mistakes first-time founders make:
- Assuming linear growth (real growth is lumpy)
- Underestimating CAC (customer acquisition cost) by 2-3x
- Forgetting one-time costs like setup fees, legal, or tooling
- Not modeling a pessimistic scenario at all
- Building month 1 projections that somehow already hit $50k MRR
Be specific. If you're modeling 10 new customers in month 3, write down how you're getting those 10 customers. If you can't answer that, your model is fiction.
How to Project Revenue for a Pre-Revenue Startup
This is where most people freeze. You have no data. You don't know what your conversion rate will be. You don't know your churn. So how do you project revenue?
You build from first principles.
Step 1: Define your acquisition channel. Start with one. Maybe it's content. Maybe it's cold outbound. Maybe it's a community you're active in. How many people will realistically see your product per month?
Step 2: Apply realistic conversion rates. Industry benchmarks help here. SaaS free-to-paid conversion is typically 2-5%. Cold email reply rates are 1-3%. Inbound conversion rates from well-optimized landing pages range from 2-8%. Pick a conservative number and defend it.
Step 3: Multiply by price. Take your conversions times your monthly or annual price. That's your new MRR for the month.
Step 4: Add churn. For SaaS, monthly churn for early-stage startups is often 3-8%. Model it in from month one. Founders who don't account for churn end up with hockey stick projections that fall apart under scrutiny.
Step 5: Layer in growth. As you invest in marketing and word-of-mouth builds, your acquisition volume will increase. Model this conservatively in year one (maybe 10-15% month-over-month growth in leads) and more aggressively in years two and three.
This won't be perfect. It doesn't need to be. It needs to be a testable hypothesis.
How to Calculate Burn Rate and Runway
Burn rate is how much cash you're spending each month, net of any revenue. Runway is how many months you have left at that burn rate.
Monthly burn = total monthly expenses minus monthly revenue
If you're spending $15,000 per month and making $3,000 in MRR, your net burn is $12,000.
Runway (in months) = cash in bank divided by net monthly burn
With $120,000 in the bank at $12,000 monthly burn, you have 10 months of runway. That's it.
Why does this matter? Because 10 months sounds like a lot until you realize fundraising takes 3-6 months minimum, customer development eats another 2, and suddenly you're making panicked decisions with 2 months left.
The rule of thumb most experienced founders use: always know your runway, and start your next fundraise or revenue push when you have at least 6 months left. Not 3. Not 2. Six.
What to Include in Your Cost Model
Costs fall into two categories: fixed and variable.
Fixed costs don't change much month to month. Rent, salaries, SaaS subscriptions, your own salary (yes, pay yourself something, even if it's small). These are predictable.
Variable costs scale with activity. Payment processing fees, hosting costs (if usage-based), contractor work for specific campaigns, ads spend. These require assumptions too, usually expressed as a percentage of revenue or a cost-per-unit.
For most early-stage SaaS startups, your biggest costs will be:
- Salaries (50-70% of total costs if you have a team)
- Marketing and advertising
- Software and infrastructure
- Legal and compliance (one-time spikes, usually)
- Contractor and freelance work
One thing first-time founders consistently forget: employer taxes and benefits. If you're paying someone $80,000 in salary, your actual cost is closer to $95,000-$105,000 once you add payroll taxes, health insurance, and other benefits. Model this correctly, or your burn rate will constantly surprise you.
Three Scenarios Every Startup Model Should Include
A single projection is not a financial model. It's a wish list.
Build three scenarios: base, bull, and bear.
Base case: Your most likely outcome if things go reasonably well. Not optimistic, not pessimistic. Grounded.
Bull case: What happens if your paid conversion rate is 20% higher than expected, churn is lower, and a big customer or channel opens up? This is your upside.
Bear case: What if customer acquisition takes twice as long? What if your churn is higher than expected? What if a key hire doesn't work out? This is your survival scenario.
The bear case is the most important one. It tells you the minimum conditions for survival and gives you an early warning system. If your bear case model runs out of cash in month 8, you need to either cut costs or change your fundraising timeline now, not in month 7.
Tools like Foundra, or a well-structured spreadsheet with three scenario tabs, can help you run these projections side by side so you can see the differences clearly.
The Key Metrics Worth Tracking in Your Model
Financial models for startups live and die by a handful of metrics. Know these cold.
Monthly Recurring Revenue (MRR): Total predictable revenue each month from active subscriptions. Track new MRR, expansion MRR (upgrades), contraction MRR (downgrades), and churned MRR separately. The breakdown tells you a lot more than the total.
Customer Acquisition Cost (CAC): Total sales and marketing spend divided by new customers acquired in the same period. If you spent $5,000 on marketing in March and got 20 new customers, your CAC is $250.
Lifetime Value (LTV): Average revenue per customer divided by monthly churn rate. If your average customer pays $50/month and you churn 5% per month, LTV is $1,000.
LTV:CAC ratio: This is the health check number. You want LTV to be at least 3x your CAC. Below 3x, you're probably losing money on every customer you acquire at scale. Above 3x, you have room to grow.
Payback period: How many months does it take to recover your CAC? Under 12 months is generally healthy for B2B SaaS. Over 18 months is a red flag for most early-stage companies.
You don't need all of these from day one. But build your model so they're easy to calculate.
Common Mistakes to Avoid
A few things that trip up nearly every first-time founder building their first model.
Top-down instead of bottom-up. "The market is $2 billion so if we get 1%, that's $20 million." This tells you nothing. Build from the customer up: how many customers can you realistically reach, convert, and retain?
Not updating the model. A model you build once and never look at is decoration. Update it monthly with actual numbers. Compare actuals to projections. When you miss, figure out why.
Hiding bad assumptions. Some founders build optimistic models and refuse to stress-test them. If your model only works if everything goes right, it's not a plan. It's a fantasy.
Confusing revenue and cash. Revenue is when you earn it. Cash is when you receive it. Annual contracts paid upfront are great for cash but can look misleading in a revenue model. Monthly invoices can create timing gaps. Track both.
Key Takeaways
- A financial model is a set of connected assumptions, not a prediction. The value is in the thinking, not the precision.
- Build revenue projections from the bottom up: acquisition channel, conversion rate, price, churn.
- Know your burn rate and runway at all times. Start your next fundraise or revenue sprint with at least 6 months left.
- Build three scenarios. The bear case is the most useful.
- Update your model monthly with actual numbers. A static model is useless.
- Track LTV:CAC ratio from the start. It tells you whether your business model actually works at scale.
FAQ
Do I need a financial model before I have any revenue?
Yes. A pre-revenue financial model is really a set of hypotheses about how your business will work. Building it forces you to make your assumptions explicit, which makes them testable. Investors also expect to see one, even at the idea stage.
How far out should a startup financial model project?
Most early-stage startups model 18-24 months in detail. Beyond that, projections become less reliable. Three-year models are common for investor presentations, but the third year should be treated as directional rather than precise.
What tools should I use to build a startup financial model?
Google Sheets or Excel work fine for most early-stage companies. If you want a structured framework with built-in formulas for unit economics and projections, tools like Foundra walk you through the key sections, or you can find templates from Visible.vc, Causal, or Baremetrics. The tool matters less than the rigor of your assumptions.
What's the difference between a financial model and a business plan?
A business plan describes what your business does and how you'll operate. A financial model quantifies it. Most investors care more about the financial model, because it shows you understand your unit economics. The best approach: a short written plan plus a solid financial model.
How accurate does my financial model need to be?
Not very, in the sense that nobody expects your projections to be exactly right. What matters is that your assumptions are reasonable, your logic is consistent, and you can explain every number. If an investor asks "why do you assume 4% monthly churn?" you should have a real answer.
When should I update my financial model?
Monthly, at minimum. Update actuals, compare to projections, and note the variance. Every significant miss is information about what's working and what isn't. Treat it as a feedback loop, not just an admin task.
Top comments (0)