Financial models scare founders who don't have a finance background. They shouldn't. Most SaaS models are 6 tabs of math you already understand intuitively, formalized into spreadsheet form. The trick is knowing which numbers matter and where assumptions hide.
The 6 tabs every SaaS model has
- Assumptions. The levers — pricing, growth rates, churn, hiring plan, fixed costs. Everything else flows from this. The most important tab.
- Revenue. MRR build. New customers × ACV - churn - contraction + expansion = net new MRR. Aggregates to ARR.
- Costs. Headcount-driven (with payroll loading) + variable costs (hosting, tools, contractors).
- P&L. Revenue - COGS = gross profit. - OpEx = operating profit. Monthly, rolling up to quarterly and annual.
- Cash. Opening cash + revenue collected - cash out = ending cash. Runway falls out of this.
- Balance sheet. Simplified for early-stage. Mostly cash, deferred revenue, and equity. More important at later stages.
Where assumptions actually live
Models look complex but the core assumptions are 8-10 numbers:
- Average contract value (ACV). What a typical customer pays per year.
- Sales cycle length. Days from first touch to close.
- Win rate. Percentage of qualified opportunities that convert.
- Net new customers per month. Output of pipeline × win rate.
- Monthly logo churn rate. Customers leaving each month.
- Net revenue retention (NRR). Existing customer revenue change including expansion.
- Gross margin. What's left after delivering the service.
- Headcount plan. When you hire, in what role, at what salary.
Find these on the Assumptions tab. Change one and the rest of the model recalculates.
The numbers investors actually probe
In rough order of how often they come up:
- Net new MRR by month for the last 12 months. Is growth accelerating or decelerating?
- Cohort retention curves. Logo and dollar retention by signup month. The single best signal of SaaS health.
- CAC payback period. Months to recover acquisition cost. Below 18 months = healthy.
- Burn multiple. Net burn ÷ net new ARR. Below 1 = excellent; above 2 = under scrutiny.
- Magic number. (Quarterly net new ARR × 4) ÷ quarterly S&M spend. Above 1 = healthy.
- Rule of 40. Growth rate + profit margin should be ≥ 40%. Mostly applies at scale ($10M+ ARR).
Red flags in your own model
Self-audit before showing to investors:
- "Hockey stick" growth without underlying drivers. If MRR triples in year 2 with no change in headcount, marketing spend, or sales motion, the model is fiction.
- Churn assumed to drop without intervention. Models often show churn falling from 6% to 2% with no plan for how it happens. Investors notice.
- Hiring plan that doesn't match growth. Tripling revenue with the same team = magical thinking.
- Gross margin expanding without explanation. Margins improve from 60% to 80% over 3 years — why? Specific driver?
- CAC payback period under 6 months at scale. Possible early on, rare at scale. Investors will assume aggressive assumptions.
How to debug a model that's wrong
When something looks off, work backward:
- Find the driver. Click on the formula and trace back to which assumption feeds it.
- Sanity-check the assumption. "Is 25% MoM growth realistic given our actuals?"
- Adjust and propagate. Change the assumption and re-read the downstream cells.
- Recheck the cash row. Almost everything ends up in the cash row. If runway changes dramatically with a small assumption tweak, you've found a sensitive lever.
Common model-reading mistakes
- Trusting the bottom line without auditing assumptions. A model is only as good as its inputs. Always start with the Assumptions tab.
- Ignoring the cash row. Profit and cash are different. A profitable company can run out of cash if working capital is bad.
- Single scenario thinking. Always model 3 scenarios. Investors will ask.
- Static models. Update monthly with actuals. A model 6 months out of date is worse than no model.
- Building from scratch. Start with a template (ours), customize, ship. Building structure from zero wastes weeks.
Pair this guide with our financial model templates — start with a template and you'll have a working model in a weekend.
FAQ
Do I need a financial model before raising?+
Yes by seed. Investors expect a 3-year monthly model showing revenue, expenses, headcount, and cash. They'll often build their own version, but having yours signals you understand the business.
Should I hire a fractional CFO to build the model?+
Helpful but not necessary at seed. A good template (like ours) plus a clear vision is enough. Hire a fractional CFO ($3-8K/month) when you're approaching Series A or have complex unit economics.
What's the difference between a financial model and a budget?+
A budget is what you plan to spend; a financial model projects revenue, costs, and cash from first principles. The model contains the budget but extends to growth, unit economics, and fundraising scenarios.
How conservative should my projections be?+
Investors discount projections 30-50%. Better to be conservative on revenue and aggressive on costs — you'll beat conservative projections, which builds trust. Aggressive projections you miss become trust problems.
Should I model multiple scenarios?+
Yes — base, upside, downside. Base is your honest plan; upside is best case (+30% revenue); downside is what happens if revenue is 30% lower and costs are 20% higher. Investors care about the downside more than the base.