The 3-statement model — and which startups need all three
A full 3-statement model links the P&L, balance sheet, and cash flow statement so that changes in one flow through to the others. Most early-stage founders only build a P&L. Here is when you need each.
P&L (Income Statement)
Always. Revenue, gross margin, operating expenses, EBITDA. This is the minimum. Build it monthly for year 1, quarterly for years 2-3. Show assumptions separately — never bake growth rates into the numbers without surfacing them.
Cash Flow Statement
From seed stage onwards. The P&L shows profit; cash flow shows the money actually moving. They diverge when you have deferred revenue (SaaS annual contracts), long payment terms, or capex. Investors care more about cash runway than P&L profitability at early stage.
Balance Sheet
Required from Series A onwards. Pre-Series A: a simplified version showing cash position, deferred revenue, and outstanding liabilities is sufficient. Post-Series A: a full balance sheet is expected and will be checked against your management accounts.
How to model each revenue type
SaaS
Formula
MRR = (opening MRR + new MRR + expansion MRR) − churned MRR
Key assumptions
Monthly growth in new MRR, monthly churn rate, expansion rate from existing customers
Practical tip
Model at the cohort level, not just top-line. A flat MRR number that hides improving cohort retention is a better story than you're showing.
Marketplace
Formula
Revenue = GMV × take rate. GMV = supply side volume × average order value
Key assumptions
Supply growth rate, demand-side conversion, average order value, take rate evolution
Practical tip
Model supply and demand separately. Most marketplace models fail because they assume demand follows supply linearly — it doesn't.
Services
Formula
Revenue = billable headcount × utilisation rate × day rate
Key assumptions
Headcount hiring plan, time to ramp, utilisation target (typically 70-80%), day rate by seniority
Practical tip
Always model utilisation conservatively. 80% utilisation sounds achievable — 65% is what most services businesses actually run at in year one.
E-commerce / DTC
Formula
Revenue = orders × AOV. Orders = (new customers + returning customers) × frequency
Key assumptions
Customer acquisition cost, repeat purchase rate, average order value, return rate (deducted)
Practical tip
Separate new customer revenue from repeat customer revenue. They have entirely different margin and CAC profiles.
Common financial model mistakes
Hockey stick with no drivers
A revenue line that goes up 5x in year 3 with no corresponding change in headcount, marketing spend, or channel mix. Investors will ask "what drives the inflection?" and you must have an answer that is specific and testable.
No sensitivity analysis
A single-scenario model signals inexperience. Build base, bear, and bull cases. Show what happens if growth is 30% slower or churn is 2% higher. The model that survives stress-testing is more credible than the one that shows unbroken optimism.
Ignoring working capital
Common in services and e-commerce: you bill in arrears, pay suppliers upfront, and have 30-60 day payment terms. Your cash flow can go negative even when you're profitable. Model the working capital cycle explicitly.
Headcount not tied to revenue
If your revenue 3x's but your team only grows by one person, that is not a hiring plan — it is wishful thinking. Each department's headcount should follow a logical ratio to the activity it supports (sales headcount / pipeline, engineering headcount / product milestones).
Gross margin ignored
Top-line revenue in, EBITDA out, nothing in between. Gross margin is where investors look first. If you're a SaaS business at 40% gross margin, that needs explanation. The model should show why and when it improves.
The 3 scenarios rule — and which one to show investors
Every financial model should have three scenarios. Investors who see only one scenario know you modelled backwards from a target revenue number. The scenario work is as much about your own planning discipline as investor presentation.
Base case
What you actually expect to happen, built bottom-up from your current growth rate and hiring plan. This is what you present to investors. It should be achievable without perfect execution.
Bear case
Growth is 30-40% lower, key hire takes 3 months longer, churn is 2-3% higher. Your bear case should show you survive — you reach profitability later, but you don't run out of cash mid-round. If you do, investors will require covenants.
Bull case
Everything breaks your way: viral growth, enterprise contract closes early, take rate improves. Use this internally to understand your upside. Do not present this to investors as the base case — they will notice and discount everything you say afterwards.
DeckBrief
Generate standardised financial summary slides from your model outputs
Paste your P&L projections into DeckBrief and get the financial summary slide, KPI grid, and runway projection formatted for board reports and investor presentations.
Generate Your Financial Slides Free →Frequently asked questions
How far out should a startup financial model go?
Three years for most seed and Series A raises. Year 1 should be monthly (so investors can track actuals vs model after the round). Years 2 and 3 can be quarterly or annual. Going beyond 3 years is usually theatre — the assumptions become too uncertain to be meaningful, and investors know it. Five-year models are more common at Series B and growth stage, where unit economics are proven and the only question is scale.
Do investors expect profitability in the model?
Not for early-stage raises. What investors want to see is a path to profitability that is internally consistent — the business reaches gross margin positive, then EBITDA positive, at a point that aligns with the capital you're raising. A model that shows perpetual losses with no inflection point is a red flag. A model that shows profitability in year 1 for a VC-backed business signals you're not thinking big enough about the market opportunity.
What's the difference between revenue and ARR?
Revenue is what you invoice. ARR (Annual Recurring Revenue) is the annualised value of your current recurring contract base — it excludes one-time revenue and professional services. If you have £50K MRR in recurring SaaS subscriptions plus £20K this month in implementation fees, your MRR is £50K and your ARR is £600K (£50K × 12). Never include non-recurring revenue in your ARR — investors will spot it and it destroys credibility.
How do I model CAC and LTV?
CAC (Customer Acquisition Cost) = total sales and marketing spend in a period ÷ new customers acquired in that period. LTV (Lifetime Value) = average revenue per customer × gross margin % ÷ monthly churn rate. The ratio that matters: LTV:CAC should be 3:1 or better for most SaaS businesses. Payback period (CAC ÷ monthly gross profit per customer) should be under 18 months. Both should be improving over time as your brand, word of mouth, and sales efficiency improve.
From model to board pack
Turn your financial model into a board-ready slide in 60 seconds
DeckBrief formats your financial projections into the standardised output that investors and board members expect.
Generate Your Financial Slides Free →