A three-statement model links your income statement, balance sheet, and cash flow into one internally consistent projection.
The three-statement model is finance's fundamental artifact: a spreadsheet (or increasingly a specialized tool like Runway, Mosaic, or Pigment) where the income statement (P&L), balance sheet, and cash flow statement are linked such that changing any driver — hiring plan, pricing, churn rate, customer acquisition cost — flows through all three consistently. For an operating startup, it serves two functions: (1) the fundraising deliverable investors expect to inspect and stress-test, and (2) the operational planning tool that lets you answer scenario questions in minutes rather than days. Most startups have neither a real one nor a plan to build one; they have a P&L, a hiring plan, and a wish that the two connect.
Income statement (P&L): revenue minus expenses over a period, ending in net income. Answers 'are we profitable?' Balance sheet: assets, liabilities, and equity at a point in time. Answers 'what do we own and owe?' Cash flow statement: how cash moved over a period, split into operating, investing, and financing. Answers 'where did the cash actually go?' The three connect: net income flows into retained earnings on the balance sheet; changes in balance sheet items (AR, AP, deferred revenue) drive working capital in the cash flow statement; ending cash on the cash flow reconciles to the cash line on the balance sheet.
A model is only useful if drivers are separated from calculations. Structure: a 'drivers' or 'assumptions' tab (growth rates, headcount plan, ARPU, churn, gross margin, sales cycle length, CAC by channel), separate from the calculated statements. Change a driver, watch the statements update. Anti-pattern: hardcoding future revenue as a single number ('$5M in 2027') without underlying units-and-price drivers. Correct pattern: 'starting ARR × (1 + monthly growth × 12) × (1 - annual churn) = ending ARR,' with each factor a driver you can vary. Two-way flexibility (change growth OR change churn, both flow through) is what makes the model a tool rather than a document.
Standard corporate three-statement models are built for stable businesses; startups need adaptations. (1) Bookings vs. billings vs. revenue vs. cash — these diverge widely in SaaS with annual/multi-year contracts. Model each separately. (2) Deferred revenue and accounts receivable as first-class items on the balance sheet — for SaaS, deferred revenue is often the largest liability. (3) Cohort-based revenue projection — new logo revenue by acquisition cohort with retention curves, not a single blended MRR line. (4) Headcount as the primary driver of opex — model hires by role, month, fully-loaded cost. (5) Runway calculation clearly visible — cash balance / monthly burn, refreshed automatically. A well-built startup model makes runway, default-alive analysis, and unit economics answerable from the same file.
Build the model to switch between named scenarios: Base (management's realistic case), Upside (favorable execution and market), Downside (things go wrong — slower growth, higher churn, longer sales cycles). Each scenario is a set of driver values, selectable via a dropdown that flows through the entire model. For fundraising, publish the Base scenario and be prepared to walk through Upside and Downside on request. For operating decisions (should we hire the next 5 engineers?), model each scenario's implication on runway and default-alive status. Anti-pattern: 'we'll adjust the drivers manually to show scenarios' — inevitably breaks under time pressure, produces inconsistent numbers between conversations.
(1) Cash doesn't reconcile — the ending cash on the CFS doesn't match the cash line on the balance sheet. Any sophisticated investor will find this in 5 minutes and lose confidence in everything else. (2) Revenue growth without corresponding cost — modeling 5x revenue growth with 20% headcount growth. Unless you're modeling extreme leverage explicitly, it looks naive. (3) Missing working capital dynamics — customers paying annual upfront changes cash flow massively; ignoring it produces useless runway projections. (4) Assumptions inconsistent with actual data — projecting 5% monthly churn when actual is 8%. Reconcile every driver to the last 3-6 months of actuals as a sanity check. (5) No sensitivity analysis — a model where you can't see 'what if CAC is 30% higher' is a document, not a model.
Investor directory · Fundraising library · Articles A–Z · Company funding database