How to build a startup financial model from scratch
A bottom-up, assumption-driven financial model in Google Sheets — three statements, three scenarios, and the cells every investor will change before they read anything else.
A startup financial model is not a forecast — it's a thinking tool. The goal isn't to predict the future accurately (you can't) but to make your assumptions explicit so you and your investors can argue about them. A good model lets a reader change one cell and see the impact across the next 36 months.
This guide builds a model in Google Sheets from a blank tab. By the end you'll have an assumptions sheet, a monthly P&L, a cash flow forecast, and a simple balance sheet — plus base, upside and downside scenarios.
The three sheets you actually need
Assumptions: every number a reader might want to change (price, churn, CAC, headcount, salary bands, COGS %, payment terms). Nothing in the P&L or cash flow should be a hard-coded number — everything should link back here.
P&L (monthly, 36 months): revenue, COGS, gross profit, operating expenses by function, EBITDA. Stick to monthly granularity; quarterly hides too much.
Cash flow (monthly, 36 months): starting cash, cash in (revenue × collection lag), cash out (expenses × payment lag), ending cash. The single most important number on the model is 'months of runway remaining'.
Why scenarios matter more than the base case
Every founder is wildly optimistic. Investors know this. Showing only a base case signals naivety. Showing base, upside (+30% revenue), and downside (-30% revenue and +20% costs) signals you've thought about what could break and how you'd respond.
The downside case is the most important one. Investors aren't trying to model your win — they're trying to model how their capital survives if you don't.
Step by step
- 01
Create three sheets in Google Sheets
Tabs: Assumptions, P&L, Cash. Add a Scenarios cell at the top of Assumptions with a dropdown for Base/Upside/Downside. Every assumption below will reference this.
- 02
Build the assumptions sheet
Group inputs by section: Revenue (price, conversion, churn, ARPC), Acquisition (channels, CAC, leads/month), People (headcount by month, salary by role, benefits %), Costs (COGS %, hosting, software, rent), Capital (starting cash, funding events). Use named ranges so formulas stay readable.
- 03
Model revenue bottom-up
Don't type a top-line growth rate. Model: new customers/month = (paid traffic × conversion) + (organic × conversion). Total customers = previous + new − churned. Revenue = total customers × ARPC. This drives every other revenue number.
- 04
Model costs by category
COGS scales with revenue (hosting, payment processing, fulfilment). Operating expenses scale with team (salaries, benefits, tooling per head). Fixed costs (rent, audit, legal) stay flat. Keep one line per category — don't over-detail.
- 05
Build the cash flow
Critical difference from P&L: timing. If you bill annually but recognise revenue monthly, cash in lumps; revenue smooths. If suppliers give you 30-day terms, cash out lags by a month. Get this right or runway numbers will be off by months.
- 06
Add the runway calculation
Two cells at the top of the cash flow: current cash, monthly burn (average of the last 3 months' net cash). Runway = cash ÷ burn. This is the number investors look at first, every time.
- 07
Build the three scenarios with one formula trick
Use a CHOOSE() or IFS() formula at each assumption that returns a different value based on the Scenarios dropdown. Now one dropdown change recalculates the whole model.
- 08
Add a one-page summary sheet for investors
Headline metrics only: monthly revenue, monthly burn, runway, headcount, key efficiency ratios (gross margin, LTV/CAC, payback period). This is what gets emailed. The full model is the appendix.
- 09
Pressure-test with three questions
1) If we miss revenue by 30%, when do we run out of cash? 2) What's the latest we can decide to cut burn? 3) What single assumption, if wrong by 50%, breaks the model? If you can't answer all three in 60 seconds, the model isn't built right yet.
Key takeaways
- Every number in the P&L and cash flow should be driven by an assumption cell — no hard-codes.
- Monthly granularity is mandatory. Quarterly models hide cash crunches.
- Three scenarios beat one. The downside case is the most important one.
- Runway = cash ÷ trailing 3-month burn. Put it at the top of every sheet.
- Bottom-up revenue (price × customers) beats top-down growth-rate assumptions every time.
Troubleshooting
Frequently asked questions
+How far out should I forecast?
36 months in monthly detail is the standard for a Series A and earlier. Pre-seed can get away with 18 months. Anything beyond 36 months should be annual and treated as illustrative — no one believes month-by-month numbers four years out.
+Excel or Google Sheets?
Google Sheets, almost always. Investors can comment inline, share without version chaos, and review on mobile. The only reason to use Excel is if the company has an existing finance team that runs on it.
+Do I need an accountant to build the model?
No. A solid founder model is the right thing for fundraising and operations. Bring in an accountant once you're raising a priced round, applying for SBA debt, or hitting $1M+ ARR and need GAAP-clean books.
+How is a startup model different from a small business model?
Small business models emphasise stable margins, predictable cash flow, and tax planning. Startup models emphasise growth assumptions, runway, and unit economics (CAC, LTV, payback). Both have P&L and cash flow, but the questions they answer are different.
+Should I include funding rounds in the model?
Yes. Model the round as a cash inflow on a specific month and the resulting runway extension. Showing 'pre-round runway' and 'post-round runway' explicitly is much more credible than a magical cash-never-runs-out forecast.
+What if I don't know my CAC yet?
Use a placeholder based on the industry benchmark (e.g. $200–500 for SMB SaaS, $2k–10k for mid-market, $20k+ for enterprise) and flag the cell yellow. The act of naming the assumption is more important than getting it perfect — you'll refine once you have channels live.
We write investor-ready plans, financial models and decks that close meetings.
Book a planning call