Most seed-stage financial models are built backwards. The founder decides how much money they want to raise, reverse-engineers a growth rate that justifies the ask, and produces a spreadsheet that shows the right number on the right slide. Investors see hundreds of these. They know the pattern.
A model that earns trust works the other way: it starts with what you actually know, builds from first principles, and shows the investor exactly how you think. The numbers will be wrong; all projections are, but the reasoning should hold. This guide walks through how to build one.
What a seed-stage financial model is actually for
The model is not a prediction. It is a structured argument. It says: given these assumptions about how we acquire customers, how much they pay, and how long they stay, here is what the business looks like in 18-24 months; here is how much capital we need to get there; and here is the burn rate that determines whether we make it to the next round before running out of money.
Investors evaluate the model on three dimensions: are the assumptions reasonable and defensible, does the founder understand what the key levers are, and is the business fundable at this stage? A model that passes all three does not need to predict the future accurately; it needs to demonstrate clear thinking.
The five components of a seed-stage financial model
1. Revenue model
Start with your go-to-market motion and build revenue from the bottom up. The structure depends on your business model.
For SaaS: Monthly new customers x average contract value x retention rate. Break new customers into channels (organic, paid, outbound) and use your current conversion rates to project forward. If you do not have data yet, use conservative benchmarks and say so explicitly.
For transactional/marketplace: Monthly active users x average transaction value x take rate. Model GMV and revenue separately.
For usage-based: Monthly active users x average usage per user x price per unit. Show how usage scales with the product's value delivery.
The most common mistake: applying a percentage growth rate to revenue rather than modelling the underlying drivers. If you cannot explain exactly where the revenue comes from, the model is not ready.
2. Cost structure
Separate fixed costs from variable costs. Variable costs should be tied directly to revenue or customer count; the cost of goods sold (hosting, payment processing, customer success for high-touch models) scales with the business. Fixed costs (salaries, rent, tools) do not.
Build your headcount plan explicitly. List every hire, when they start, and what they cost (salary + benefits + employer taxes, typically 120-130% of salary in the US). Headcount is usually 60-70% of seed-stage burn. If you cannot justify each hire's contribution to the model, cut it.
3. Gross margin
Gross margin = (Revenue - Cost of Revenue) / Revenue. For software businesses, target gross margins above 60% at seed stage, with a path to 70%+ at scale. Margins below 40% at seed stage in a software business raise questions about the scalability of the model and whether the business is truly software or services.
Show gross margin by month. Investors will look for the inflection point where margins expand as fixed costs in COGS are amortised across more customers.
4. Runway and burn rate
Monthly burn = total cash out - cash in. Runway = current cash / monthly burn. This is the number the investor uses to assess urgency and set the terms of the conversation.
For a seed round, most investors want to see 18-24 months of runway post-investment. Less than 18 months, and the pressure to raise again before achieving milestones is too high. More than 24 months, and the question becomes why you are raising so much.
5. Key metrics and milestones
Show the metrics that matter for your business: ARR, MRR, CAC, LTV, payback period, and net retention. Then show what the metrics look like at 12 months and 24 months post-investment, and what those metrics imply about Series A readiness.
Investors assess seed deals against a Series A benchmark. For SaaS, this typically means $1-3M ARR with strong net retention (>100%) and a clear path to efficient growth. Build your model so you can show where you will be against that benchmark at the time of your expected next raise.
The assumptions that get challenged most
Churn rate. Founders consistently underestimate churn. If you are pre-product, use conservative benchmarks (5-8% monthly churn for SMB SaaS, 1-2% for enterprise) and explain why your retention will be at the better end.
Sales cycle length. For B2B, a sales cycle you are modelling as two weeks is often six to eight weeks in practice. A compressed sales cycle assumption compresses the cash conversion cycle and makes the model look better than it will be.
Conversion rates. Especially for outbound. A 10% email-to-demo rate sounds reasonable until investors ask what your data shows.
Headcount ramp. New hires take time to reach full productivity. A sales rep model that shows full quota attainment from month one will be questioned.
Structuring the model for investor review
Three tabs are the standard structure: Assumptions (all input variables in one place, clearly labelled), Model (the month-by-month projection built from those assumptions), and Summary (a dashboard showing the key metrics over time, suitable for a quick read).
Use colour coding: blue for inputs, black for formulas. Investors sometimes ask to stress-test assumptions; blue cells tell them exactly where to look.
Share the model in Excel or Google Sheets, never PDF. An investor who cannot see the formulas cannot trust the model. A model that cannot be interrogated is a red flag.
Put your fundraising documents, including the financial model, in a structured data room so investors can find everything in one place. For data on what metrics European seed investors are benchmarking against in 2026, the Pitchwise State of Fundraising is a useful reference when stress-testing your projections.
Frequently asked questions
How many months of projections should a seed-stage model show?
24-36 months is standard. 24 months shows enough to demonstrate the trajectory and Series A readiness. 36 months can be useful if your business has a longer ramp (enterprise sales, marketplace cold start). More than 36 months at seed stage strains credibility -- the further the projection, the less reliable.
Do I need an accountant to build my seed model?
No. Most seed-stage models are founder-built, and investors prefer it that way -- a model the founder built and can explain in detail demonstrates understanding of the business. Use an accountant or CFO to review for errors before sharing, but not to build the original model.
What is the right amount to raise at seed?
Raise enough to reach the milestone that makes your Series A a strong raise, plus a 20-30% buffer for model error. The milestone is typically a combination of ARR, user growth, or technical progress. Work backwards from the milestone, calculate the headcount and costs required to get there, and that number, plus buffer, is your target raise. Use your data room to package the supporting documents alongside the model.
