How to Build a Financial Model for Your UK SaaS Startup
How UK SaaS founders should build a financial model for fundraising: 6-step process, three scenarios, unit economics, and UK tax lines investors check.
For the impatient - the four rules of a fundraise-ready SaaS model:
- Build order MRR roll-forward, then revenue, then hiring plan with employer NIC and pension, then monthly P&L, then cash, then runway, then raise ask
- Run 3 reconciliation checks before sending to investors MRR schedule ties to P&L revenue every month; balance sheet balances every month (assets = liabilities + equity); closing cash equals balance sheet cash
- Structure Bottom-up from sales-rep capacity and ACV, three scenarios in parallel columns (base, optimistic, pessimistic), formulas not hardcoded numbers
- Two model rule The 3-5 year model is for raising and exits. Run the business on a 12-month operational forecast updated monthly.
The full 6-step process, scenario design, UK tax lines and the shape investors look for are below.
A 3 to 5 year financial model goes out of date within months, and investors know it. They ask for one anyway because they want to see your vision in the numbers - how you think, where the business is going, and whether the plan holds together. Investors do not need perfect numbers; they need a plan that makes sense and a founder who understands it.
And they will test that. Angels usually review the model themselves, while VCs have analysts who go through it line by line to check whether the numbers and assumptions hold up. Here is the process ScaleWithCFO follows when building models as a fractional CFO, and the checks investors run on the other side of the table.
| Step | What you do | What investors check |
|---|---|---|
| 1. Start from your history | Clean and reconcile 12-24 months of P&L, balance sheet and MRR schedule | Does the model reconcile to the accounts? |
| 2. Build revenue bottom-up | Clients x pricing, with churn and expansion from your actual data | Are the assumptions defensible? |
| 3. Build costs from the hiring plan | Headcount drives 60-75% of costs, including employer NIC and pension | Does spend scale with the growth story? |
| 4. Run three scenarios | Base, worst and best - each with its own hiring plan | What happens when things slip? |
| 5. Produce the statements | P&L, balance sheet, cash flow and unit economics | Does the balance sheet balance and cash reconcile? |
| 6. Check the story the numbers tell | The benchmarks below | Does the shape say scalable business? |
Start from your history
No forecast will ever be taken seriously if the history is a mess. Before building anything forward-looking, clean and reconcile 12 to 24 months of P&L, balance sheet and MRR schedule, because if your last 12 months show 25% growth and the model shows 60%, the model has to explain exactly where the acceleration comes from. Our guide on financial data accuracy covers what to check.
Build revenue bottom-up
"The UK market is £2 billion and we will capture 1%" is the line that gets models dismissed. Most credible models show 5-10% of a realistically defined market, and even 10% is on the high side. Build from your actual sales motion instead: start with the MRR schedule, drive new MRR from leads, conversion rate and deal size, and take churn and expansion from your own data.
And never hardcode a number. Do not type "£50,000 revenue" into a cell - use formulas, clients x pricing, so the whole model updates when an assumption changes. Hardcoded numbers are the first thing an analyst finds and the fastest way to lose credibility. For the tab structure, see our SaaS financial model template.
Build costs from the hiring plan
Headcount is 60-75% of total costs, so the hiring plan is the cost model. Each role needs a start date, a salary with employer NIC at 13.8% and pension on top, and a department. Link hires to revenue milestones - one customer success manager per 15 new customers, the next engineer at £500K ARR. The spend ratios everything else should follow are in the benchmarks table below.
Run three scenarios
Base, worst and best - and each scenario needs its own hiring plan, not just revenue scaled up and down. The drivers change between scenarios: acquisition rate, churn, hiring pace. They also have to stay consistent - more customers in the best case means more support staff and higher hosting costs.
Produce the statements
A monthly P&L for years one and two, a balance sheet that balances in every month, and a cash flow where closing cash equals balance sheet cash. Watch cash timing: revenue recognised is not cash received - annual upfront invoicing brings cash in month one but revenue over twelve months, and 30-day payment terms delay every receipt.
Unit economics must be derived from the model, never hardcoded. A quick example: £30,000 a month on sales and marketing landing 10 new customers means £3,000 CAC; at £500 monthly revenue per customer and 75% gross margin, that is £375 of monthly gross profit per customer, so CAC payback is 8 months - comfortably inside what UK investors expect at Seed and Series A.
The story the numbers must tell
Investors read the shape of the model before they read the cells. Growth, margins and spend ratios either tell the story of a scalable business or they do not.
| What investors read | What good looks like |
|---|---|
| Growth (VC-funded) | Roughly 3x, 3x, then 2x year on year - at £1m ARR, the next milestone is £2m+ within 12 months |
| Growth (bootstrapped or angel-funded) | At least 60% a year, ideally with months close to profitability |
| Revenue shape | Steady month-on-month growth, never lumpy, with recurring revenue rising as a share of total |
| Sales & marketing | 20-30% of revenue, growing in line with the growth you claim |
| Customer support | 10% of revenue at most - a product that needs more support does not scale |
| Gross margin | Climbing steadily to 70-80% - anything near 90% means costs are hiding in the wrong place |
| R&D and product | The biggest share of spend, because that is what builds long-term value |
| Efficiency | CAC falls as you scale and revenue per employee rises |
| Consistency | The same metric shows the same number in every tab - one mismatch and investors doubt everything |
Two models, not one
The 3 to 5 year model is for raising and exits. You run the business on a 12-month operational forecast, updated monthly with actuals, which tells you how long your cash lasts and what to do about it. Trying to do both jobs in one giant model usually means it does neither well.
Before it goes to an investor
Run three reconciliation checks: the MRR schedule ties to the P&L, the balance sheet balances in every month, and closing cash on the cash flow equals cash on the balance sheet. A simple model that passes those beats a 50-tab model that fails them.
Then two rules for the conversations themselves. Know how the model works even if someone else built it, because your connection with a new investor is thin at the start, and one wrong answer about your own numbers can break it. And do not send the full model after the first call - start with the pitch deck, and share the model once they are seriously interested.
Building a fundraise-ready model takes two to four weeks with clean data, and it should not retire after the round closes: updated monthly with actuals, it becomes the tool you actually run the business with.