Financial Forecasting UK —
3-Year Models That Get Funded & Drive Decisions

‘We only need to capture 1% of a £10bn market’ is the most common (and most immediately discredited) revenue forecasting approach. Investors and lenders want to see revenue built from the bottom up — number of customers, average contract value, conversion rates from pipeline, churn rates — not a percentage of a large number. We build all revenue forecasts from identifiable customer-level assumptions.

A P&L forecast shows profitability; a cash flow forecast shows when money arrives and leaves. Profitable businesses regularly run out of cash because of working capital timing — debtors paid slowly, creditors paid quickly, capex hitting before revenue. Every financial model we build includes a fully integrated monthly cash flow statement.

Investors and bankers always ask ‘what if revenue is 20% below plan?’ A model with no sensitivity analysis forces you to rebuild it on the spot or admit you haven’t considered downside scenarios. We include sensitivity analysis on the three most impactful assumptions in every model.

Gross margin assumptions that aren’t explained — ‘we expect 70% gross margin’ without explaining why — are the first thing experienced investors challenge. Every margin assumption must be tied to a specific cost structure, pricing model or comparable benchmark. We document every margin assumption with its underlying rationale.

Our standard 3-year model includes: monthly P&L (revenue by product/channel/customer, COGS, gross margin, operating expenses, EBITDA), monthly cash flow statement (operating cash flow, working capital movements, capex, financing activities, closing cash balance), and a balance sheet at month end that reconciles with both. All three statements are fully integrated — a change in any revenue or cost assumption flows automatically through to cash and balance sheet.

For an existing business: 2-3 years of historical P&L and balance sheet, current management accounts, headcount and payroll data, key customer and revenue data, known cost commitments and capex plans, any existing debt schedule. For a startup: unit economics (CAC, LTV, pricing), headcount plan, cost structure assumptions and any existing financial projections. The quality of input data directly affects the quality of the output model.

Bank credit teams use financial forecasts to assess: debt service coverage ratio (EBITDA / annual debt service — typically must be above 1.25x), leverage (debt/EBITDA — typically below 3-4x for SME lending), cash runway (how long until cash runs out without the facility), and stress testing (does the business survive a 20% revenue decline?). We build bank lending models specifically to address all these assessment criteria.

Usually not with the same document — investor models and bank lending models have different formats, different KPI emphasis and different audiences. Investor models focus on growth, unit economics and scalability; bank lending models focus on cash flow stability, debt service and asset coverage. We typically build a primary model for one purpose and adapt it for the other — at reduced additional cost.

For businesses with a model in active use: monthly actuals should be loaded and compared to forecast (actuals versus budget), with a full reforecast quarterly. Annual rebase — rebuilding the 3-year model with updated assumptions — should happen before each financial year end. For businesses preparing for a raise: update the model to reflect current trading before sending to any investor. We offer ongoing model maintenance as a monthly retainer service.