Building revenue assumptions assessors won't reject

There is a specific sentence that quietly sinks business plans: "We expect to reach 1,000 customers in year one, growing 20% a quarter thereafter." It sounds precise. It is, in fact, a guess wearing the costume of a forecast — and an endorsing body assessor will treat it as one. The question they ask next is always the same: why 1,000? If your answer is a shrug dressed up as optimism, the viability of your whole plan takes the hit. The alternative is not a better guess. It is a different kind of number entirely — one built from drivers you can defend.

What endorsers are actually looking for

The viability pillar of the endorsement assessment is not testing whether your numbers are big. It is testing whether they are real. Guidance on the Innovator Founder business plan is explicit that viability assessment looks for realistic financial projections with clear assumptions, backed by evidence of market research — customer discovery, competitor analysis, industry data — and a clear explanation of your revenue model and pricing strategy (DavidsonMorris business plan guide).

Read that carefully and you notice it is a list of drivers, not a request for a bigger top line. Market research tells you how many customers exist and how they buy. Competitor analysis and pricing tell you what you can charge. Your revenue model tells you how the money actually flows. A forecast that is disconnected from these — a single growth rate applied to an invented starting point — cannot be evidenced against any of them, which is precisely why it reads as guessed.

The anatomy of a driver-based model

A driver-based model breaks your revenue into the operational inputs you control and then does the arithmetic. For most businesses the chain looks like this:

• The unit. Decide what you actually sell — customers, subscriptions, transactions, contracts, seats. Everything else hangs off this.
• Acquisition. How many prospects reach you each month, from which channel, at what cost. "40 qualified leads a month from paid search at a £35 cost per lead" is a driver. "Word of mouth" is not.
• Conversion. What proportion of prospects become paying customers, benchmarked against a comparable or your own early data. "8% of leads close" is testable.
• Price. The average revenue per unit, anchored to comparable market pricing. "£2,500 average annual contract" can be checked against competitors.
• Retention or churn. How many customers you keep, which compounds heavily over three years. "3% monthly churn" comes from an industry range or a pilot.

Multiply these through and the top line emerges. Crucially, it emerges from the assumptions rather than being imposed on them. Change the conversion rate and the total moves; change the price and it moves again. The number is now a function you can explain, not a figure you defend by insisting.

Notice the strong column produces a smaller, slower first-year number than the guessed one — and it is far more credible for it. That is the recurring pattern: evidenced assumptions are usually more modest than optimistic ones, and assessors reward the honesty.

Why drivers win at the contact point

The Innovator Founder process does not end when you submit the plan. Endorsing bodies conduct contact-point reviews and interviews, and this is where guessed forecasts fall apart. When an assessor asks "why do you expect 1,000 customers?", the founder with a top-line growth rate has nowhere to go — the number was never built from anything, so there is nothing to point at. The driver-based founder walks through the chain: here is the channel, here is what leads cost, here is the conversion rate and where it comes from, here is the price and the comparables. The answer exists because the model was built to contain it.

This is also what separates a defensible model from an AI-generated one that looks plausible but is invented. A driver-based model is deterministic by construction: every output traces to a stated input. If you cannot trace it, you cannot defend it.

Use the free assessment at /assess — five assessments, no card — to test whether your revenue assumptions hold up against the viability criteria before you submit. For more in this series, see /insights.

Connecting revenue to the rest of the model

A driver-based revenue line is the foundation the rest of your model sits on. Your acquisition spend and headcount are operating costs; your growth pace determines your 24-month runway; and the range around your assumptions is what you flex in a sensitivity analysis to show you have stress-tested the plan. A guessed top line breaks all three of those downstream — you cannot stress-test a number that has no drivers.

Once your revenue is built bottom-up, the classic credibility failures become easy to avoid — most of the financial model red flags assessors look for are, at root, symptoms of a revenue line that was never built from drivers. For the tax consequences that follow from the revenue you model, the UK tax map for founders.

For general best-practice references on driver-based and bottom-up forecasting, the Corporate Finance Institute's revenue-modelling resources are a solid technical starting point, alongside GOV.UK's overview of the Innovator Founder route.

Key takeaways

• Endorsers assess whether revenue is realistic and evidenced, not whether it is large — a driver-based model answers that test directly.
• Build the top line from operational drivers you control: the unit, acquisition, conversion, price and churn, each tied to a source an assessor can interrogate.
• Evidenced assumptions usually produce a more modest first-year figure than a guess — and are far more credible for it.
• Driver-based models win at the contact-point review and interview because every output traces to a stated input, so "why that number?" always has an answer.
• Attach a cost to every acquisition driver — a lead volume with no marketing budget behind it is a contradiction an assessor will find.

• revenue-forecasting
• financial-model
• viability
• unit-economics
• assumptions