How lenders assess recurring revenue

Why this is the whole argument

Every sector on this site has the same shape. The value of the business is not in what you can see. It is in income that arrives again next year: a fee block, a maintenance book, a monitoring base, a service round, a schedule of compliance visits.

That makes one question decisive. Not how much recurring revenue there is, but how much of it a lender will actually treat as recurring. Those are different numbers, and the gap between them is where deals are won and lost.

The word does a lot of work

"Recurring" gets used for at least three different things, and only one of them is what a lender means.

• Contracted revenue. A signed agreement, a term, a price, a notice period. The customer has committed.
• Habitual revenue. No contract, but the customer has come back every year for a decade. Genuinely valuable and genuinely not the same thing.
• Repeat-ish revenue. Work that happened more than once. This is not recurring, and a schedule that quietly includes it is the single most common reason a valuation and a lending decision end up in different places.

Sellers rarely separate these, not usually out of dishonesty but because inside the business they all feel the same. A buyer's first job is to insist on the split, from the ledger, before price is discussed.

Test one: is it contracted, and on what terms

A lender wants the document, not the summary. What it is reading for:

• Term and notice. A rolling arrangement terminable on a month's notice is worth materially less than a three year term with a fixed renewal date.
• Price mechanism. Whether the price can move with costs. In labour-heavy sectors this single clause decides whether a contract stays profitable, which is why it matters so much in facilities management.
• Change of control. Whether the customer can walk, or must consent, when the business is sold. This is checked contract by contract on the largest ones, and it is a frequent late surprise in HVAC and fire and security.
• Who the counterparty is. A managing agent, a main contractor and an end client carry different payment risk.

Test two: can you prove it renews

Assertion is worth nothing here. What a lender wants is renewal history, by value, over at least three years: what was on the book, what renewed, what was lost, and what replaced it.

Two things separate a good answer from a poor one. The first is that it is measured by fee or contract value rather than by count, because those diverge exactly when it matters, which is when the departures were the large ones. The second is that it reconciles to the accounts. A schedule that totals something different from the sales ledger does more damage than no schedule at all, because it makes everything else you have provided suspect.

Test three: why does the customer buy

This is the test that separates lenders who know a sector from lenders who do not, and it is where the sector specialism on this site actually bites.

Work that exists because the customer has a legal, insurance or audit obligation renews far more reliably than work that exists because it is sensible. A food business needs documented pest control because its own audit requires it. A commercial landlord needs fire safety servicing because the law and the insurer require it. A company needs its accounts filed because the deadline comes round again.

A lender who has funded the sector before asks this question early, because it tells them what the income is really made of. One who has not tends to treat the whole figure as turnover, look for tangible security, find very little, and price the uncertainty rather than the risk.

Test four: whose relationship is it

The question every lender eventually asks: if the seller left the day after completion, how much of this leaves with them?

The honest answer determines the structure more than almost anything else. Where the relationship is genuinely with the business, a lender can lend against the income. Where it is with a person who is leaving, the deal needs a retention, a clawback that outlasts the risk, and a handover with the introductions actually scheduled rather than promised. See deferred consideration and earn-outs for how that money is then treated.

This is at its sharpest in accountancy practices, where a long-standing partner may be the only reason several clients have never looked elsewhere, and it is the reason clawback is drafted as carefully as it is in that sector.

Test five: how concentrated is it

A book where the largest client is a fraction of a per cent behaves like an annuity. One where the top three are a third of it behaves like three relationships. Both can be funded. They are not funded the same way, and the difference belongs in the structure rather than in an argument at credit stage. See how lenders assess customer concentration.

The trend, which almost nobody presents

A recurring book is rarely static, and lenders read the direction as carefully as the level. A book that has grown steadily is evidence of a business that wins work as well as keeping it. A book that is flat because new contracts are replacing lost ones is a different business again, and the churn underneath the flat line is the thing worth explaining.

The one that catches buyers is the slowly shrinking book. Attrition of a few per cent a year is easy to miss inside a business and easy to price wrongly from outside it, because the current year's revenue looks fine. A lender modelling a five year facility is not lending against this year. It is lending against the fourth and fifth years, by which time a gentle decline has become the whole margin.

If the book is shrinking, say so and explain why. An owner who has been winding down towards retirement and has not chased new work is a completely different proposition from a business losing customers to a competitor, and the second one needs a plan rather than an explanation. Presenting the trend yourself, with the reason, is far stronger than letting diligence find it.

What good evidence looks like

If you produce one document for a lender on a business of this kind, make it a reconciled schedule with, for every contract or client: who they are, what they pay, how often, since when, on what contractual terms, and with what notice period. Totalled, and tied back to the accounts.

It is unglamorous and it does more work than any forecast. It converts the central claim of the whole deal from something you are asserting into something a lender can check, and a lender that can check something stops discounting it.

The rest of the pack is in the document checklist.

Why the same schedule gets two different answers

Everything above describes a competent assessment. What it does not capture is how differently two lenders will read the identical schedule.

One has funded fee blocks and maintenance books before. It knows what a good renewal history looks like, it knows which clauses matter, and it is comfortable lending against contracted income with little tangible security behind it. The other is applying a general SME template, and a business with almost no assets and no stock does not score well on it.

Bank of England figures for June 2026 show the effective rate on new bank lending to SMEs sitting well above the wider corporate market, at 6.36% against 5.42% across UK private non-financial corporations[1]. Part of that is real risk. Part of it is the cost of being read by somebody who does not recognise the asset.

We know this market and we know specific people who have funded recurring books in these sectors before, so the schedule goes to someone who reads it correctly. That is the entire service, and why Reads is straight about its limits.