Two firms can walk into the same buyer's office with the same adjusted EBITDA on the front page of their information memorandum — £500,000, say — and walk out with offers nearly £1.6m apart. Nobody has moved the goalposts on either firm's size. What has moved is the buyer's confidence that the £500,000 will still be there next year, under someone else's ownership, without the founder in the room. That confidence is quality of earnings, and it is a bigger lever on price than growing the EBITDA line itself.
Founders preparing for a sale spend a great deal of energy trying to make the EBITDA number bigger. That is not wasted effort, but it is the smaller half of the job. The larger half is making sure a buyer's diligence team believes the number they are already looking at — and that comes down to four things: how much of the revenue behind it repeats without being re-won, how much sits with one client, how much depends on the founder personally, and how much of the number was manufactured by add-backs that will not survive scrutiny.
Smaller UK accountancy and IFA practices are usually priced on a multiple of gross recurring fees, which our guide to GRF multiples covers in detail. Once a firm is large enough to interest an institutional or private-equity buyer — typically once it has real management depth and a fee base running into the millions — the conversation moves onto the same footing as any other business acquisition: an enterprise value expressed as a multiple of adjusted EBITDA. The mechanics of the bridge from enterprise value to the cheque a seller receives are covered in our note on locked box and completion accounts; this piece is about what happens before that bridge is even built — how the EBITDA figure itself gets tested, accepted, discounted or rebuilt.
That shift matters because an EBITDA multiple is far more sensitive to earnings quality than a GRF multiple is. GRF is a reasonable proxy for quality on its own, because it already excludes most one-off work. EBITDA does not exclude anything by default — it is a bottom-line profit figure that has to be pulled apart, line by line, before anyone can agree what it is really worth a multiple of.
The simplest question a buyer asks of any fee line is whether it would arrive again next year if the client did nothing at all. Statutory accounts, tax compliance, payroll, audit, an ongoing legal retainer, an IFA's annual review fee on assets already under advice — all of these repeat without being re-won. A restructuring assignment, a single transaction, a discrete advisory project, however well it paid, has to be sold again from a standing start.
A firm built mostly on the first kind of work is, to a buyer, close to an annuity. A firm built mostly on the second is a business that has to re-earn a meaningful share of its income every twelve months, which is a fundamentally riskier thing to buy at any price. This is also where founders quietly inflate their own sense of quality: work that has recurred for five years by habit is not the same as work that is contractually or structurally likely to recur, and a diligence team will ask for evidence of the difference rather than take the pattern on trust.
Our note on how UK accountancy practices are valued sets out the general principle: a firm where no client is a meaningful share of fees is worth more than one where a single relationship carries the business. The reasoning is straightforward. A concentrated client can leave for reasons that have nothing to do with the buyer — a change of ownership at the client, a personal falling-out with the founder who is leaving, a decision to bring work in-house — and when they go, they take a fixed slice of EBITDA with them on day one of new ownership. A buyer prices that risk by discounting the earnings attributable to the concentrated relationship, by applying a lower multiple to the whole firm, or both. Diversifying the client base is slow, deliberate work, and it is one of the few quality improvements that genuinely cannot be done in the run-up to a sale.
A buyer's diligence team will ask, directly, who a client would call if the founder disappeared for three months. In an owner-dependent firm, the honest answer is "me" — for technical sign-off, for the relationship, for the one difficult client who has never dealt with anyone else at the firm. That answer is not something that shows up as a cost in the accounts, so it never gets caught by an add-back adjustment, but it is one of the most consistent reasons a buyer applies a lower multiple to an otherwise clean set of numbers.
The fix is structural rather than financial: a second tier of people who hold their own client relationships, technical review delegated away from the founder on routine files, and a founder who can genuinely take a fortnight's holiday without the phone ringing. Firms that can point to this before a sale process starts are pricing from a position of strength rather than promising it will happen after completion. Our note on what happens to your team when you sell covers the related question of how that second tier is protected once a deal is agreed.
An add-back is a cost in the statutory accounts that the seller argues would not exist under new ownership or will not recur, and it is added back to reported profit to produce "adjusted EBITDA." Some add-backs are entirely legitimate: a redundancy payment tied to a one-off restructuring, professional fees on an aborted acquisition, a genuine below-market or above-market owner's salary that needs adjusting to a fair rate for the role. Others are the first thing a diligence team strips back out.
The standard test a buyer applies is whether an item has appeared more than once across the last three years of accounts. A cost labelled "one-off" that shows up in two of the last three years is not one-off, whatever the seller calls it. Family members on the payroll without a defined, evidenced role get the same treatment, as do personal expenses run through the company and management charges to connected entities that do not reflect an arm's-length service. None of this is unusual or aggressive on the buyer's part — it is the ordinary discipline of a quality of earnings review, and any seller who has not been through the exercise before consistently underestimates how much of a claimed add-back list survives it.
Illustrative figures, constructed to show the mechanism. Not drawn from any particular transaction.
Both firms present an adjusted EBITDA of £500,000 at heads of terms. Both are asked, in due diligence, to justify the add-backs and the revenue mix behind that number.
| Firm A — compliance-led | Firm B — same headline | |
|---|---|---|
| Adjusted EBITDA claimed | £500,000 | £500,000 |
| Recurring fee income | 91% | 58% |
| Largest client, share of fees | 4% | 21% |
| Add-backs claimed | £35,000 | £220,000 |
| Add-backs accepted after review | £30,000 | £48,000 |
| Confirmed adjusted EBITDA | £495,000 | £328,000 |
| Multiple applied | 6.5x | 5.0x |
| Indicative enterprise value | £3,217,500 | £1,640,000 |
Firm A's £35,000 of add-backs was a redundancy payment from closing a satellite office, backed by a settlement agreement, with one small item — a training course — rejected as an annual cost rather than a one-off. The confirmed EBITDA barely moved, and because recurring revenue was high and no client carried real weight, the full multiple applied.
Firm B's £220,000 of claimed add-backs was made up of an owner's salary uplifted to a claimed market rate, a spouse's salary against an undefined role, a "one-off" systems project that had in fact run in two of the last three years, and a bad debt write-off treated as non-recurring for the third year running. The buyer's review accepted a reduced salary benchmark and a small genuine part-time bookkeeping element, and rejected the rest outright, taking confirmed EBITDA down by nearly a third. Concentration and owner-dependency then took a further half-turn off the multiple. The result: an identical number on the front page of the information memorandum, and a gap of £1,577,500 in what each firm was actually offered.
None of this changes the size of the business. It changes how much of the EBITDA a buyer is willing to believe, which is the number the multiple is actually applied to.
This article is general information about how buyers assess earnings quality in a professional-services acquisition, not legal, tax or investment advice, and Founder Capital LLP is not authorised or regulated by the Financial Conduct Authority. Nothing here is an offer or invitation to invest; any investment opportunities we reference are available only to professional investors, certified high-net-worth individuals and self-certified sophisticated investors, are not open to retail investors, and capital is at risk and returns are not guaranteed. The transaction figures in this article are illustrative and are not drawn from any particular deal. Please take specific advice from a qualified corporate finance adviser and accountant before you sign anything.
It means testing whether the adjusted EBITDA in your accounts is a fair estimate of what the firm will genuinely earn next year, under new ownership, without you personally holding it together. A buyer's quality of earnings review does not ask whether last year's number is accurate — it usually is — it asks whether the number is repeatable. Revenue that has to be re-won, costs that quietly recur despite being labelled one-off, a client base concentrated in one relationship and a founder who is the only person clients will deal with all lower the quality of an EBITDA figure without changing the figure itself. Two firms can report an identical adjusted EBITDA and be worth very different amounts, because quality, not size, is what the multiple is actually paid for.
Work that repeats without being re-won each year: statutory accounts, tax compliance, payroll, audit, ongoing legal retainers and IFA review fees on assets already under advice. Project work — a one-off restructuring, a single transaction, a discrete advisory engagement — has to be sold again from scratch, so a buyer discounts it heavily or excludes it from the base entirely. The practical test is simple: if the client did nothing at all, would the fee arrive again next year without a fresh conversation? If yes, it is recurring. If it depends on winning a new instruction, it is not, however reliably it has turned up in the past.
It moves both the EBITDA a buyer will accept and the multiple applied to it, which compounds. A firm where the largest client is a low single-digit share of fees is, in a buyer's eyes, close to interchangeable client by client. A firm where one relationship is a fifth of fee income is a single point of failure, because that client's decision to leave — for reasons that may have nothing to do with the buyer — removes a slice of the business overnight. Buyers respond by discounting the earnings attached to the concentrated client, applying a lower multiple to what remains, or both, which is why concentration is worth fixing years before a sale, not disclosing at due diligence.
The ones with paper behind them and no recent history of repeating. A redundancy payment tied to a signed settlement agreement, in a year with no comparable cost before or since, is a clean add-back. An owner's salary genuinely uplifted to a benchmarked market rate for an equivalent principal is usually accepted, though often at a lower figure than claimed. What gets rejected is anything that shows up more than once across the last three years' accounts — that is the buyer's standard test for whether “one-off” is true — and family salaries that do not map to a defined, evidenced role. An add-back a seller cannot document with an invoice, contract or minute is treated as if it does not exist.
It can be fixed, but it takes years, not months, which is why it needs to start well before a sale process. The core moves are the same across accountancy, legal and IFA firms: put a second tier of people in front of the largest clients so the relationship is with the firm rather than one person, delegate technical sign-off on routine files, and step back from being the only person who can handle a complaint or a difficult renewal. Buyers test this directly by asking who clients would call if the founder were unavailable for three months. A confident, specific answer protects the multiple; a hesitant one invites a discount.
