What Multiple Should You Pay for a Business?

The immediate answer

There is no universal multiple you should pay — it depends on sector, size, growth, quality of earnings and how the deal is funded, and anyone quoting a single number without those inputs is quoting a rule of thumb, not an answer for your deal. What you can actually work out, and what matters more than a market average, is your own ceiling: the highest price your funding structure and your required return can support without the deal being fragile. That ceiling is set by working forward from comparable evidence and backward from your own numbers, and taking the lower of the two.

Why the comparable median is not the answer

A comparable-transactions multiple tells you what a set of similar businesses actually sold for — genuinely useful evidence, and the starting point for any sensible conversation about price. But a median is, by construction, the middle of a range that includes deals done under better and worse conditions than yours: some buyers overpaid for strategic reasons that will not apply to you, some sellers accepted a discount because they needed to exit quickly, and every comp carries its own quality-of-earnings and growth profile that your target may or may not share. Treating the median of a handful of comparable deals as the price you should pay skips the part of the analysis where you ask whether your target sits above or below that median on the factors that actually move multiples — growth rate, customer concentration, margin quality, owner dependency — and whether you can even fund a price at that level in the first place.

A useful comparable-transactions read needs a genuine sample: a handful of deals in the same sector and a similar size band, not one anecdote from a broker or a single deal you heard about at a conference. Business valuation methods used in M&A covers how the comparable-transactions method sits alongside the other four, and why a workspace with too few relevant comps should get a valuation identical to one that ignored comps altogether, rather than a number derived from one unreliable data point.

The constraint the comps range does not show you: what you can actually fund

The multiple a comparable range suggests and the multiple your deal can actually support are two different numbers, and the gap between them is where most acquisitions that later feel overpriced actually went wrong. Funding capacity is set by the target's own balance sheet and cash flow, not by what you would like to borrow: a lender will size senior debt against a loan-to-value the target's assets support and against how comfortably projected cash flow services the resulting repayments, and an unknown or thin asset base should cap what a lender is willing to advance rather than being ignored. Equity is the other half — how much you and any co-investors are actually prepared to put at risk, plus the fees and costs of the transaction itself, which are funded from equity and reduce what is available for the price.

Once debt and equity are sized, four covenant-style thresholds test whether the resulting price is one the deal can actually carry once it is trading, not just one that clears on day one: leverage (total debt against EBITDA), interest cover (EBITDA against interest due), debt service cover (cash available against total debt repayments) and fixed-charge cover (cash available against every fixed obligation the year creates — debt service plus any earn-out or deferred instalment falling due). A structure that clears comfortably on leverage but fails debt service cover in a normal trading year is not a safe deal at that price, whatever the comparable range implied it should sell for.

A framework for setting your ceiling

1. Build a genuine comparable range, not a single data point — several deals in the sector and size band, and note the spread, not just the median. A tight spread is more trustworthy evidence than a wide one built from dissimilar deals.
2. Establish the earnings figure the multiple actually applies to. Adjusted EBITDA, unadjusted EBITDA and seller's discretionary earnings are not interchangeable — see EBITDA vs SDE — and a multiple quoted against one figure applied to another overstates or understates the implied price before any negotiation starts.
3. Size the funding stack against the target's own balance sheet, not a percentage assumption — how much senior debt the assets and cash flow actually support, capped rather than assumed, plus what equity you are prepared to commit including transaction costs.
4. Test the resulting structure against your covenant thresholds — leverage, interest cover, debt service cover and fixed-charge cover — in a normal trading year, not the best year in the target's history.
5. Solve for the maximum price the structure clears every threshold at, and compare that ceiling against the comparable range. Where the funding-constrained ceiling sits below the comps range, that is your real number; where it sits above, the comps range — not your funding capacity — is what should discipline the price.
6. Decide how much of any gap a contingent structure can close. An earn-out or a subordinated vendor note can let you agree a higher headline price while keeping the price you actually fund at completion inside your ceiling — see how to structure a business acquisition.

A worked example

What genuinely moves the multiple, beyond the sector average

A few factors reliably shift where a specific business should sit within — or outside — its sector's comparable range, and are worth pricing explicitly rather than leaving to instinct. Growth and its durability matter more than the trailing year alone: a business growing on a handful of contracts that expire next year is a different proposition from one growing broadly across its customer base. Customer concentration is a direct discount — see customer concentration risk — because a business dependent on one or two accounts is worth materially less than the same revenue spread across many. Owner dependency, the quality of the management team left behind, and how clean the add-backs in the EBITDA calculation actually are — see how to calculate EBITDA — all move a genuinely comparable multiple up or down before funding capacity is even considered.

Common mistakes

• Treating the comparable median as a target price rather than a reference range. The median tells you what happened on average, not what you should pay for this specific business with its own strengths and risks.
• Ignoring debt capacity until after a price has already been agreed. Sizing the funding stack after the headline price is settled, rather than before, is how buyers find out too late that the deal they agreed cannot actually be funded on sensible terms.
• Testing leverage but not debt service or fixed-charge cover. A structure can clear a leverage multiple comfortably and still fail on whether the year's actual cash covers what is owed, especially once working capital absorption and any earn-out instalment are accounted for.
• Confusing the earnings figure the multiple applies to. Applying a multiple sourced from EBITDA-based comparables to an SDE figure, or vice versa, moves the implied price by a margin that has nothing to do with the negotiation.
• Pricing the deal at the cap of an earn-out rather than an expected attainment. What is an earn-out covers why modelling the maximum contingent payout as certain overstates the price actually being committed to.
• Not stress-testing a normal trading year. A structure that only clears its covenants in the target's best historical year is not a structure with any margin for a below-plan year, which is the year the covenants exist to test for.

How TrueValue models the ceiling

The deal economics engine runs both halves of this together rather than as two separate spreadsheets. The comparable-transactions method draws on the workspace's own recorded deal history — a minimum of three relevant comps, the same sector, a size band around the target, and the median rather than the mean — alongside the other four valuation methods, so a thin comp set never quietly substitutes for real evidence. Separately, the funding side solves from the target's own balance sheet: secured draws are capped at what the accounts actually support, with an unknown asset base capping nothing rather than inventing a ceiling, and a senior loan above a sensible leverage default is reported rather than silently capped. Four covenants — leverage, interest cover, debt service cover and fixed-charge cover, the last including any earn-out or deferred instalment falling due — are tested by year against the resulting structure, and the engine can solve for the maximum price the structure clears every threshold at, alongside the price implied by comparable evidence. The same figures carry into the IC memo, so the number a committee sees is the number the structure was actually tested against.