The earn-out is where good deals go to die: how to structure one that actually pays

An earn-out is presented as a compromise on price. It is almost never that. It is a transfer of risk from the buyer to a seller who has just given up control of every lever the payment depends on — and it is agreed, in most owner-led deals, at the precise moment the founder is least inclined to argue about anything.

The arithmetic is worth stating plainly. A founder accepts a headline number where a third or more sits behind two years of performance, then hands over the business. The new owner integrates it, re-prices it, changes the channel mix, cuts or redirects the marketing spend, and loads in a share of group overhead. The metric the founder is being paid on is now being set by someone else, and that someone else has a legitimate commercial interest in the number landing below the hurdle. This is not bad faith. It is structure.

The metric decides the outcome before anything else does. Earn-outs written on EBITDA are the easiest to miss and the hardest to argue about, because EBITDA is the line most exposed to decisions the seller no longer makes — allocated overhead, management fees, restructuring costs, capitalisation policy, the timing of investment. Revenue is far harder to engineer and far easier to verify. Gross profit sits in between and is usually the fairest compromise where a buyer is genuinely worried about a seller buying revenue at any price. If the buyer insists on EBITDA, the definition has to be written out in full in the agreement, with every permitted and excluded adjustment named. A one-line reference to "EBITDA determined in accordance with the buyer's accounting policies" is not a definition. It is a blank cheque.

You cannot be paid on a number you no longer control. Either you keep the levers, or the metric has to be one the buyer cannot move.

Control is the term nobody negotiates, and the one that matters most. If a founder is going to be measured on the performance of the business, the agreement has to say who runs it during the earn-out period. Who signs off pricing. What the minimum marketing and trade spend will be. Whether the product can be pulled from a channel, re-listed under a different brand, or bundled into a group range. Whether the buyer can move the manufacturing, change the pack, or reposition the price point. Each of these can move an earn-out metric by a wide margin, and each of them is a decision the new owner is entitled to make unless the contract says otherwise. Protective covenants around the operating envelope are not an insult to the buyer — they are the only thing that makes the contingent consideration real.

Change of control, strategy shift, and the acceleration clause. Trade buyers get bought. Private equity owners change plan, change management, and change the reporting entity. If the business the earn-out was written against ceases to exist in a recognisable form — because it has been merged into a larger division, sold on, or had its strategy reset — the earn-out must accelerate and pay out in full or on an agreed formula. Without that clause, a seller can do everything right and be paid nothing because a decision was taken two levels above the business they built. This is the single most commonly omitted protection we see, and the one that costs the most when it is missing.

Information rights and a dispute mechanism that is not litigation. A seller in an earn-out is a creditor with no visibility unless the agreement gives them some. Monthly management accounts for the earn-out entity, prepared on a stated and unchanging basis. Separate books for the acquired business rather than a merged ledger nobody can unpick. A right to inspect and query. And when the parties disagree — they will — an expert determination clause that puts the question to an independent accountant on a fixed timetable, rather than a dispute resolution process that only a party with more money than the other can afford to run. A seller who has to sue to get paid usually settles for less than they were owed.

The alternative most founders never think to ask for. Before negotiating the mechanics, it is worth asking whether the earn-out should exist at all. In our experience a buyer offering a contingent structure will often trade it for a lower but certain headline — and a certain number today is frequently worth more than a larger number that depends on someone else's execution. Where the buyer will not move, there are better instruments than a naked contingent payment: deferred consideration secured against the assets or supported by a bank guarantee; a portion of the price taken as equity in the acquiring entity, so the seller participates in the outcome rather than betting on a single line item; or a shorter earn-out on a simpler metric with a lower ceiling. The right question is not "how do I win the earn-out" but "what is the least contingent form this consideration can take".

Why we take this view. We have seen founders write off seven-figure sums with a signature, not because the buyer behaved badly, but because the structure did what it was designed to do and nobody stress-tested it before it was signed. The terms that decide an earn-out are agreed in a week, usually late, usually when everyone is tired and keen to sign. That week deserves more attention than the six months of preparation that preceded it.

If you have a term sheet in front of you with a contingent number in it, that is a document worth having read properly before you sign it — and a conversation worth having while the terms are still open. You can also have these notes sent to you by email.