On paper, an earn-out looks neat: pay some cash now, pay the rest only if the business keeps performing after the sale. In practice, it often means the person who built the company hands over the keys and then waits to see whether the new owner chooses to drive it into a wall, politely or otherwise.
This tension explains why earn-outs appear in so many deals. They are a price bridge, but also a control transfer with a delayed bill attached. Once the buyer owns the business, the founder may still be tied to the outcome, while having far less say over the levers that decide it.
What the business was
An earn-out is a slice of the sale price paid only if the acquired business hits agreed targets after completion. Targets are usually measured over one to three years and are typically plain enough to sound objective: revenue, gross profit, EBITDA, net income, customer retention, or a product milestone.
This objectivity is the sales pitch. The buyer says, “If the business performs the way you told us it would, you get the rest of the money.” The seller hears, “If I am right, I get paid properly.” Both can sign the same document and still interpret it as two different deals.
The structure is common enough to matter. A 2022 SRS Acquiom report stated earn-outs appeared in 24% of private M&A deals worth between $10 million and $1 billion, with the contingent part averaging 15% of total value. This is not a niche clause tucked into the back of the paperwork. It is a standard way of pushing risk out of the purchase price and back onto the seller.
HP’s 2011 purchase of Autonomy for about $11.1 billion is the cautionary poster child. HP’s 2012 Form 10-K later detailed the deal and the ugly aftermath, including an $8.8 billion write-down. The fight over Autonomy’s true worth became a public argument about valuation, accounting, and who gets blamed when a promised future does not arrive.
What actually happened
Buyers usually want an earn-out because it lowers the cost of being wrong. If the target business is growing fast, but the future is still foggy, the buyer can offer a higher headline price without paying all of it upfront. The seller gets the promise of upside. The buyer postpones part of the cheque until the numbers prove themselves.
This sounds balanced until you remember who controls the machine after closing. The buyer holds the sales team, the budget, the pricing, the product roadmap, the reporting systems, and usually the right to decide how much shared overhead gets shoved onto the acquired business. If the earn-out depends on EBITDA, a bigger management charge can quietly make the target harder to hit. If it depends on revenue, rerouting leads to another product line can do the same job more elegantly.
Earn-outs stop being a valuation tool and start looking like a custody battle over the future of the business. A founder may still be working inside the company, but the important decisions now belong to someone whose incentives are different. The buyer may be trying to improve the whole group. The earn-out only cares about the piece that was sold.
Microsoft’s $69 billion acquisition of Activision Blizzard in 2023 was not a classic earn-out deal, but it still showed the same instinct at scale. Buyers do not like to write a cheque and then leave value creation to chance. They try to manage the timing, the control, and the integration risk first, then pay as the picture clears.
Where the money went
The money in an earn-out can disappear without anyone stealing it. That is the irritating part.
A buyer can move salespeople to another division, bundle the acquired product into a wider package, reprice the offering, delay a launch, or allocate more corporate overhead to the target. Each move can be defended as sensible for the larger business. Each move can also make the earn-out number worse.
Courts have seen this movie before. In Western Filter Corp. v. Argan, Inc. in 2009, the seller alleged that the buyer had diverted contracts and resources away from the business, depressing the result tied to the earn-out. The point of the case was not that every integration decision is suspicious. It was that once one side owns the levers, it can do damage without leaving fingerprints.
The accounting side matters just as much. If the agreement does not lock in how revenue is recognised, how costs are allocated, and which accounting policies will be used for the earn-out calculation, the buyer can change the measured outcome without changing the underlying business very much at all. Accounting consistency clauses are therefore important. Without them, the numbers can be made to tell a different story from the trading desk.
What was knowable at the time
Founders do not need hindsight to know the basic risk. The risk is built into the structure. The buyer’s ideal outcome is to own the business and not owe the deferred piece. The seller’s ideal outcome is to still benefit from growth after losing control. These interests do not line up by accident.
The contract language matters more than the cheerfully simple price headline. The important clauses limit what the buyer can do after closing. A covenant to operate, often framed as a “best efforts” or “commercially reasonable efforts” promise, tries to keep the business running in a way that does not sabotage the earn-out. “Best efforts” is the harder standard. “Commercially reasonable efforts” gives the buyer more room to argue that its decision was rational.
Anti-diversion clauses try to stop the buyer from redirecting customers, leads, or staff elsewhere. Information and audit rights let the seller see the numbers and challenge the calculation. A dispute process, ideally with an independent accountant or arbitration, keeps the whole thing from becoming a years-long courtroom grind. A change of control clause matters too, because if the buyer sells the business before the earn-out ends, the seller should know whether the promised payment survives the handover.
There is also a legal floor in the background: the implied covenant of good faith and fair dealing. Sellers have used it when they believe the buyer has gamed the arrangement. In Viking Global Investors LP v. Baidu, Inc. in 2018, the court did not find a breach even though the buyer’s actions affected the earn-out. Bad feeling is not the same as legal proof.
What it means for anyone running something similar
The sharp question is not whether earn-outs exist. They will keep existing because they solve a real problem for buyers who do not want to overpay for uncertain growth. The sharper question is how much control a founder is willing to give away in exchange for a cheque that depends on someone else’s behaviour.
A founder signing one of these deals is selling a company and also signing up for a ruleset about who controls the sales engine, how costs are charged, what counts as performance, and who gets to interpret the spreadsheet when the numbers land slightly inconveniently. The clause that looks boring in the draft may be the one that decides whether the last chunk of the price is real or decorative.
If the buyer owns the levers, why should anyone be surprised when the levers get pulled in the buyer’s favour?