Earnouts that survive contact with the operator
Most earnouts pay zero. Here is how to draft the ones that pay what they should and how to spot the ones that will not.
Earnouts are the part of a deal where everyone agrees in principle and nobody agrees in practice. The buyer wants protection on the price. The seller wants a path to the rest of the purchase price. Both sides walk away from the closing dinner feeling like they got something. A year later one of them is on a call with their lawyer asking what just happened.
Most earnouts pay zero. That is not a defect. It is a feature. The earnout is a price-discovery mechanism that lets the parties hedge a disagreement about what the company is worth. When the disagreement resolves against the seller, the earnout pays nothing, which is the price the seller agreed to pay for getting the deal done at all.
But there is a separate set of earnouts that pay zero for the wrong reasons. The metric was unmeasurable. The window was too short. The buyer changed the underlying operation and the seller had no recourse. Those are the earnouts that produce litigation, ill will, and a generation of associates learning to draft around the last lawsuit.
Pick a metric the operator can read
The single biggest earnout mistake I see is choosing a metric that requires an audit to compute. Adjusted EBITDA with twelve add-backs is great for a closing model. It is terrible as a milestone the operator has to hit. By the time the metric is computed, the operator does not remember whether they were tracking against the right number, and the seller's accountants are already preparing to argue.
The cleanest metrics are the ones an operator can read off a dashboard. Revenue. Number of customers. Number of units shipped. Renewal rate. Bookings. Cash collections. Each of these can be tied to an existing system of record that produces the same number every quarter without lawyer intervention.
If the metric you want is EBITDA, fine, but lock the add-backs at signing. List them. Cap them. Specify the accounting policies. Anything else is an invitation to a fight.
Pick a window that matches the deal
Twelve-month earnouts almost always fail. The integration takes three months, the new operating cadence takes another three, and by the time the operator is running a normal year there are four months left to perform. That is not enough runway to demonstrate anything.
Twenty-four months is better. Thirty-six is better still for deals where the value is in growth rather than steady-state. The longer window costs the buyer less because the discount rate eats the value of deferred payments. It costs the seller less in disputes because there is time for the operation to find its feet.
Specify what the buyer cannot do
The buyer has the keys to the operation after closing. If the earnout depends on metrics the operator controls, and the buyer controls the operator, the earnout is effectively at the buyer's discretion unless the contract says otherwise.
The fix is a list of operational covenants. The buyer cannot terminate the seller-CEO without cause. The buyer cannot relocate the headquarters. The buyer cannot impose a corporate-level cost allocation that depresses the metric. The buyer cannot move the customer relationships into another entity. The buyer cannot wind down a product line that drives the earnout.
These are not friendly provisions. They constrain the buyer's flexibility for the earnout window. They are also the only provisions that make the earnout enforceable. A buyer who refuses to agree to any operational covenants is telling the seller that the earnout is a fiction.
Specify how to dispute
When the earnout produces a number, somebody is going to disagree. Draft the dispute mechanic before it matters.
A clean dispute mechanic has three layers. First, the buyer delivers the computation. Second, the seller has a defined window to object, and a mechanism for accessing the underlying data. Third, an independent accountant resolves any remaining dispute within a defined window and at a defined cost split.
The mechanic should be self-executing. If the buyer fails to deliver the computation, the seller's number wins. If the seller fails to object, the buyer's number wins. The accountant's decision is binding. Nothing else.
The reason for all of this is that the alternative is litigation, and litigation over an earnout takes longer than the earnout window itself. The mechanic exists to avoid the courthouse.
What survives
The earnouts that pay what they should share three properties. The metric is readable. The window is long enough. The operational covenants protect the seller from the buyer's discretion. The dispute mechanic is short and self-executing.
The earnouts that do not share those properties are not earnouts. They are conditional gifts.
Walter Allison is a corporate attorney in Denver. He writes here about M&A, private equity, and venture capital structure.
Follow on LinkedIn ·
Firm bio ·
More writing