Negotiation.

The framework

1. 01

   Most of the time, nobody is bluffing.

   Most stuck deals are one of two things, and the repairs don't overlap. A value disagreement is about what the thing is worth now: different needs, different alternatives, different urgency. A forecast disagreement is about a fact neither of you has yet — the volume, the approval, the churn rate in year two. Usually both sides can show their reasoning, nobody is bluffing, and more argument moves nobody, because nothing is knowable yet. So ask what would have to happen for you to change your mind. If the answer is an event rather than a document, you're not going to win this by talking. And where you do suspect the forecast is theatre rather than belief, the same table is the test: offer to price their number and watch whether they will sign to it.

2. 02

   Price both worlds before you price the deal.

   The instrument is a payoff table, not a paragraph. Three rows: what they pay in their forecast, what they pay in yours, how it moves in between — bounded by a floor and a cap, hung on one metric with a date it gets read. Start from the base, what they would pay in their own world — the floor you have both agreed. Everything above it is the argument you hand to the evidence. Cap the top; uncapped upside is how a good instrument becomes a grievance. The familiar forms are this table in legal clothing: earn-outs, milestone payments, royalty steps. Fill the rows; let the lawyers name it.

3. 03

   Four conditions, checked before anyone drafts.

   Four tests decide whether a contingent term is worth writing, and each one quietly kills a deal that fails it. Enforceable: you can still collect in three years, from a party that still exists, in the jurisdiction that would hear it. Measurable: the payoff row has one reading, off one record, named before signing. No perverse incentive: nobody is paid for the outcome the deal exists to prevent — a trigger on a cost line rewards the cancelled project as readily as the efficient one. Survivable: if nothing above the floor ever arrives, this deal still has to work for you. Run all four as a gate, not a review.

4. 04

   Nobody has to lie for the number to disappear.

   The payer usually produces the number, and that alone is not the flaw. The flaw is discretion: whether they can move it by a decision nobody would ask them to explain. An earn-out on net profit offers plenty. Head-office costs land on the acquired unit, a launch slips a quarter, and the profit meant to pay you never appears — nobody having lied, nothing breaching a term you wrote. A trigger the paying side can steer doesn't weaken this instrument, it empties it. So hang the money higher up their accounts, above where the discretion lives — units shipped, gross revenue. The question is not who writes the number, but what they can do to it.