Every accelerator, lawyer, and YC essay tells you to sign a cofounder agreement on day one. It’s good advice in principle. In practice, the documents most cofounders actually sign are templates filled out hastily, never updated, and structurally incapable of resolving the conflicts that actually break startups apart. They don’t fail because they’re missing a clause. They fail because they assume disputes are about contract terms, when they’re almost always about something else.
Templates can’t anticipate the real conflicts
Standard cofounder templates cover equity splits, vesting, IP assignment, and decision rights. They rarely cover the situations that actually rupture partnerships: one cofounder underperforming and the other wanting them out, fundamental disagreements about company direction after product-market fit, divorce-related stock claims, mental health crises, or a cofounder who technically meets their commitments but has lost belief in the project. These are messy human situations that don’t reduce to clauses. Even when the agreement names a process — board vote, mediation, buyout formula — invoking it usually destroys the working relationship before any “resolution” is reached.
Vesting clauses help, but not the way founders think
Four-year vesting with a one-year cliff is the standard, and it’s better than nothing. It at least prevents a cofounder who quits in month six from walking away with 25% of the company. But vesting doesn’t solve the harder cases: the cofounder who stays exactly long enough to vest substantial equity while contributing little, or the cofounder you can’t legally fire because they’re a director. Acceleration clauses, good-leaver / bad-leaver definitions, and repurchase rights matter more than the basic vesting schedule, and templates often handle them poorly. By the time you discover the gap, it’s too late to renegotiate.
The agreement matters less than the relationship
Functional cofounder partnerships have something most agreements can’t manufacture: a habit of direct conversation about uncomfortable topics. Founders who can talk openly about performance, money, life changes, and disagreement rarely need to invoke their agreement. Founders who can’t will discover that no document forces a productive conversation. The agreement is useful as a forcing function — drafting it surfaces assumptions early — but the document itself is closer to a prenup than a manual. It defines fallback positions for when things go badly, not a path back to working well together.
What to actually do
Sign a real agreement, drafted by a startup lawyer, with thoughtful vesting and clear IP assignment. Then schedule recurring honest check-ins between founders — quarterly is reasonable — covering performance, satisfaction, and life circumstances. Treat the agreement as insurance, not as a substitute for ongoing alignment. And accept that if a cofounder relationship goes sideways, the legal document will determine the messy outcome but won’t prevent the mess.
The bottom line
Cofounder agreements get oversold as the safeguard against partnership disasters. They’re closer to a backstop than a shield. The actual work is in the relationship, the candor, and the early warning signs the document can’t detect.
Leave a Reply