The 50/50 Deadlock: What to Do When Two Co-Founders Can't Agree

This article is part of our guide to co-founder conflict.

You split it down the middle because anything else would have required a conversation neither of you wanted to have in week one. Equal partners, equal risk, equal reward. It was the decision that let you start.

It is also the reason that, three years later, a disagreement about pricing has been open since March.

Short answer: is a 50/50 co-founder split a mistake?

No. Co-founder deadlock is what happens when two founders with equal voting power disagree and have no agreed way to break the tie, so the disagreement waits until one of them gives in. The split itself is not the problem; a split with no tiebreaker is.

Two founders holding equal equity have equal voting power, which means that on any question where you genuinely disagree, neither of you can carry the decision. There is no vote to win. The disagreement does not resolve — it waits.

Opinion in the ecosystem is split on the split itself. Y Combinator's partners have long favoured equal founder equity, on the grounds that it signals commitment and removes a standing grievance. Startup lawyers tend to flag a pure 50/50 as a structural flaw and suggest a casting vote, a neutral third director, or a 51/49 split. Both camps agree on the thing that matters: a company that cannot make a decision under disagreement is a company that cannot move at the speed its market requires. The concern is not that you will fight. It is that you will stall.

What matters is not the number. It is whether you have agreed, in advance and in calm, on what happens when the two of you do not agree.

What deadlock actually looks like day to day

Deadlock rarely announces itself. Founding teams in deadlock usually describe their problem as something else entirely:

• Decisions that get reopened. Something was settled in April. It is back on the agenda in June, because it was never actually agreed — one of you stopped arguing.
• Slow strategy, fast execution. You ship well. You cannot decide what to ship. Operational questions move; directional ones sit.
• Proxy arguments. A disproportionate fight about a hire or a vendor. The intensity does not match the stakes because the real disagreement is the one underneath, and it has nowhere to go.
• Decisions made unilaterally, then defended. One founder stops asking, because asking produces stalemate. The other experiences this as a breach of the partnership, which it is — but the alternative was paralysis.
• A growing list of unmentionables. Subjects that reliably produce stalemate get quietly removed from the agenda. The list only grows, because nothing ever comes off it.

If more than one of those is familiar, you do not have a communication problem. You have a decision-rights problem wearing a communication problem's clothes.

The stamina problem: how deadlock picks a winner

Here is the part most founders miss. The Stamina Problem: deadlock does not actually stop decisions. It outsources them to whoever can tolerate more discomfort, which means the same founder concedes every time and starts keeping a ledger.

When there is no mechanism, the disagreement continues until one person concedes. Over enough iterations, the person who concedes is reliably the same one — the one who finds unresolved tension harder to live with, or who is more invested in the relationship, or who simply has less appetite for the argument.

This produces two things, both corrosive.

The founder who concedes starts keeping a ledger. Not deliberately. But a running private tally accumulates: I gave way on the pricing model, the first big hire, the raise timing. That ledger is invisible to the other founder, who experiences the same history as a series of reasonable discussions that went the right way.

And the founder who prevails learns, without ever intending to, that persistence works. They are not being manipulative. The mechanism rewarded a behaviour and the behaviour increased.

Then one day the ledger comes out, usually during an argument about something minor, and it arrives as a list going back years. The other founder is genuinely blindsided — and both of their accounts of the last three years are accurate.

This is what makes 50/50 deadlock dangerous. It does not feel like a structural flaw while it is happening. It feels like a character flaw in the other person.

Four ways to break a deadlock

The four deadlock-breakers are domain ownership, a named tiebreaker, the reversibility test, and structural provisions in your agreement — in that order of usefulness. None of these require changing your cap table.

1. Domain ownership

The strongest and most under-used fix. Split decisions, not percentages.

Each founder owns a set of domains outright — product, go-to-market, hiring, finance, engineering. The owner decides. The other founder gets full input, real argument, and no veto. The question is not "do we agree?" but "whose call is this?"

This feels uncomfortable to propose because it sounds like a loss of power for both of you. In practice it is the opposite: two founders who each truly own half the decisions argue far less than two who jointly own all of them, and both get to move faster in their own half.

The test of whether you have real domain ownership: can the owner make a call the other founder disagrees with, and have it stick without a negotiation?

2. A named tiebreaker

One person, agreed in advance, who settles anything the two of you deadlock on. An experienced operator you both respect, a board member you both trust, an advisor with no financial interest in the outcome.

Two rules make this work. Agree on the person while you are calm — choosing a referee mid-dispute is itself a dispute. And use them rarely; a tiebreaker invoked monthly is not a tiebreaker, it is a third founder.

3. The reversibility test

Borrowed from decision theory and badly underused by founding teams. Before arguing, classify the decision:

Reversible? Whoever cares more decides, now. Cheap to undo, expensive to debate.

Irreversible? It gets the full process — written positions, a slept-on night, the tiebreaker if needed.

Most founder deadlock is spent on reversible decisions being treated as permanent ones. Sorting them costs nothing and removes a surprising fraction of the friction.

4. Structural provisions

The legal layer: deadlock resolution clauses in your operating agreement or shareholders' agreement, buy-sell mechanics, a mediation step before escalation. A good startup lawyer can draft these in an afternoon.

Worth having. Worth understanding what they are for: these are the fire exits. They settle what happens if the partnership fails. They do not help the two of you decide what to do about pricing on Thursday, and a founding team that reaches for the legal layer to solve an operating problem has usually skipped the three mechanisms above. We compare what a lawyer, a mediator and a co-founder coach each fix here.

The conversation that has to happen first

Any of those four will fail if the two of you install it without first saying the true thing.

The true thing is usually some version of: there is a ledger, and here is what is on it.

That conversation is difficult precisely because both founders' accounts are accurate. The one keeping the ledger genuinely conceded repeatedly. The other genuinely did not notice, because from where they sat, those were just conversations that concluded.

Neither of you is lying. You are describing the same events from inside two different experiences of them, and the only way through is for both accounts to be in the room at once — which is hard to do in a relationship where every exchange is also a negotiation.

This is the specific situation where having someone outside the two of you present is worth more than any amount of goodwill. Not to adjudicate. To make it possible for both versions of the last three years to exist in the room without one of them having to be wrong.

If you are pre-launch: splitting equity without building this in

If you have not incorporated yet, you have an advantage worth using.

Decide decision rights at the same time as equity. They are separate questions and conflating them is the original error. You can hold equal equity and still have unequal, clearly-owned decision domains. Equity is about reward for risk. Decision rights are about operating speed. There is no reason they have to match.

Vest everything, including the founders. Standard four-year vesting with a one-year cliff, both of you, no exceptions. Not because you distrust each other — because the alternative is a company where someone can leave in month eight holding half of it.

Write down what you are building toward. Numbers and years. The commonest ambition divergence is not discovered until a term sheet arrives, and by then it is a crisis rather than a conversation.

Agree how you will break ties, before you have one. Five minutes of work in week one. Years of friction avoided. The full list of questions to settle before they settle you is here.

FAQ: 50/50 splits and deadlock

Is a 50/50 equity split a bad idea for co-founders?

Not in itself. Equal splits are common and many investors, including Y Combinator's partners, prefer them because they signal equal commitment. The problem is an equal split with no agreed way to break a tie. Decide decision rights and a tiebreaker at the same time as equity, and a 50/50 split works fine.

Should we change our equity split to break a deadlock?

Usually not. Renegotiating equity to fix a decision problem is an expensive answer to the wrong question, and it tends to make the contribution ledger explicit and permanent. Fix decision rights first. If what you actually have is a contribution imbalance rather than a decision problem, that is a different conversation.

Do investors refuse to fund 50/50 teams?

Refuse, rarely. Ask about it, routinely. What they are testing is whether you have a mechanism. A founding team that says "we're 50/50, and here's how we break ties" answers the concern completely.

How do you break a deadlock between two 50/50 partners?

Four mechanisms, none of which touch the cap table: give each founder sole ownership of defined decision domains; name a tiebreaker in advance; sort decisions by reversibility so only the irreversible ones get the full process; and put deadlock provisions in your shareholders' or operating agreement as a backstop. Then have the conversation about why you had no mechanism, because the mechanism alone holds for about a quarter.

What if my co-founder will not agree to any tiebreaker?

That refusal is information. Someone unwilling to be overruled under any mechanism is telling you what the partnership actually is. Raise it directly, as a pattern rather than an accusation, and if it does not move, you are no longer discussing deadlock — you are discussing whether this partnership works.

Is 51/49 better than 50/50?

Mechanically, yes: it breaks ties. Relationally, it can create the same resentment it was meant to prevent, because the minority founder learns they can always be overruled. Clear domain ownership usually achieves more than a one-point difference, with less cost to the relationship.

We are deadlocked on something right now. What do we do today?

Classify it. If it is reversible, let whoever cares more decide and move. If it is irreversible, write your positions down separately before discussing them — written positions are harder to escalate and much harder to misremember later — then bring in your tiebreaker if you have one. Then, separately and not in the same week, have the conversation about why you had no mechanism.

Next step

Deadlock is a mechanism problem with a relationship problem underneath it. Fixing the mechanism without touching the ledger tends to hold for about a quarter.

Aether Practice offers co-founder sessions for founding teams in San Francisco, New York City and Norway — both of you in the room, online, working on how you decide, disagree and divide the weight.

Keep reading in this cluster: The complete guide to co-founder conflict · When one co-founder stops pulling their weight · Renegotiating co-founder equity · Co-founder coaching in San Francisco and New York

Research and further reading

• Hellmann, T. & Wasserman, N. (2017). The first deal: The division of founder equity in new ventures. Management Science, 63(8), 2647–2666. In a sample of 511 ventures, roughly a third of founding teams split equally, and teams whose members differed in idea generation, prior founding experience or capital contributed were less likely to do so.
• Seibel, M. (2015). How to Split Equity Among Co-Founders. Y Combinator. The case for equal or near-equal splits: the value lies in the years ahead, and a lopsided split costs motivation.
• Techstars. How to Split Co-Founder Equity (The Right Way). Techstars blog. The accelerator argument against a reflexive 50/50, and for vesting every founder.
• Breugst, N., Patzelt, H. & Rathgeber, P. (2015). How should we divide the pie? Equity distribution and its impact on entrepreneurial teams. Journal of Business Venturing, 30(1), 66–94. Eight founding teams followed for six months: whether the split was perceived as fair, not its size, set off the positive or negative spiral in how the team worked together.
• Wasserman, N. (2008). The founder's dilemma. Harvard Business Review, 86(2), 102–109. The trade-off between keeping control and building value that sits underneath most arguments about who gets the casting vote.
• Also read: Renegotiating co-founder equity · The co-founder agreement questions · Co-founder coaching

Last updated: October 2026.