The Co-Founder Conversation You Keep Postponing: How to Practice It Before You Have It

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

Short answer: how do you practice a hard conversation with your co-founder?

Write the one sentence you want to say, as a question about the partnership rather than a verdict on the person. Write down the strongest thing they will say back, which is usually a counter-charge or a retreat to process. Say the opening out loud until it sounds like you and not like a board memo. Then practice the exchange against a counterpart who plays your co-founder at their most defensive, find the sentence where you either sharpened or backed off, and replace it. Two minutes at a time, a few times, then have it, ideally with someone in the room whose only interest is the partnership.

The aim is not to win against your co-founder. It is to say the true thing without saying it as a prosecution, and to still be saying it when they push back.

Why founders practice the wrong version for months

Founders are good at preparing. That is the problem. The equity conversation gets a document. The performance conversation gets a list with dates. The exit conversation gets a decision tree. And every one of those artefacts is built alone, from one side's memory, in a version of the conversation where the other founder does not interrupt.

By the time it is said, it is a case. Your co-founder, who has also been preparing alone, hears the opening and recognises a case, and responds with their own. Two prepared positions, no plan for what happens when the other person does not accept one. That is how a conversation that should have been "I want to look at this with you" becomes the end of the company.

Practice is the correction. Not more preparation; the opposite. Saying the opening out loud, to something that answers back the way your co-founder will, and discovering in the first thirty seconds that the sentence you have been polishing lands as an accusation.

A note before the three conversations. Decide first whether what you want is a conversation or a decision. Practice makes you better at holding a position. If your position is "you have to go", the method below will make you better at a conversation that is really a notice, and the separation sequence is the right page, not this one.

The equity conversation

The sentence. The version that lands is a question about a forecast, not a claim about a ledger. "The split we agreed assumed something about how the work would divide. It hasn't divided that way for a year, and I want us to look at it together." Not "I do most of the work." Not "the split is unfair." Those are verdicts, and the forecast frame exists because verdicts about equity end partnerships.

The real objection. The weak one is "it's not that unequal." The real one is: "We agreed. Now that it's worth something, you want to change the deal." It is powerful because it is half true, and because it recasts you as opportunistic. Practice against that one.

Where founders fold. On "we agreed." The most common response is "you're right, forget it", which keeps the resentment and loses the conversation. The second most common is to produce the list, hours, commits, deals closed, which turns a conversation about a forecast into a trial about a ledger, and the trial has no judge.

The line that holds. "We did agree, and the agreement rested on an assumption that turned out wrong. I'm not asking you to concede anything today. I'm asking us to look at it with that in mind, and I'd rather do it now than carry it for another year." Acknowledge, restate, and name the alternative, which is carrying it. Then stop talking.

Practice the second objection. It is usually "so what do you want the split to be?" Do not answer with a number. The number is the end of the conversation, not the opening of it. "I don't have one. I want us to agree how we'd think about it before either of us says a number."

The performance conversation

The sentence. Not "you're not pulling your weight." A description of what is not happening, with no diagnosis: "Three things I expected to be done by now aren't, and I don't know why, and I want to." The whole value of the sentence is the "I don't know why," because there are four different things that look identical from outside, burnout, mismatch, disengagement and a quiet departure already under way, and they need opposite responses.

The real objection. "You have no idea what I'm carrying." Sometimes said; usually meant. The counter-charge is the reflex, and it is often partly right. The second-order objection is a retreat to process: "let's put it in the Monday meeting." Both are attempts to move the conversation out of the room.

Where founders flare. On the counter-charge. "I know exactly what you're carrying, I'm carrying it too" is the sentence where the performance conversation becomes a fight about whose life is harder, and nobody has ever won that one. The founder who raised it gets sharp, the other goes defensive, and the three specific things are never discussed.

The line that holds. "You might be right that I don't see all of it. I'd like to. I'm still asking about these three things, because I need to understand what's going on with them." It accepts the possibility of the charge without conceding the request, and it keeps the question a question.

Practice the silence. This conversation often goes quiet at the point where the real answer is "I'm not sure I want to do this any more." If you fill that silence with reassurance, "it's fine, we'll sort it", you will not hear it. Practice saying the sentence and then saying nothing. Count to five.

The conversation about whether you are still in

The sentence. This one is different, because you may be the one with the doubt. Two openings, depending on which side you are on. If you suspect your co-founder has checked out: "I think something has changed for you about this, and I'd rather hear it from you than guess." If it is you: "I'm not sure I want to keep doing this, and I'm telling you before I've decided, because you should be part of how I decide."

The real objection. From the other side of the first opening: "Where is this coming from?", which is a request for your evidence, and the trap is to give it, because a list of signs reads as surveillance. From the other side of the second: "So you're leaving.", which jumps to the end and makes the undecided thing decided.

Where founders fold. On "so you're leaving." The founder who was not sure says "no, no, I'm not going anywhere", because the alternative is unbearable in the moment, and the honest conversation is over before it started. The two terminal conditions are worth knowing before this one, because the difference between a finished partnership and a long bad stretch feels identical from inside.

The line that holds. "I don't know. That's why I'm saying it now rather than after I've decided. I want to work out with you whether this is fixable, and I can't do that if I pretend I'm certain either way." Hold the uncertainty. Uncertainty, said plainly, is the most honest position in this conversation, and the one most people abandon within a sentence.

The two ways founders lose the conversation

Every piece of advice about founder conflict optimises for one failure: escalation. Stay calm, assume good intent, separate the person from the problem. It is good advice, and it is half the picture.

The other half is the fold. The equity question withdrawn on "we agreed." The performance question dropped on "you have no idea what I'm carrying." The doubt retracted on "so you're leaving." Each of those is a conversation ended by one founder trading the thing they came to say for the end of the discomfort, and each of them keeps the problem. Founders who never raise their voice can lose every one of these conversations this way for years and call it maturity.

AET01 Conflict, the co-founder conversation simulator, scores both failures. Escalating loses points. So does giving in to make it stop. Clear, respectful and immovable scores highest, which is the register these three conversations need and the one founders practice least, because it is the one that feels, from inside, like the discomfort is a mistake. It is not. Discomfort is not evidence of error.

AET01 Conflict

Fix the fight before you have it.

Practice it against an AI partner who doesn't back down. Two minutes, a score, and the line that would have held.

Open AET01 ↗Coming to iOS and Android

Practice, then the session

The seven steps apply. For co-founders, three adjustments.

Practice against the reasonable voice. Your co-founder will not shout. They will stay calm, concede nothing, and move the conversation to a document. The counterpart should do that, because the reasonable voice is harder to hold a line against than anger, and founders are unprepared for it.

Practice without the evidence. Leave the list at home, in practice and in the real conversation. The moment you produce it, the conversation becomes a trial. Practice holding the question with nothing but the question.

Stop at two minutes. The opening is the part you control. Practice it until it sounds like a question, three or four runs across a week, then have the conversation. The real one will be longer, and it will go somewhere you did not practice, which is fine; the register will have been set.

Then the session. Practice changes how you hold a position; it cannot put the other founder in the room. Co-founder coaching is the two of you, online, with someone whose only interest is the working partnership, in a room with no cap table in it. Most founding teams come for exactly one of the three conversations above, and most find that saying it with a third person present makes the thing sayable that has been unsayable for a year. Aether works with founding teams in San Francisco and New York City, both founders present, with a free 15-minute consultation first and no minimum engagement.

FAQ: practicing the co-founder conversation

How do I bring up equity with my co-founder?

As a question about a forecast, not a claim about a ledger. "The split we agreed assumed something about how the work would divide. It hasn't divided that way, and I want us to look at it together." Practice it against the real objection, "we agreed, and now you want to change the deal", and have a line ready that acknowledges the agreement and restates the request. Do not open with a number, and do not bring the list.

How do I tell my co-founder they are not pulling their weight?

Do not tell them; ask. Describe three specific things that are not happening and say you do not know why. There are four different problems that look identical from outside, burnout, mismatch, disengagement and a quiet departure, and they need different responses, so the first conversation is for finding out which it is. Practice holding the question when the answer is a counter-charge about what they are carrying.

How do I practice a difficult conversation with my co-founder?

Write the one sentence you want to say, as a question about the partnership. Write the strongest objection they will make. Say the opening out loud until it sounds like you and not like a board memo. Practice the exchange against a counterpart who plays your co-founder at their calmest and most immovable, find the sentence where you sharpened or backed off, and replace it. Two minutes at a time, a few times, then have it.

What if my co-founder goes quiet or moves it to a meeting?

That is the real objection for most founder conversations: a retreat to process. Hold the conversation in the room. "I'd rather not put it in Monday's meeting. I want to talk about it with you, now, without an agenda." Practice that line specifically, because the pull toward the document is strong for both of you and it is where the conversation usually dies.

Should I practice with an AI before talking to my co-founder?

For the opening, yes, and against something built to hold a position rather than agree with you. AET01 Conflict plays the other side at a level you choose, scores the exchange against a fixed standard, and names the sentence where you escalated or gave in. It is a practice tool, not a coach and not therapy, and it will not put your co-founder in the room; co-founder coaching does that.

Should we do this conversation ourselves or with a coach?

Practice yourself; have the conversation with the two of you present, and if it has been postponed for more than a quarter, with a third person in the room. The conversations founders postpone longest are the ones where both sides have prepared a case, and a neutral person whose only interest is the partnership is what makes it possible to compare the two versions without one of them having to be wrong.

Next step

Pick the conversation you have been postponing. Write the sentence as a question. Then rehearse it in AET01 Conflict, which will play your co-founder at their most reasonable and immovable, and find out where you fold.

Then book the room. Aether Practice offers co-founder sessions for founding teams in San Francisco, New York City, Stockholm, London and Norway: both of you, online, sixty minutes, with a free 15-minute consultation first.

Keep reading in this cluster: The complete guide to co-founder conflict · Renegotiating co-founder equity · When one co-founder stops pulling their weight · Co-founder breakup: separating without killing the company · How to practice a difficult conversation: the seven steps · Why general AI fails at practicing difficult conversations

Last updated: November 2026.