The Cracks You Can't See Yet

I've been doing some match-making these last couple of weeks. Not the social kind - the startup kind. A founder I work with needed a co-founder. I knew someone who could fill it - complementary skills, the kind of temperament that holds up when things get hard, real experience in the space. I made the introduction. They hit it off. Now we're working through the fine print - equity split, roles, vesting, leaver terms, decisions that need both signatures, what happens if one of them wants out.

I know both of them well. I think this is going to work. And I still can't be sure.

That gap between "I think this works" and "I know this works" is where the co-founder problem lives. You can't test a partnership in advance. The things that make a co-founder relationship succeed or fail don't tend to show up until there's real pressure on the company, and by then you've already committed.

So we're spending time on the fine print now, while everyone is optimistic. Not because I'm worried - because it's easier to think about partnership when you're not testing it.

Two kinds of friction

Every co-founder team has friction. Rough weeks, arguments that get heated, stretches where one person is carrying more than the other. That's normal. A partnership that claims to be conflict-free either isn't telling you the whole truth or hasn't been tested yet.

The question isn't whether friction exists. It's what the friction is about, and where it's going.

Some friction stays on the work. You disagree about pricing, about hiring, about which customers to chase. The argument is sharp, sometimes uncomfortable, but it's about the decision in front of you. When it ends, you're clearer than when you started, and the decision is usually better because both of you pushed on it. You might walk out of the meeting annoyed, but you come back the next day and keep building together.

Other friction isn't really about the decision. It's about the person making it. They don't care enough. They're trying to control this. They don't understand the business. Once the conversation moves from "I think you're wrong about this" to "I think you're the wrong person for this," you've crossed into something else, and it doesn't tend to move back on its own.

A few patterns I've learned to watch for. Decisions the team thought were settled keep resurfacing, sometimes weeks later, with neither founder quite remembering why they were closed in the first place. One founder starts venting about the other to employees or advisors instead of raising things directly with their partner. Unilateral calls get made first and explained after, the partner informed rather than consulted. And the texture of the relationship shifts over time - arguing with each other slowly becoming avoiding each other.

I should be honest about the limits of this. You can't always tell in the moment. Any of these things can happen in a healthy partnership going through a bad stretch. What you can usually read is the direction. Is the friction getting clearer, or is it getting quieter?

Why it stays hidden

Reading direction sounds simpler than it is. Most writing on this topic assumes the problem is visible. The harder case is when it isn't - and the founders themselves are usually the last to see it clearly.

I've watched this from a few different seats - founder, investor, board member, advisor. The case that stays with me was from the investor side, of a company with three co-founders. On the surface, everything looked fine - clean board updates, aligned team, founders who presented well together. It was only when the business started to struggle that the cracks between them became visible to me, and when I talked to the team later, most of them had sensed it for much longer than I had. The three founders had been good at performing unity. They were less good at actually having it.

The company didn't make it. The product had problems, the market was hard. But the co-founder dynamic made every one of those problems harder to solve. Decisions got delayed because nobody wanted to be the one to push. The team picked up on the tension and started hedging their own bets. By the time the founders were willing to name what was happening between them, there wasn't much company left to save.

What I didn't fully appreciate at the time was how much the performance itself was costing them. Keeping up a unified front in front of employees, the board, and investors uses energy that should be going into the business. You're running the company, and you're running a second, invisible project to hide that the first one is cracking. That's the bind. The performance is exhausting, but the alternative - being the one to name it, sitting across from someone you've been through a lot with and saying "I don't think this is working" - is harder still. It's not so different from the end of a marriage; the question usually isn't whether both people know, but who's willing to say it first.

So founders wait. Months, sometimes years, hoping something external will resolve what neither of them is willing to bring up.

Why delay compounds

The cost of waiting isn't linear. It gets steeper the longer you sit with it.

Equity is the most mechanical part. At the very beginning, before vesting has kicked in and before the company has meaningful value, separating co-founders is a conversation and some paperwork. A year in, with a seed round closed and vested shares on the cap table, the same separation involves negotiation, sometimes a buyback, sometimes lawyers. Two years after that, with a Series A and employees holding options, you're dealing with the board, the investors, and a lot of moving pieces.

Narrative is the part founders underestimate. Investors and future hires see a co-founder change differently depending on when it happens. Early, it tends to read as a team finding its shape. Later, it reads as governance failure - something that was allowed to fester. I read it that way myself when I'm looking at a company. The cleaner the story, the less it costs in future rounds. And the story gets harder to tell the longer you wait, because the length of the delay raises its own questions about why.

The cost shows up in the team before it shows up anywhere else. The senior hire who leaves six months in for "chemistry reasons" isn't leaving because of chemistry with her manager. The board member who stops pushing for introductions isn't distracted by something else. Tension at the top leaks quietly, and you usually find out how much only in retrospect.

Acting while the signal is faint

The time to build a partnership that can survive trouble is while there isn't any - proper vesting, clear role definition, leaver terms that handle both the good-leaver and bad-leaver cases, a written understanding of who makes which decisions when the two of you can't agree. It's the startup equivalent of a pre-nup, drafted in honeymoon mode, and for the same reason - the right time to do it is when neither of you thinks you'll need it. The paperwork isn't a lack of faith; it's what keeps faith from being your only tool when things go sideways. Vesting and leaver terms are cheap to sort out when you don't need them, painful when you do.

When trouble does show up, the thing that seems to matter most is having the actual conversation. Directly, with the other person, not through a mutual advisor or a coach or a sympathetic investor.

In my mineral exploration days, my co-founders and I had what we called "working on our marriage" sessions built into the schedule - not as a response to trouble, but as a default. We'd realized that the relationship between us was going to make or break the company more reliably than any market or product question would. The sessions surfaced things we'd been avoiding and at one point made it clear that the founding team needed to change shape. That part wasn't easy. But it happened early enough to be a choice rather than a crisis.

What I took from those years isn't that every founding team needs scheduled sessions. It's that the work isn't a single hard sit-down when things go wrong - it's the practice of talking honestly often enough that nothing has to wait for a crisis to be named. It didn't always change outcomes. It changed whether the outcomes were something we shaped, or something that happened to us.

When direct conversation has stopped being productive, a third party can help - a board member with standing, an experienced advisor, sometimes a professional. The job isn't to take sides or deliver a verdict. It's to facilitate a conversation that has become too hard to have alone, and to keep both people focused on the actual question rather than the accumulated grievance. I do this occasionally for founders I work with. It's never pleasant work. It tends to be, in retrospect, the work that mattered most.

And sometimes, after all of that, the answer is separation. A clean separation handled early and framed honestly is almost always better for the company than a slow dysfunction everyone is pretending not to see. The founders who got this right tended to stop asking "can we make this work" and started asking "what does the company actually need." Those are different questions, and they often have different answers.

Back to the matchmaking

The fine print the two of them are working through isn't there because I think this partnership is fragile. I think it's solid. I'm pushing on it now because the version of them I want to see in two years isn't the one that learns to talk honestly under pressure - it's the one that already knows how.

That's the work I keep coming back to with founders. The paperwork matters, but the practice underneath it matters more: naming small things before they become large, and being willing to name them with the person it's hardest to name them with. The founders who handle this well aren't the ones who avoid trouble. They're the ones who built the habit before they needed it.

If you're at the start of a partnership, that's the work. If you're already in one and haven't built the habit, the next hard conversation is where you begin.

I help founders navigate strategy and funding decisions when the path isn't clear. If you're there, let's talk.