When two founders part ways, there is a story they tell the team, and a story that is actually true, and they are rarely the same story. The official version is almost always about vision. We wanted different things. We grew apart on strategy. We had philosophical differences about the direction of the company. It is dignified, vague, and lets everyone save face. In my experience working with founders, it is usually a cover for something far more ordinary and far more heartbreaking: two people who fit together perfectly at 10 employees never stopped to renegotiate who they were to each other at a 100 employees. The company changed underneath them, their roles changed underneath them, and the relationship that governed it all was never once renegotiated. It just quietly went out of date, like a contract nobody remembered to renew, until the day the resentment was too loud to ignore.
I want to make the case that the co-founder relationship is the single most under-maintained asset in most growing companies, and that the fix is neither dramatic nor expensive. It is a conversation. You just have to have it on purpose, and you have to have it more than once.
Why the founding deal expires
Think about what the original co-founder arrangement typically is. Two people, usually with complementary strengths, divided up a company that barely existed yet. One takes operations, one takes sales. One became the inside person, one became the outside person. Often it was never even spoken aloud; it just emerged from who grabbed which problem first in a company small enough that you could grab problems by hand.
That arrangement was perfectly designed for the company you had when you made it. The trouble is that a company of 10 people and a company of a 100 are not the same company. They are different organisms that happen to share a name. At 10 people, the founder who “does sales” is on the phone closing deals personally. At 100, that same function is a VP of sales, a team, a pipeline, a compensation philosophy, and the founder’s real job is to hire the person who does what the founder used to do and then get out of their way. That is not a bigger version of the old job. It is a different job that requires the founder to give up the thing they were good at, the thing that made them feel useful, the thing that was, in a real sense, their identity in the partnership.
Here is where the fracture starts, and it rarely starts as a disagreement. It starts as a feeling. One founder feels the other is holding on too tight, still meddling in a function that has outgrown them. Or one founder feels quietly obsolete, watching the thing they built get handed to someone else, unsure what their role is anymore. Neither of them says this out loud, because they are afraid to bring it up. So, it goes underground and ferments, and 18 months later it surfaces as “different visions.”
The founders who kept renewing the deal
Consider the counterexample, because it proves the thing can be done. Patrick and John Collison started Stripe in 2010 and built it into one of the most valuable private companies in the world, with Patrick as CEO and John as president. This partnership has held for well over a decade at a scale that breaks most co-founder relationships. What is striking is not that they avoided change. It is that they kept choosing it.
In the early years, the division was concrete and hands-on: John handled partnerships and sales, Patrick focused on engineering and being the public face. But as Stripe grew, both of them systematically handed off the very responsibilities that had defined them. Fundraising, once a core Collison duty, got handed to their CFO. Board preparation was handed off to someone else, too. The company’s biggest partnerships and international expansion were handed to a chief business officer. Patrick put it in almost startling terms a few years in: on many of the most important things Stripe does, he said, he and John might be twentieth or fiftieth on the list of people driving the work, or absent from it entirely. This is a founder describing, without apparent grief, the deliberate act of designing himself out of his own former jobs.
That is the move most founders cannot make, and it is worth being precise about why the Collisons could. It was not sibling telepathy, and it was not that they never had to change. It is that they treated the handoff as the normal condition of scaling rather than as a loss to be mourned or resisted. Each stage of growth, they renegotiated what they were for. The relationship stayed current because they kept updating it, on purpose, before the outdated version could curdle into resentment. They re-contracted so continuously that there was never a single expired deal sitting around long enough to poison anything.
How to re-contract before it breaks
You do not need to be siblings, and you do not need to be preternaturally calm to do this. You need to treat the co-founder relationship the way you treat any other critical system in the company: something that gets reviewed on a schedule, not something you touch only when it is already on fire. Here is what that looks like in practice.
Put it on the calendar, and tie it to scale, not to time. The trigger for a re-contracting conversation is not “once a year.” It is every time the company roughly doubles, because doubling is what silently rewrites everyone’s job. At 20 people, at 50, at 100, at 200, the founders sit down specifically to ask: what is each of us actually responsible for now, and is that still what we agreed to, and is it still what the company needs from us? The point is to have the conversation while it is still theoretical, before either person is already hurting.
Name the loss out loud, because the loss is the real content. The hardest part of scaling for a founder is not learning new skills. It is grieving the old identity, the version of you that personally closed the deals or shipped the code. When a founder resists letting go, they are rarely being irrational about the business. They are protecting a self. A good re-contracting conversation makes room for that explicitly: what are you giving up at this stage, what did that thing mean to you, and what are you moving toward that could mean as much? Resentment is grief that was never allowed to be spoken. Speak it, and it loses most of its power.
Separate the two questions that always get tangled. When co-founder tension surfaces, it almost always arrives as a strategy argument: we disagree about the roadmap, the market, the raise. But underneath a strategy argument there is frequently a role argument in disguise: this is really about who gets to decide, about whose territory this is, about whether I still matter here. If you try to resolve the surface disagreement without naming the role question underneath, you will win the argument and lose the relationship. Ask directly: is this a disagreement about the decision, or about who owns the decision? They require completely different conversations.
Write down the new deal. The original founder arrangement was usually implicit, and implicit is exactly what let it expire without anyone noticing. When you re-contract, make it explicit enough to point at later: here is what each of us owns now, here is what we have each let go of, here is how we will make the calls that fall in between. It does not need to be a legal document. It needs to be a shared, spoken agreement that both people could repeat back, so that six months from now, when the old instinct creeps back in, either of you can name it without it becoming an accusation.
The relationship is the company
Founders spend enormous energy maintaining everything except the one relationship the whole enterprise rests on. They renegotiate vendor contracts, revisit strategy quarterly, rebuild org charts, and update investor decks, while the co-founder partnership runs on a handshake from a garage that no longer exists. Then one day it breaks, and everyone calls it a difference in vision, and a company that should have survived does not.
The two of you are not the same people who started this, and thank goodness, because the people who start a company are almost never equipped to run the one it becomes. That is not the threat to the partnership. Pretending otherwise is. The founders who make it to the far side of scale are not the ones who never changed. They are the ones who kept sitting down, again and again, to renew a deal they were honest enough to admit had expired. Do it on a schedule. Do it before the resentment. Do it while you still like each other. It is the cheapest insurance you will ever buy, and it is the one almost nobody remembers to renew.
If this resonated with you and you’re thinking about what it might look like to bring more of this into your work, I’d invite you to stay connected.
Join my newsletter
Or book time to connect.