Starting a business with a friend: why partnerships break — and the honest case for going solo

It almost never starts as a business decision. It starts as two friends, a good idea, and a moment of optimism — and then someone says “let’s just split it 50/50” and they shake hands. Robin Chase, who co-founded Zipcar, later called her own version of that moment “a really stupid handshake… that first handshake caused a huge amount of angst over the next year and a half.” She is not unusual. In Noam Wasserman’s research, 73% of founding teams split the equity within the first month — at the exact point they know least about who will actually carry the weight (the equity-split research behind The Founder’s Dilemmas).

The friendship-killing decision is often made before there is even a business. This is a piece about what happens next — why so many business partnerships fracture, what the data actually says, and why a striking number of people come out the other side as a company of one.

It’s not the market that kills you — it’s the conversation you avoid

Ask why startups die and the recorded cause is usually “ran out of cash.” But dig into the root and you find people. Wasserman’s data on roughly 10,000 founders found that about 65% of startup failures stem from “people problems” rather than product or market — with co-founder conflict the dominant strand. (Be precise: “people problems” is the category; it’s routinely mis-quoted as a clean “65% fail from co-founder conflict.”) Tom Eisenmann’s later HBS study, Why Startups Fail, names one of its six fatal patterns “Good Idea, Bad Bedfellows” — including co-founders who can’t decide who’s in charge and slow the company to death by deciding everything jointly.

And here’s the cruel twist for friends specifically: closeness makes it worse. Wasserman found businesses run by friends, family and couples failed most often — because the very bond that made them start together made them avoid the hard truths that would have saved them. The couples therapist Esther Perel, writing for First Round, puts it perfectly: “Conflict isn’t the problem. Avoiding conflict is.” Healthy partnerships move from rupture to repair; doomed ones just quietly accumulate the rupture, each friend thinking “I can’t bring this up — we’re mates.”

The pattern is always some mix of the same ingredients: a static equity split that stops matching who’s actually doing the work; one person treating it as a side-hustle while the other wants to build an empire; roles that were never defined because defining them felt like distrust; and money or control disputes that a frank early agreement would have pre-empted. None of it is exotic. All of it is avoidable. Most of it isn’t avoided.

Going solo isn’t the consolation prize — the data says it might be the better business

Here’s the part that reframes the breakup. The instinct after a partnership sours is to treat solo as second-best — what’s left when the dream team fell apart. The evidence disagrees.

In “Sole Survivors”, Jason Greenberg (NYU Stern) and Ethan Mollick (Wharton) followed Kickstarter-funded ventures and found that solo-founded companies tend to survive longer and generate more revenue than founder pairs — two-person teams fared worst of all. Their likely explanation is exactly the tax described above: solos don’t dissolve over co-founder conflict, because there’s no co-founder to fall out with.

This is genuinely contested, and honesty demands the other side. Paul Graham of Y Combinator calls the single founder his number-one startup mistake: “the low points in a startup are so low that few could bear them alone… you need colleagues to brainstorm with, to talk you out of stupid decisions, and to cheer you up when things go wrong.” And investors plainly prefer teams — they raise more money, almost regardless of outcome. So the scoreboard reads: investors bet on teams; survival data favours solos; friend-and-family teams are the shakiest of all. Going solo isn’t just emotionally cleaner after a fallout — there’s real empirical cover for it being the more durable structure.

What you actually trade when you go solo

Don’t romanticise it. Going solo resolves the relationship problem and, per the data, often the business problem too — but it hands you a permanent trade. Everything Paul Graham listed as the reason to have a co-founder becomes your standing cost: no one to share the emotional load, no complementary skillset, no one to talk you out of the bad call at 2am. The solopreneur’s well-documented burdens — isolation and wearing every hat — are not a separate problem from the autonomy you wanted. They are the autonomy. By various surveys a large share of solopreneurs report real loneliness, and that’s the honest price of owning all the upside and all the decisions. It’s worth going in clear-eyed: the loneliness of solopreneurship is the recurring cost on the other side of “it’s all mine.”

That’s the whole trade in one line: a partnership bets that shared load beats sole control; solo is the opposite bet. When the friction wins, the opposite bet starts looking less like defeat and more like fit.

If you do partner — or partner again — build the boring scaffolding

The protective structure feels cold between friends, which is precisely why friends skip it and precisely why you shouldn’t:

• Vest the equity. The standard is four years with a one-year cliff — nothing vests for the first year, then it accrues monthly (Carta explains it well). A departing co-founder keeps only what they earned, so a fallout doesn’t leave dead equity on the cap table.
• Write the founders’ agreement — roles, ownership, decision rights, capital, and a pre-agreed exit/buyout. The most common omission is never discussing how one of you leaves.
• Treat equity as dynamic, not a handshake. Contributions diverge; the split should be able to.
• Have the hard conversations early and on a schedule — “renew the vows.” Avoidance is the failure mode, so build the talk into the calendar.

And if you’re the one rebuilding alone after it ended, the move is the same one everyone making a fresh start uses: work out the maths of a solo business (how few customers you actually need), then follow the step-by-step setup for a one-person business in Europe — and pick the legal structure for one owner, not two, in sole trader vs OÜ vs freelance.

The clean, unsentimental takeaway: a business with a friend can absolutely work — but only if you’re willing to be a little less of a friend and a little more of a partner about it. And if it doesn’t work, going solo isn’t the wreckage. For a lot of people, it’s the version that was always going to suit them better — whichever kind of one-person business you are.