How to find a cofounder

Where do most cofounders meet?

In a Censuswide survey of 500 people for Starling Bank, published in April 2026, 60% said they met their cofounder by chance. The routes they named most often were friends and family (34%), a former workplace (26%) and school or university (25%). Community events, coworking spaces and networking events came in at 15 to 17% each (startups.co.uk).

Research from Cornell and Wharton points the same way. In a survey of 3,562 Kickstarter creators, teams with at least one member who had been a stranger were more than twice as likely to fail as teams of friends, family or coworkers (Cornell Chronicle). The study looked at Kickstarter projects rather than venture-backed startups, so treat the exact figure with care.

What the two findings have in common is history. If you've worked with someone, you already know what they do when a deadline slips or a customer complains. With someone new you have to find that out on purpose, and most of this guide is about how.

1. Work out what you need

Steve Blank, in an interview with First Round Review, suggests a simple exercise: list what the business has to do well, then list what you can cover yourself. Whatever is left over is your cofounder's job.

For most people the skills gap is the obvious part: a founder who sells needs someone who builds, or the other way round. The less obvious part is how you'll work together: the hours each of you can give, how fast you want to grow, how you make decisions, and what you will and won't give up.

Write it down before you talk to anyone. Four or five sentences is enough. For example:

I'm building booking software for independent physiotherapy clinics. I spent six years selling software to clinics, and ten of them have agreed to try a first version. I need someone who can build and run the product. I'm full-time from January and would want you full-time within six months. I'm in Pune and open to remote work, with a few days together each month.

Then mark the must-haves and keep that list short. Gloria Lin, who wrote up her own search in The Founder Dating Playbook, allowed herself three. Decide on hours early. If you need someone full-time within six months, that belongs on the list, and it's a question for the first call.

2. Ask the people who know your work

A quarter of the cofounders in the Starling survey met at a former workplace and another quarter at school or university, so go back through those people first. Ask two questions: would you consider starting something, and if not, who's the best person you've worked with who might?

Someone happy in their job may say no to the first question. They can still answer the second.

Then share your note where people who know you will see it: LinkedIn, an old team's group chat, your university's alumni network. A post that says "looking for a technical cofounder for a SaaS startup" doesn't make anyone think of a particular person. The physiotherapy version might.

3. Meet new people where they're looking too

If your network comes up empty, go where people are looking for a cofounder or want to build something with strangers. These are the places we'd use. We build Foundinity, so it comes first, and we've kept its entry to what it does.

Foundinity. You state what you need in a cofounder, and answer an assessment about how you work: commitment, working style, ambition, personality, goals, values, alongside your skills and interests. A multi-trait algorithm built on established psychological compatibility models ranks the people you see, and every match shows where the two of you align and where you differ. When you match, a 10-day clock starts, and you can book the intro call in the app. Only members compatible with you can see your profile. It's free to get started, with 5 likes a week and up to 3 active matches; Plus is $5 a week or $15 a month. More on how it works.

Founder Institute's online cofounder events. Networking events open to anyone. Founder Institute says it doesn't guarantee you'll find a cofounder, which is true of every option on this list.

Startup Weekend and hackathons. Techstars Startup Weekend gives you 54 hours to pitch ideas, form a team and build something. It's the quickest way we know to see how someone behaves under pressure, which you can't learn from a profile.

Indie Hackers. A free forum for people building in public. Its dedicated partner-finding group has closed, so post your note and search older posts. The useful part is that you can read someone's progress updates before you get in touch.

Full-time programmes. Entrepreneur First and Antler take individuals and help them form teams. Both are selective and expect you full-time, so they only suit people ready to leave their job now.

If you're weighing up cofounder platforms, we compare the main ones, with costs, in our comparison of cofounder platforms.

4. Use first calls to learn how they work

First calls are usually friendly, which makes them easy to misread. Ask questions whose answers you'll be able to check later:

• Why this problem, and why now?
• How many hours a week can you give this now, and in six months?
• How long could you work on it before you need an income?
• When did you last disagree with someone you worked with, and how did it end?
• What would make you walk away?

For 20 more, with what to listen for in each answer, see questions to ask a potential cofounder.

If you can, go further. In the First Round playbook, both people answer a 50-question questionnaire on their own and then compare. Lin and her eventual cofounder spent 6 to 10 hours over three to five sessions going through theirs. Set against the years you might spend working together, that isn't much.

If you matched on Foundinity, the compatibility breakdown already shows where the two of you differ, so the first call can start there.

5. Work on something real together

Talking only gets you so far. Steve Blank's advice is to work on an intense project together first, then run a trial period of about 30 days. The founders of Whalesync ran a six-week trial with weekly check-ins, and agreed the goals, the length and the check-in schedule before they began. Lin worked with her cofounder for four months before committing.

Pick a project with a deadline and a result someone outside the two of you can judge: ten customer interviews in two weeks, a landing page that has to collect sign-ups, a prototype that real users try.

While you work, watch the small things. Does the other person tell you early when something will be late? Do they do what they said without a reminder? When you disagree, do you end up with a decision?

6. Decide and commit

At the end of the trial, decide. If you both want to continue, agree when each of you will be working the hours you discussed, and say it plainly. If you don't, say why, and ask whether they know anyone who might fit.

Parting ways is a normal result. Lin met six potential cofounders over about a year, and one of those ended after a couple of coffee chats.

Signs to slow down

• They want a title before doing any work.
• Their hours are vague, or shrink each time you ask.
• They won't do a trial project.
• They go quiet for days during the trial without saying why.