How to find a cofounder with no network: start from zero and meet the right person in 8 weeks

Most people who start a company do it with someone they already knew. That's fine advice if your old team is full of engineers itching to leave, or your university friends now work in startups. It's less useful if you moved to a new city, work in an industry where nobody starts companies, studied something unrelated, or simply never met anyone who builds things.

This guide is for that situation. It covers why having no network is a smaller problem than it feels, where to meet people when you know nobody, how to make yourself worth replying to, how to approach a stranger, how to judge someone with no shared history, and a 6 to 8 week plan. If you do have colleagues or classmates who might fit, start with our main guide on how to find a cofounder instead.

Where people actually meet their cofounders

In a Censuswide survey of 500 cofounders for Starling Bank, published in April 2026, the most common routes were friends and family (34%), a former workplace (26%) and school or university (25%). Smaller groups met through a professional association or community event (17%), at a coworking space (16%) or at a dedicated networking event (15%) (startups.co.uk).

Two other numbers from the same survey matter more if you're starting from zero:

• 60% met their cofounder by chance, not during a deliberate search.
• Only 35% were actively looking when they met. Another 43% were "open to finding" a cofounder but not looking, and 19% weren't looking at all.

Read that from the other side. Most of the people who would make a good cofounder for you aren't searching. They're working in your industry, contributing to projects, answering questions in forums and turning up at events for reasons of their own. You won't find them by searching harder. You find them by being somewhere they already are, often enough that they notice what you do.

Why no network is a smaller problem than it feels

"I have no network" usually means "I have no startup network". Most people know more people than they think: former classmates, people from jobs outside tech, clients, suppliers, people in the same online groups, friends of friends. These are weak ties, the people you know a little rather than well.

Weak ties have a long record in research on how people find opportunities. One of the largest tests ran on LinkedIn: a team of academic and LinkedIn researchers, publishing in Science in 2022, analysed experiments that changed how many weak ties appeared in the networks of more than 20 million people over five years. They found that moderately weak ties, people you share some connections with but don't know well, did the most for people's chances of moving jobs (Stanford Digital Economy Lab). The study measured job moves, not cofounder searches, but the mechanism is the same: new opportunities tend to come through people slightly outside your circle, because they hear about things you don't.

So the job isn't to build a big network. It's to build a small set of weak ties in the right places, and make sure those people know what you're looking for.

The real risk with strangers

There's one honest reason to take extra care. Jason Greenberg of Cornell and Ethan Mollick of Wharton studied 3,562 Kickstarter creators and found that teams with at least one founder who had been a stranger before forming the company were more than twice as likely to fail as teams of friends, family or coworkers. The stranger teams brought wider networks, but relationship problems outweighed that advantage (Cornell Chronicle). The study covered crowdfunded projects rather than venture-backed startups, so treat the exact figure with care.

The lesson isn't to avoid strangers. It's that shared history does a job you now have to do on purpose. Someone you worked with for two years has already shown you how they behave when a deadline slips. With someone new, you have to create those moments and watch them, which is what the second half of this guide covers.

Before you look: make yourself worth replying to

With no mutual contact to vouch for you, your work has to do it. Before you contact anyone, get three things ready.

A short note. Four or five sentences: what you're working on, who it's for, what you've done so far, what you bring, and what kind of person you need. "Looking for a technical cofounder" gives nobody a reason to reply. "I've spent six years running operations at logistics firms, I've interviewed 20 warehouse managers about returns, and I'm looking for an engineer who wants to build tools for them" does.

Something people can look at. A page, a write-up of customer interviews, a prototype, a public repository, a few posts about what you've learned. It doesn't need to be polished. It needs to show that you've started and that you finish things.

Your must-haves. Two or three, written down: the skills gap you need filled, the hours you expect, and where the person needs to be. Everything else is a preference. Knowing the difference stops you rejecting good people for small reasons and accepting poor fits because they're keen.

Where to meet a cofounder when you know nobody

Pick two or three of these, not all of them. Going back to the same places every week builds the kind of familiarity that a single visit to ten places never does.

Foundinity

We build Foundinity, so it comes first. It suits people with no network because everyone there is looking for a cofounder, and nobody needs an introduction. 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, so you start with compatible people instead of scrolling through anyone who signed up.

Every match shows where the two of you align and where you differ, which replaces some of what a mutual friend would have told you. When you match, a 10-day clock starts, and you book the intro call in the app, so a conversation with a stranger turns into a call instead of a thread that fades. Your profile is visible only to compatible members, and an "exploring" status lets you look while you're still in a job. It's free to get started; Plus is $5 a week or $15 a month. See how it works.

Online communities in your field

Founder forums, subreddits, and Slack or Discord groups for your industry or your craft (developers, designers, marketers, people in healthcare or logistics). The value isn't the "looking for a cofounder" channel. It's that you can watch how people think, write and treat others for weeks before you message anyone, and they can watch you.

Answer questions you know the answer to. Share what you learn from customer conversations. After a month, the people who reply to you most thoughtfully are your shortlist, and some of them will have noticed you first.

Open source projects

If you build software, or want to meet people who do, contribute to an open source project in the area you care about. Issue threads, code reviews and pull requests show how someone explains a decision, takes criticism and follows through, in public and over time. Many projects label beginner-friendly issues, so you can start small. Non-technical contributions such as documentation, design, testing and community work count too, and they put you in front of the maintainers.

Building in public

Post regular, specific updates about what you're working on: what you tried, what customers said, what failed. Pick one place where people in your field read, and keep a steady rhythm, such as one post a week. Over a couple of months this does two things. People who care about the same problem find you, and when you do message someone, they can see months of evidence that you're serious.

Hackathons and Startup Weekend

A hackathon is the closest thing to a trial project you can get in a weekend. Startup Weekend, run by Techstars, is a three-day event where people pitch ideas, form teams and build for 54 hours. Major League Hacking says it runs more than 1,000 events a year (its own figure) and lists upcoming hackathons on its events page.

Go to build, not to recruit. Join a team with people you don't know, take on a real part of the work, and notice who you'd want to work with again on Monday. Follow up within two days.

Local meetups, talks and coworking spaces

In the Starling survey, 17% met through a community event and 16% at a coworking space. Choose one regular meetup in your field rather than general startup mixers, and go three or four times. The second and third visits are where conversations stop being introductions. Give a short talk if you can: speaking once makes more people come to you than an evening of handing out cards.

Your industry, not the startup scene

If you know an industry well, its conferences, trade groups and professional associations are full of people who understand the problem you want to solve, and almost none of them are being approached by founders. A nurse who knows how rotas really work, or an engineer who has maintained a hospital's scheduling system, may be a better cofounder for a health product than anyone at a startup event. Many of them are in the 43% who are open to the idea but not looking.

Full-time programmes that build teams from individuals

Some programmes are designed for people who arrive alone, with no idea and no cofounder. They bring a group of individuals together, usually on site for several weeks, and run a process to help members form teams. They expect you to commit full-time, so they suit people who are ready to leave a job.

How to approach someone you don't know

A message from a stranger competes with everything else in someone's inbox. The ones that get replies have a few things in common.

1. Lead with them. Mention something specific they wrote, built or said. It shows you paid attention, and it's the reason you're writing to them and not someone else. 2. Say who you are in one line. What you've done that's relevant to them, not your whole background. 3. Ask for something small. A 20-minute call, feedback on a page, an opinion on a problem. "Want to be my cofounder?" from a stranger is too big to say yes to. 4. Be honest about why. Say you're looking for someone to build this with. People respect it more than a disguised pitch, and it saves both of you time. 5. Make it easy to say no. One message, one follow-up a week later, then stop.

Two examples, written for this guide:

To an open source contributor: "Your pull request on the offline sync issue last month was the clearest explanation of the conflict problem I've read. I'm working on field-service software for small plumbing firms, where technicians lose signal in basements all day, and I've talked to 15 owners so far. I handle sales and operations, and I'm looking for an engineer to build it with. Would you be up for a 20-minute call to tell me whether my approach to sync is naive?"

To someone from an industry group: "I enjoyed your answer in the procurement group about why approval chains break in mid-sized firms. I've been building a prototype to fix exactly that and would value 20 minutes of your view on it. Full disclosure: I'm also looking for someone with your background to start this with, but a call to pick your brain is all I'm asking for."

Our guide on how to ask someone to be your cofounder has more example messages and covers the conversation that follows.

How to judge someone you've never worked with

Without shared history, you're building a picture from scratch. Collect evidence from several directions, and give it time.

Read their public work. Code, posts, talks, past products, how they answer questions. It's the closest thing you have to a work history, and it's harder to fake than a good first call.

Talk to people who have worked with them. Ask for two names, a former manager and a peer, and ask those people what the person is like on a bad week. If someone can't name anyone, ask why.

Vary where you talk. One video call shows how someone presents. A working session, a walk, a call after a long day and a call when something has gone wrong show how they are.

Watch small promises. If they say they'll send something on Thursday, see whether it arrives on Thursday, or whether you hear early that it won't. Small promises predict large ones.

Ask the hard questions early. Hours, money, where each of you will live, how fast you want to grow, what would make each of you walk away. Our questions to ask a potential cofounder has a full list.

Run a longer trial. Sean Murphy of SKMurphy suggested in 2010 that a friend or former coworker might take a month or two to reach a working arrangement, a friend of a friend three to six months, and a stranger met at an event "three to nine months working on a few projects of escalating complexity" (SKMurphy). You don't need nine months, but plan for more than one short project. Our guide to running a cofounder trial project covers how to set one up, step by step.

A 6 to 8 week plan to find a cofounder with no network

This is a plan, not a promise. It gets you from knowing nobody to a shortlist and the start of a trial. Many searches take longer, and that's normal.

Week 1: prepare. Write the note and your must-haves. Put up one piece of work people can look at: a short write-up of what you've learned about the problem is enough.

Weeks 1 and 2: pick your places. Set up your Foundinity profile and answer the assessment honestly, including the parts you think make you look less impressive. Join two communities where people in your field talk, and book one event or hackathon.

Weeks 2 to 4: show up and be useful. Post once a week. Answer questions. Contribute to one project. You're not looking for a cofounder in these weeks so much as becoming someone people recognise.

Weeks 3 to 5: reach out. Message the people whose work you've noticed, one at a time, with something specific. Reply to your matches. Aim for five or more first calls. Keep a simple list of who you've spoken to and what you learned.

Weeks 5 and 6: go deeper. Meet the two or three most promising people again, in a different setting. Ask about hours, money and how they make decisions. Talk to someone who has worked with them.

Weeks 6 to 8: start a trial. Choose one person, write a one-page brief with a finish line and a decision date, and start. Keep other conversations warm and tell them honestly where you are.

Mistakes people make when they start with no network

• Collecting places instead of people. Joining 12 communities and posting once in each builds nothing. Two places, every week, for two months, builds weak ties.
• Asking before giving. A first post that says "looking for a cofounder" in a group you joined yesterday reads as spam. Contribute first.
• Pitching the idea instead of the work. Strangers can't judge your idea, and many won't care about it yet. They can judge what you've already done.
• Messaging everyone the same way. A copied message to 50 people gets fewer replies than five messages that mention something specific.
• Treating chemistry as proof. A great first call with a stranger tells you they're good at first calls. Look for evidence across weeks.
• Committing too fast to end the loneliness. Searching alone is tiring, and the first person who says yes is tempting. A short trial costs weeks; a wrong cofounder can cost years.
• Skipping references because it feels awkward. It's normal, and anyone serious will understand. Offer your own references in return.
• Hiding from your existing weak ties. Tell former classmates and colleagues what you're doing, even if none of them are in startups. Someone they know might be.

FAQ

How do I find a cofounder if I don't have a network?

Use places where meeting strangers is normal, such as a cofounder matching platform like Foundinity, hackathons and communities in your field. Then build a small network on purpose by showing up in the same two or three places every week, being useful, and contacting people whose work you've seen. Make up for the missing history with a longer trial project.

Can you find a cofounder without connections?

Yes. In a 2026 survey of 500 cofounders for Starling Bank, 17% met through a community event, 16% at a coworking space and 15% at a networking event. Those routes need no prior connection, only time spent in the right places.

Where can I meet a cofounder online?

Cofounder matching platforms, founder forums, Slack and Discord groups for your industry, open source projects, online hackathons and any place you build in public. Choose the ones where people with the skills you need already spend time.

How do I approach a potential cofounder I don't know?

Mention something specific they wrote or built, say who you are in one line, ask for a short call or their opinion rather than a partnership, and be honest that you're looking for someone to build with. Send one follow-up at most.

Is it risky to start a company with a stranger?

It carries more risk than starting with someone you've worked with, because you have no history to go on. A Cornell and Wharton study of Kickstarter teams found stranger teams were more than twice as likely to fail. You reduce the risk by checking references, talking in different settings and running a longer trial.

How long does it take to find a cofounder with no network?

Expect a couple of months to build a shortlist and longer to decide. Our 6 to 8 week plan gets you to the start of a trial; the trial itself adds weeks or months. More detail is in how long it takes to find a cofounder.

Should I join a programme to find a cofounder?

If you can commit full-time, programmes that build teams from individuals are made for people who arrive without a team. If you need to keep your job while you look, a matching platform and part-time trial projects fit better.

What should I do before reaching out to strangers?

Write a short note on what you're building, what you've done and who you need, put up one piece of work people can look at, and decide on two or three must-haves. That gives strangers a reason to reply.

How do I build a startup network from scratch?

Pick two places where people in your field gather, show up every week, help others before asking for anything, and share what you're learning. Weak ties built this way often lead to the person you're looking for, or to someone who knows them.

What if I'm a first-time founder with no startup experience?

That's common, and it isn't a barrier. Bring evidence of the problem instead: customer conversations, industry knowledge and something you've started. Be clear about what you can do and what you need, and look for someone whose strengths cover the gap.

Can I find a technical cofounder with no network?

Yes, but engineers get many requests, so show progress first: customer interviews, a waitlist or a simple prototype. Our guide on how to find a technical cofounder goes through the process.

Sources

• Starling Bank and Censuswide survey of 500 cofounders, reported by Katie Scott in "Most co-founders meet when they're not actively looking for business partners, data reveals", startups.co.uk, 29 April 2026, checked 12 November 2026.
• Rajkumar, Saint-Jacques, Bojinov, Brynjolfsson and Aral, "A Causal Test of the Strength of Weak Ties", Science, 2022, via the Stanford Digital Economy Lab, checked 12 November 2026.
• Tom Fleischman, "Familiarity breeds success for fledgling companies", Cornell Chronicle, 30 September 2025, on Greenberg and Mollick, Journal of Business Venturing, checked 12 November 2026.
• Sean Murphy, "Finding A Co-Founder", SKMurphy, 25 June 2010, checked 12 November 2026.
• Major League Hacking, home page and 2027 season events, checked 12 November 2026.
• Techstars, Startup Weekend page (techstars.com/communities/startup-weekend), checked 12 November 2026.
• Foundinity plans and how it works from foundinity.com, checked 12 November 2026.