What to look for in a cofounder: 10 criteria beyond skills, and how to check each one

Paul Graham opened his "Startups in 13 Sentences" with cofounders: "Cofounders are for a startup what location is for real estate." You can change the idea, the product and the market. Changing the people you started with is much harder (Paul Graham).

Elad Gil, writing about the same choice, called it the most important early decision a founder makes and compared it to a marriage (Elad Gil). Yet most people choose on what they can see in an hour: a good CV, an easy conversation, enthusiasm for the idea.

This guide sets out what to look for in a cofounder beyond skills, why each criterion matters, what a good sign looks like, and how to check it with evidence rather than a feeling. It's the criteria. For specific signals, see our guides to cofounder green flags and cofounder red flags; for the conversation itself, our questions to ask a potential cofounder.

What to look for in a cofounder: the short answer

Skills tell you who can do the job. The rest tells you who you can actually build with for years.

Criteria about skills

1. Skills that cover your gaps

Skills are where everyone starts, and they matter. Steve Blank's way in is to list what your business needs to do and what it needs to own, then compare that with what you bring: the gap "reveals the expertise you need to find in your co-founder" (First Round Review). Gil makes a similar point: avoid overlapping roles unless you're all engineers, and avoid three business founders at a technology company, because there isn't enough business work to go around.

So look for someone who is strong where you're weak, and who wants to do that work. A good engineer who wants to stop writing code is a poor fit for a company that needs one.

How to check skills. Ask to see something they made and talk through a decision they took in it. Better still, give the trial a part only they can do (see our guide to running a cofounder trial project). References help here too: "What was their part in that project?" is a question colleagues can answer precisely.

Once the skills fit, the remaining nine criteria decide whether the partnership lasts. Hustle Fund, summarising a conversation between its partner Eric Bahn and coach Ben Tauber, calls skills and experience the obvious criteria and values the less obvious priority (Hustle Commons).

Criteria about time and drive

2. Commitment

Why it matters. If one of you works 50 hours a week and the other works 10, resentment tends to build quickly, however well you get on. Commitment is also the easiest criterion to check early and the one people most often leave vague.

What good looks like. Their hours now and their hours in six months are specific, and they match yours, or the gap has a date attached. They know how long they can go without a salary. If they have a job, they can say what would need to happen before they leave it.

How to check it. Compare numbers, not adjectives: hours a week, months of runway, the date or milestone that changes things. Then look for evidence. Commitment shows in what someone has already moved aside for it: a weekend kept free, a side contract turned down, a conversation at home about money. Once you start working together, their calendar tells you more than anything they say.

3. Follow-through

Why it matters. Early companies run on small promises: "I'll send the deck tonight", "I'll call those three clinics this week". A partner whose small promises slip creates work for you every day.

What good looks like. They do what they said, when they said. When something will slip, they tell you before the deadline, not after it.

How to check it. Keep a private note for three weeks: each time they promise something, write down what and by when, then whether it happened. This is different from asking whether they've finished projects in the past (our green flags guide covers that). Here you're measuring how they treat commitments made to you. Ten small promises tracked honestly beat any interview answer.

4. Determination that can bend

Why it matters. Paul Graham's essay on what he looks for in founders puts determination first, calling it "the most important quality in startup founders", and pairs it with flexibility: the ability to "modify your dreams on the fly" (Paul Graham). You want both. Determination without flexibility keeps building something nobody wants. Flexibility without determination gives up when the first channel fails.

What good looks like. They kept going through a stretch where nothing worked, and they changed the plan when the evidence said so. They can tell you which part they held on to and which part they let go.

How to check it. Look at what they did after a setback: a rejected application, a failed launch, a lost job. What did they do in the next month? A trial project usually produces one small setback of its own, such as interviews that don't get booked or a test that fails, and how each of you reacts in the following week is direct evidence.

Criteria about how you work together

5. Working style

Why it matters. Two capable people can still wear each other out. One plans in detail, the other improvises. One writes everything down, the other prefers a call. One works at 6 a.m., the other at midnight. None of these is wrong, but some combinations are tiring every single day.

What good looks like. You can name your differences and they don't bother either of you much. Where they do, you can agree a simple rule, such as a written update every Wednesday or two fixed hours of overlap.

How to check it. Work in the same room, or on the same video call, for a full afternoon. Then run a week where most of the work happens in messages. Notice the rhythm: how quickly each of you replies, how much detail you each need before starting, what happens when a plan changes halfway through.

6. How they handle conflict

Why it matters. You will disagree, often and about things that matter. Gil's advice is to agree up front who makes the final call in each area and "a pre-agreed upon way to resolve conflicts" (Elad Gil). The Hustle Fund piece lists what to watch for: whether someone becomes avoidant under pressure, gets upset easily, or is "comfortable being wrong sometimes".

What good looks like. Arguments stay on the problem. Reasons change minds. The decision holds the next day, and nobody brings it up again a week later as a grievance.

How to check it. Wait for the first real disagreement, which tends to come early in any shared project, often over what to do first or what to cut. Before it arrives, agree how you'd settle a tie, then see whether you actually use that method when it counts.

7. Personal maturity

Why it matters. Gloria Lin, describing her search for a cofounder, had two must-haves, and the first was personal maturity (First Round Review). It covers how someone handles stress, praise, blame and a bad week. In a two-person company, one person's bad week becomes the other person's problem.

What good looks like. They admit a mistake and say what they'll do about it. They stay roughly the same person when things go badly. They don't need to win every point.

How to check it. References are the best source: ask a former colleague what this person was like during the worst month they worked together. In a trial, notice what happens the day after something goes wrong.

Criteria about where you're each headed

8. Ambition

Why it matters. Gil warns that mismatched goals, such as one person wanting a quick sale and the other wanting to build something large, cause serious conflict. Ambition here means the size and shape of the company each of you wants: a profitable business that pays two salaries, a company that raises outside money and grows fast, or something in between.

What good looks like. You describe the same company when you each imagine year five, and you'd both be satisfied with the same outcomes.

How to check it. Ask each other to describe two futures: the outcome you'd be happy with and the outcome you'd regret. Write them down separately first, then compare. A gap here rarely closes on its own.

9. Personal goals

Why it matters. A company sits inside two lives. One of you may want to move cities in two years, start a family, or keep a part-time job for security. None of that rules anyone out, but it changes what each of you can give, and when.

What good looks like. Your plans for the next three to five years point in the same direction, and anything that could pull one of you away is said out loud early.

How to check it. Talk about the life around the company: where you want to live, what money you need, who else depends on you. The people closest to them are part of this too. If their partner or family doesn't know about the plan, that's worth discussing (our red flags guide explains why).

10. Values

Why it matters. Hustle Fund's piece puts it plainly: "Values are how decisions get made when you are not in the room." Gil's question is shorter: "How ethical are they?" You'll be trusting this person with customers, money and hires, mostly when you aren't watching.

What good looks like. Faced with the same dilemma, you'd make roughly the same call, or at least respect the other person's reasons.

How to check it. Use real dilemmas rather than abstract questions. Say you can ship on Friday with a bug you know about, or tell the customer you'll be a week late: which do they choose, and what do they tell the customer? Write your answers separately, then compare the reasons, not only the choices. Our cofounder compatibility test has a set of scenarios like this. References help again: "Did you ever see them cut a corner?"

What to look for first: criteria by stage

You can't check everything at once, and you don't need to. Some criteria can be compared before you meet. Others only show after weeks of work.

At the looking stage, screen hardest on the criteria that are cheap to check and hard to change: commitment, personal goals and ambition. A mismatch there ends the conversation fairly, and finding it early saves both of you weeks. Leave follow-through, conflict and maturity to a trial, because conversation can't show them.

Must-haves, table stakes and flex areas

Lin's advice is to keep the list of non-negotiables small: "don't have more than three." She sorted everything else into table stakes, must-haves and flex areas, where flex areas "might give you pause, but they don't have to be deal-breakers."

That's a useful way to turn ten criteria into a decision:

• Must-haves (up to three). The criteria where a gap ends it for you. For most people, one of them is commitment.
• Table stakes. Things you expect from anyone, such as basic honesty and showing up on time.
• Flex areas. Differences you can work with if you name them. Working style often belongs here.

Write your three must-haves down before you start talking to anyone. Written down in advance, they're harder to bend for someone you happen to like.

What people overrate when choosing a cofounder

Some things feel like strong evidence and aren't.

Friendship. Knowing someone well is helpful, but friendship and a working partnership are different things. Adam Grant, summarising Noam Wasserman's study of nearly 10,000 founders, reports that "each friendship in a founding team increased the rate of founder turnover by 28.6%", while "the most stable teams were past coworkers" (Next Big Idea Club). If you start with a friend, test the work as hard as you'd test a stranger's.

Similarity. People tend to start companies with people like themselves. In a nationally representative US sample of 816 people starting businesses, Ruef, Aldrich and Carter found strong support for founding teams forming around shared gender, ethnicity and occupation, and no evidence that teams with mixed functional backgrounds were more common than chance (Ruef, Aldrich and Carter, 2003). Feeling at ease with someone because they're like you isn't the same as fitting them.

Credentials. A well-known university or employer shows someone did well in that setting. It says little about how they work with no team, no budget and no one checking.

Love of your idea. Enthusiasm for your idea is pleasant to hear and changes quickly. Ideas change too. Look for interest in the problem and the customers, which tends to last longer than excitement about one solution.

Availability. "They're free right now" is a reason to talk, not a reason to choose. The person with the most time this month isn't necessarily the person with the most commitment.

Our green flags guide has a related list of signals that feel good but tell you little, such as easy chemistry and instant agreement.

Should a cofounder be like you or different from you?

Both, on different things.

Different: skills, and often temperament in useful ways. One person who enjoys detail and one who enjoys selling cover more ground than two of either. A 2023 study of more than 21,000 founder-led companies, which used machine learning to infer founders' personalities from their public Twitter activity, found six founder personality types rather than one, and reported benefits for larger, personality-diverse teams (McCarthy et al., 2023; UNSW Sydney). That's a correlation across many companies, not a rule for your pair, but it argues against looking for a copy of yourself.

Close: commitment, ambition, personal goals and values. These are the criteria where a gap means one of you ends up living with a company they didn't want. Working style sits in between: different is fine if you can agree how to handle it.

How to check criteria without turning it into an interview

Evidence comes from three places, and you want all three.

1. What they've done. Past work, past teams, past setbacks. Ask about real events with dates, not opinions. Our questions to ask a potential cofounder are built for this. 2. What they do with you. A few weeks of real work together, with a written brief and a decision date. Our cofounder trial project guide sets one up. 3. What others say. Two or three people who worked closely with them. Ask about specific moments: a deadline, a disagreement, a bad month.

When the three sources agree, trust them. When they conflict, weight what you saw yourself over what you were told.

A one-page cofounder scorecard

Copy this before your first call and fill it in after each stage. Score each criterion 1 to 5, and write one piece of evidence next to every score. A score without evidence is a feeling.

• Skills that cover my gaps: score, evidence
• Commitment (hours, runway, date): score, evidence
• Follow-through (promises kept / made): score, evidence
• Determination that bends: score, evidence
• Working style: score, evidence
• Conflict: score, evidence
• Personal maturity: score, evidence
• Ambition: score, evidence
• Personal goals: score, evidence
• Values: score, evidence
• My three must-haves, and whether each is met: yes, no, not yet known

Fill in your own scorecard as well, as the other person would score you. It's a fair check on your own expectations.

Hold yourself to the same criteria

Everything on this list runs both ways. The person you're talking to is deciding about you too, and the strongest thing you can do is make your own answers easy to see: your hours, your runway, your must-haves, what you've already built and the outcome you want. Being specific about yourself also makes it easier for the other person to be specific back.

Where Foundinity fits

Several of these criteria can be compared before the first call, and Foundinity is built to do that comparison for you.

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 people who already fit on the criteria that are hardest to change.

• A compatibility breakdown on every match shows where the two of you align and where you differ, so your calls and your trial can focus on what still needs checking.
• A 10-day clock on each match keeps it from sitting unanswered, and you book the intro call in the app.
• Your profile isn't public. Only compatible members can see it, and an "exploring" status lets you look at your own pace.
• Pricing: free to get started; Plus is $5 a week or $15 a month. See how it works.

FAQ

What should you look for in a cofounder?

Skills that cover what you can't do, and nine things beyond skills: commitment, follow-through, working style, ambition, personal goals, values, how they handle conflict, determination that can bend, and personal maturity. Check each one with evidence: past work, a trial project and references.

What are the qualities of a good cofounder?

They're committed in hours and time, they keep small promises, they disagree without making it personal, they want the same kind of company as you, and their values hold when no one is watching. Paul Graham adds determination and flexibility; Gloria Lin puts personal maturity first.

How do you choose a cofounder?

Write down your three must-haves first. Screen for commitment, ambition and personal goals before you meet, test values and working style on calls, then work together on something real for a few weeks before deciding. Talk to people who've worked with them.

What is the most important quality in a cofounder?

There's no single answer that holds for every pair, but commitment is the one most worth checking first: it's easy to compare, hard to change, and a mismatch causes strain within months. Graham names determination as the most important quality in founders generally.

Should your cofounder have the same skills as you?

Usually not. Overlapping skills leave gaps in the company and create competition over the same work. The exception is a team of engineers building a technical product, where shared skills are normal as long as roles are clear.

Is it a good idea to start a company with a friend?

It can work, but friendship isn't evidence that you'll work well together. Wasserman's research, as summarised by Adam Grant, found more founder turnover in teams built on friendships and the most stability among former coworkers. If you start with a friend, run the same trial and the same frank conversations you would with a stranger.

What should you look for in a technical cofounder?

The same criteria, plus evidence they can build what you need without outside help: shipped work you can look at, and a trial where they build something real. Our guide on how to find a technical cofounder covers this in detail.

What should you look for in a business partner?

In a small or established business, the same core criteria apply: commitment, values, how they handle conflict and follow-through. Add an open approach to money and a clear view of who does what day to day.

How long does it take to know if someone is the right cofounder?

Commitment, ambition and goals can be compared in a call or two. Follow-through, conflict and maturity take a few weeks of working together, which is why most advice ends with a trial project before anyone commits.

What are signs someone will be a bad cofounder?

Patterns such as avoiding customers, changing the idea every month, or a story that doesn't match the record. Our guide to cofounder red flags lists ten, with how to spot each one.

Can a test tell you who to choose as a cofounder?

A compatibility test can show where two people align and differ on things like commitment, working style and values, which tells you what to talk about and test. It can't replace working together. Our cofounder compatibility test guide includes a version you can run yourselves.

Sources

• Paul Graham, "Startups in 13 Sentences", February 2009, checked 12 November 2026.
• Paul Graham, "What We Look for in Founders", October 2010, checked 12 November 2026.
• Elad Gil, "How To Choose A Co-Founder", 27 February 2012, checked 12 November 2026.
• First Round Review, "Looking for Love in All the Wrong Places: How to Find a Co-Founder", interview with Steve Blank, 26 January 2015, checked 12 November 2026.
• Gloria Lin, "The Founder Dating Playbook", First Round Review, 30 September 2019, checked 12 November 2026.
• Hustle Commons (Hustle Fund), "Part 1: How to pick the right co-founder", undated, checked 12 November 2026.
• Adam Grant on Noam Wasserman's research, "Why Your Best Friend Might Not Be Your Best Business Partner", Next Big Idea Club, 5 February 2016, checked 12 November 2026.
• Martin Ruef, Howard E. Aldrich and Nancy M. Carter, "The Structure of Founding Teams: Homophily, Strong Ties, and Isolation among U.S. Entrepreneurs", American Sociological Review 68(2), 2003; summary checked 12 November 2026.
• Paul X. McCarthy and others, "The Science of Startups: The Impact of Founder Personalities on Company Success", arXiv 2302.07968, published in Scientific Reports 13, 17200 (2023); abstract checked 12 November 2026.
• UNSW Sydney, "Founder personality could predict start-up success: study", 18 October 2023, checked 12 November 2026.
• Foundinity features and pricing from foundinity.com, checked 12 November 2026.