Usually not first, and almost never as cofounders. Friends and family are the easiest people to recruit and the hardest people to manage, measure, or let go — which is why the research on founding teams treats close ties as a risk to price in rather than a shortcut to take.
That does not make them unhirable. Referrals from people you trust are one of the best-documented hiring advantages that exists. The distinction that matters is structural: where the relationship sits relative to you, how much of the company's authority it touches, and whether the person could have been hired without the relationship.
Table of Contents
- What the founding-team research actually shows
- Family ties hurt in one direction and help in another
- The hiring cost you cannot see from inside
- Why close ties are so easy to default into
- The referral advantage you should keep
- If you are going to do it anyway, do these things first
- Frequently Asked Questions
What the founding-team research actually shows
The strongest evidence against starting with friends comes from the cofounder seat, not the employee seat. Noam Wasserman's study of nearly 10,000 technology and life-sciences founders, summarized by Adam Grant for the World Economic Forum, found friend-based founding teams were the least stable of any type — each pre-existing friendship on a team raised founder turnover by 28.6%. Teams of former coworkers were the most stable. The mechanism is not that friends are disloyal.
It is that friendships are built on avoiding conflict, and founding a company requires constant conflict over equity, titles, strategy and who is underperforming. A team of former colleagues has already watched each other work under pressure and has no social bond to protect by staying quiet. That matters because team composition is not a soft issue. CB Insights' review of 101 startup postmortems, distributed by the Angel Capital Association, ranked "not the right team" the third most common cause of failure at 23%, behind no market need (42%) and running out of cash (29%).
Family ties hurt in one direction and help in another
The cleanest finding on family hiring separates two relationships that get discussed as one. Ertug, Kotha and Hedström studied 4,967 new firms founded in Stockholm between 1998 and 2003 and published the results in the Academy of Management Journal. Kin ties between founders and employees hurt firm performance. Kin ties among employees improved it.
The asymmetry is about power. A relative who reports to you imports two problems: the information you get is narrower, because the person shares your assumptions and your social world, and the non-family staff read the arrangement as nepotism. Two employees who happen to be cousins carry neither cost — they get the referral benefits with none of the authority distortion. The practical translation: your sister-in-law as your first operations hire is the structurally risky version. Your operations hire's brother joining later, on the same bar as anyone else, is close to the version the data likes.
The hiring cost you cannot see from inside
Visible closeness at the top of a company changes who applies. A preregistered field experiment described in a research roundup at National Affairs found job candidates were less likely to apply to a startup cofounded by spouses, with the penalty strongest when the lead founder was male. The applicant pool shrinks before you ever see a résumé.
The reaction holds even when the relative is good. A Swiss experiment with 165 employees, published in Organizational Behavior and Human Decision Processes, found a well-qualified candidate was evaluated more negatively purely because the candidate had family ties to the company — and perceived nepotism lowered fairness perceptions and willingness to join the organization. This is the cost founders systematically underestimate, because it is invisible. You never meet the engineer who read your team page, concluded the senior roles were taken by the family, and closed the tab.
Why close ties are so easy to default into
Founders do not usually choose friends and family after weighing alternatives. The network does the choosing. Ruef, Aldrich and Carter's analysis of a representative U.S. sample of founding teams, published in the American Sociological Review, found strong ties — spouses, romantic partners and family — exerted the most pronounced constraint on who joins a team, while weak ties such as business acquaintances imposed no significant constraint. Strong ties are available at 11pm, work before there is money, and require no interview process you have not built yet.
The constraint is real and it is not a character flaw. It is the reason a deliberate first hire takes months and a default one takes a weekend. The counterweight is that similarity among close ties is not automatically fatal. Kovács and Sels studied 1,125 U.S. startups for the Strategic Management Journal and found politically heterogeneous founding teams were *more* likely to shut down than similar ones. Alignment buys cohesion; the question is what you gave up to get it.
The referral advantage you should keep
Rejecting family cofounders does not mean rejecting your network as a hiring channel. Burks, Cowgill, Hoffman and Housman examined personnel records from nine large firms across call centers, trucking and high-tech for the Quarterly Journal of Economics. Referred hires quit less often and generated more profit per worker, with raw productivity similar to non-referred hires. Read that result carefully, because it is routinely misquoted.
The gain came from retention and lower recruiting cost — not from referred candidates being more talented. A referral is a reliable signal about whether someone will stay and fit, and a weak signal about whether they can do the job. So use your friends to source candidates, and use your process to select them. A friend's recommendation of a stranger captures almost all of the referral benefit and none of the authority problem.
If you are going to do it anyway, do these things first
Many founders will hire a close tie regardless — sometimes because there is genuinely no one else who will work for equity. The goal then is to make the arrangement survivable.
The honest limit on all of this: the family-business literature has not converged. A review by Firfiray and colleagues in the Human Resource Management Review reports no theoretical or empirical consensus on nepotism's effect on performance, and notes a 163-firm Tanzanian study that found no significant negative effect on financial or market performance — with outcomes contingent on the institutional environment and the level of uncertainty a firm faces. Where formal hiring markets are thin and trust is expensive to establish, a relative can be the rational hire.
- Write the exit terms before the start date. Agree in writing, while everyone is optimistic, what happens to the role and the equity if it does not work.
- Use vesting with a cliff, same as any other equity grant. No exceptions for relatives.
- Put the person under someone who is not you, or give them a scope where you are not the only judge of their performance.
- Hire for a defined role with a defined bar, not "help out." Undefined roles are the ones that cannot be ended.
- Check who else can see the relationship. If it will appear on your team page, assume candidates will price it in.
Frequently Asked Questions
Is hiring a friend as employee number one as risky as making them a cofounder?
No. The turnover finding from Wasserman's data is about founding teams, where friends hold equity and authority and must fight over strategy. An employee with a defined role, a manager, and vesting terms is a much smaller structural bet — though the perceived-nepotism effect on other staff still applies if the relationship is visible.
What if my relative is objectively the best candidate I can get?
The Swiss experiment is the relevant warning: a well-qualified candidate was still rated more negatively solely because of family ties. Being genuinely qualified does not neutralize how the hire reads to everyone else, so plan to document the bar they met and keep them out of a reporting line that runs only to you.
Should I hide the relationship from the team or from investors?
No — concealment converts a manageable perception problem into a trust problem when it surfaces. Disclose it with the terms attached: the role, the bar, the vesting, and who evaluates performance.