Just two in 10 U.S. employees say they have a best friend at work, according to Gallup’s ongoing workplace research — a number that has held roughly steady for years even as employers pour budget into trust falls, escape rooms, and off-site retreats meant to manufacture exactly that kind of bond. The gap matters more than it sounds. Gallup has tracked the connection between a workplace best friend and how likely someone is to stay, recommend their employer, and actually engage with their job, and the pattern is consistent enough that it shapes how the firm advises companies to think about culture. What its data does not support is the assumption that a single offsite builds it.
What Gallup’s Numbers Actually Show
Gallup has been asking American workers about workplace best friendships since the early 2000s, and the trend line is one of the more stable findings in its research. In the most recent large sample — 16,586 U.S. employees surveyed in June 2022 — 44% of workers with a best friend at work said they would strongly recommend their organization as a place to work, up from 39% in 2019. Among employees without one, that figure sat at just 21% to 26% across the same years. Gallup has also found that the relationship between having a work best friend and outcomes like engagement and safety has gotten stronger, not weaker, since the pandemic pushed so much work remote or hybrid — the opposite of what you’d expect if proximity alone explained the effect.
Team Building Isn’t Useless — It’s Just Aimed at the Wrong Layer
None of this means team-building days are a waste. A meta-analysis published in Small Group Research by Cameron Klein and colleagues, pooling 20 studies covering more than 1,500 teams, found team-building interventions produce a real, moderate positive effect on team functioning overall. But the study broke that effect down by outcome type, and the breakdown is telling: team building’s impact on affective outcomes (morale, mood) and process outcomes (coordination, communication) was strong, at a correlation of .44 for each. Its impact on the interpersonal-relations component specifically — the piece closest to actual friendship formation — came in lower, at .26, similar to problem-solving skills and well behind the boost to mood and process. In other words, a retreat can genuinely make a team feel more cohesive for a day. It is not, on the evidence, the primary mechanism that turns coworkers into people who text each other outside of Slack.
Where Real Friendships Actually Form
So what does work, according to the people who’ve studied this longest? Gallup’s own guidance to employers leans almost entirely on the ordinary texture of a workweek rather than planned events: managers who visibly model friendship themselves, communication rhythms like regular team check-ins and small-group updates, and public storytelling that gives coworkers a reason to notice each other as people rather than job titles. One organization Gallup worked with lifted its best-friend-at-work scores by more than 80 percentile points using something as unglamorous as a weekly spotlight on employee friendships — not a new offsite budget line. Writing in Harvard Business Review, Gallup CEO Jon Clifton frames the stakes in blunter terms: more than 300 million people worldwide report having no one they’d call a close friend, and workplace bonds have become, in his words, “lifelines” for connection during a period when many people’s social circles have otherwise contracted. That’s a case for building friendship-friendly conditions into daily work, not for a single annual event to carry the weight.
Why the Distinction Is Worth Getting Right
The retention math is the part employers tend to skip past. Gallup’s research consistently links having a work best friend to higher engagement, better safety records, and stronger customer-facing performance — the kind of outcomes that show up on a P&L, not just an engagement survey. If a company wants more of that, the evidence points toward investing in the unglamorous, recurring stuff: giving people time and reason to talk during the actual workday, having managers who treat friendship as a legitimate part of culture rather than a soft perk, and creating small, repeated moments of visibility between coworkers. A two-day retreat can be a nice reset. It was just never going to be where the real relationship-building happened.
What the data ultimately suggests is a mismatch between where companies spend and where the return actually lives: the big, visible line items go toward one-off events, while the thing that predicts whether someone stays and recommends the job keeps forming quietly, in the ordinary minutes between meetings.
