Companies spend enormous energy on compensation benchmarking and title inflation to keep employees from quitting, but according to the 2023 SHRM Workplace Romance & Relationships Survey, one of the strongest predictors of retention has nothing to do with the paycheck at all. Employees with close friends at work are far more likely to stay put: 76% of workers with a close workplace friendship said it made them more likely to remain with their employer, and 85% said having a close friend at work had a positive impact on their career overall. Salary bumps and promotions matter, but this research suggests the office friendship down the hall might matter just as much.
Belonging turns out to be a retention metric
SHRM’s survey found that workplace friendship is closely tied to something HR departments have struggled to quantify for years: a genuine sense of belonging. Among employees with close work friendships, 80% reported feeling a strong sense of belonging at their organization, compared with just 63% of employees without one. That 17-point gap is not a rounding error — it represents a meaningfully different day-to-day experience of the same job, at the same company, depending entirely on whether a person has someone they consider a real friend nearby.
Job satisfaction follows the same pattern. SHRM found that 86% of employees with workplace friends described themselves as content in their roles, compared with 74% of those without close work friendships. Both numbers are respectable in isolation, but the 12-point gap suggests friendship is doing real work that job design, management quality, and compensation alone are not fully covering.
Why this matters more than it sounds like it should
It’s easy to dismiss workplace friendship as a soft, feel-good metric next to something concrete like salary. The retention numbers argue against that dismissal. A quarter of employees, 25%, told SHRM they would consider leaving their job if a close friend at work departed first — meaning workplace friendships are not just a pleasant byproduct of a good job, they are functioning as an active retention mechanism that can cut both ways. Lose the friend, and the company risks losing the employee attached to that friendship too.
This lines up with what organizational psychologists have argued for years about the difference between a job people tolerate and a job people are actually invested in. Compensation sets a floor for whether someone will consider leaving. Relationships appear to set the ceiling for how reluctant they are to actually do it, even when a better-paying opportunity is on the table.
The retention math companies tend to ignore
Most retention strategy still centers on the things that are easiest to benchmark against competitors: base pay, bonus structure, title, remote-work flexibility. Those levers matter, but they are also the easiest for a competing employer to simply match or beat with a slightly higher offer. A close workplace friendship is much harder to replicate somewhere else. An employee can walk into a new job with a higher salary on day one; they cannot walk in with three years of inside jokes, mutual trust, and someone who already knows how they take their coffee during a bad week.
That asymmetry is likely part of why SHRM’s numbers show such a wide gap in loyalty between employees with and without close work friends. A competitor can out-bid a salary. It is much harder to out-bid a genuine friendship an employee would have to walk away from to take the new job.
What this means for how workplaces should actually be built
None of this suggests companies should engineer friendships artificially, which tends to backfire and feel forced. But SHRM’s data does make a reasonably strong case that the informal, unscripted relationships that form between coworkers deserve to be treated as a real retention asset rather than an incidental nice-to-have. Team structures that isolate people, constant reshuffling of who sits near whom, and cultures that discourage anything resembling personal connection at work may be quietly undermining loyalty in ways no exit interview will ever fully capture, because the employee who leaves after losing a work friend rarely lists “loneliness” as the reason on the way out the door.
The practical takeaway for managers is less about programming forced bonding exercises and more about protecting the conditions that let friendships form naturally in the first place: stable teams that stay together long enough for trust to build, physical or virtual spaces where informal conversation is actually possible, and a general reluctance to treat every minute of the workday as something that needs to be optimized away from casual connection. SHRM’s numbers suggest that the return on that kind of patience shows up exactly where it matters most to a company’s bottom line — in who stays.
