How Much Time Does Friendship Actually Need?

You’ve heard some version of this: it takes 50 hours to make a casual friend, 90 to move to real friend, 200 to get to close friend. That figure gets repeated constantly, usually with no source attached, and usually as a way of explaining why your social life feels stuck. It is a genuine finding, from communications researcher Jeffrey Hall’s 2018 study on friendship formation. It is not in the source list for this piece, so treat the exact numbers as background rather than gospel. But the shape of the claim is right, and it points at something real: friendship is not made of grand gestures. It’s made of accumulated ordinary time.

That’s the actual problem. Not that you don’t want friends. That the ordinary time isn’t there.

Where the time went

Americans spend measurably less time face-to-face with other people than they used to, according to reporting on the American Time Use Survey covered by WBUR in February 2024. This decline started before the pandemic and continued after it. It isn’t a story about one bad year. It’s a longer erosion of the unstructured, low-stakes contact that used to happen without anyone scheduling it — the conversation at the bus stop, the regular at the diner, the neighbor on the porch.

Robert Putnam described this decline decades ago in Bowling Alone, tracking the fall in club membership, league bowling, and casual civic participation across the second half of the twentieth century. What he was really tracking was time: time spent around other people in settings that didn’t require an invitation. That kind of time has kept shrinking. Work has absorbed more of it, commuting has absorbed some, and screens have absorbed the rest.

The American Enterprise Institute’s 2021 survey on friendship found that 12% of Americans report having no close friends at all, up from just 3% in 1990. For men specifically, the drop is starker: those with six or more close friends fell from 55% in 1990 to 27% in 2021, and the share with zero close friends rose fivefold over the same period. That’s not a generation that stopped liking people. It’s a generation whose weeks filled up with other things.

The week has 168 hours. Most are already spoken for.

Say you sleep eight hours a night. That’s 56 hours gone. A standard working week, with a commute, easily takes another 45–50. Add basic maintenance — meals, errands, laundry, whatever keeps a household running — and you’re looking at maybe 40–50 hours left over. Some of that goes to a partner, kids, aging parents. Some goes to recovery, because being tired is not a moral failing, it’s a fact about your nervous system.

What’s left, for most people, is a handful of hours. Not fifty. Not a hundred. Maybe five to eight hours a week that aren’t already claimed. If friendship genuinely requires dozens of cumulative hours to take root, and you only have a few discretionary hours a week to spend on anything social at all, the math is unforgiving. At five hours a week, even the low estimate for a casual friendship takes ten weeks. A close one takes most of a year — and that’s if every discretionary hour goes to the same person, which it won’t, because you probably want more than one friend.

This is why casual advice to “put yourself out there” so often lands wrong. It treats the problem as a shortage of willingness. The actual shortage is of hours, and hours are not created by wanting them more.

Why remote work didn’t create more free time

There’s a version of this that seems like it should have gotten easier. No commute, in theory, frees up time. Gallup’s 2024 State of the Global Workplace found the opposite pattern on loneliness: fully remote employees report loneliness at 25%, compared with 21% for hybrid workers and 16% for those fully on-site. The hours saved on commuting didn’t reliably convert into hours spent with other people. They got absorbed back into work, or into nothing in particular. Saved time and social time are not the same thing. Time has to be pointed at another person, on purpose, or it just refills with whatever’s nearest — usually a screen.

This matters because it kills a common assumption: that if you just had more free hours, the social life would follow. Plenty of people have more unstructured time than they did five years ago and are lonelier than they were five years ago. The Harvard Making Caring Common survey from 2021 found 36% of Americans reporting serious loneliness, including 61% of young adults — a group that, by any measure, has more flexible schedules than someone raising three kids on a factory shift. Time alone doesn’t solve this. What you do with the scarce hours does.

What actually fits

If the discretionary hours are down to five or six a week, the only strategy that works is concentrating them, not spreading them.

Spreading them looks like: coffee with one acquaintance this week, a different one next week, an old college friend the week after, a work happy hour the week after that. It feels productive. It rarely produces a close friendship, because no single person is accumulating enough time with you to cross the threshold that turns “someone I know” into “someone I’d call.”

Concentrating them looks like: the same two hours, the same evening, with the same person or same small group, most weeks. A standing Tuesday. A monthly dinner that’s genuinely monthly and not “let’s find a time.” A running group you show up to on autopilot because it’s 8am Saturday and that’s just where you are now. This is less interesting to describe and far more effective. It’s the same principle Putnam was describing when he wrote about bowling leagues: not that bowling itself mattered, but that it was a fixed, recurring reason for the same people to be in the same room.

Five hours a week, spent on the same one or two relationships for ten weeks running, will get you further than fifty scattered hours spent on twenty different people over the same period. This isn’t a claim that quantity of contacts doesn’t matter at all — it does, for a broader sense of belonging. But for the specific ache of not having anyone close, repetition beats range.

The practical version

Pick one thing you can protect for the next six weeks. Not six months — six weeks is short enough to actually commit to and long enough to notice a difference. It should recur without needing to be re-negotiated each time: a class, a weekly call, a standing dinner, a group that meets whether or not you feel like it that day.

Then stop optimizing which one. The choice matters far less than whether it survives contact with a tired Tuesday. The friendship math doesn’t care how good the plan sounds. It only counts the hours that actually happened.