How Much Time Does Making Friends Actually Take

The honest question is not “how do I meet people.” It’s “where does this fit in a week that already has no gaps in it.” Most advice about making friends skips this part entirely, as if the only obstacle were nerve. It isn’t. The obstacle is time, and specifically the kind of time that has to be the same time every week, indefinitely, for it to work.

That’s the part worth being blunt about before comparing any tools: the research on adult friendship keeps coming back to repetition, not intensity. Robert Putnam’s work on the decline of American civic life was largely about people quietly dropping the recurring things — the bowling league, the union meeting, the church committee — not about people avoiding one-off events. One dinner with strangers doesn’t build a friendship. Ten dinners with the same five people, spread over ten weeks, might. So the real question for any app or route is: what does it cost you per week, and does it survive a week where you’re tired?

The one-off model: low cost per event, high cost over time

Timeleft and similar dinner-matching services ask very little of you upfront. You sign up, answer some questions, get seated with strangers at a set time. Timeleft reports having seated more than three million guests, which tells you the format works well enough to repeat at scale. Each individual dinner costs you an evening — two, maybe three hours, plus travel. That’s a manageable ask.

The catch is that you’re seated with different people most weeks, unless you and someone else decide to arrange something outside the app. Which means the actual friendship-building has to happen after the dinner, on your own initiative, with someone you’ve met once. That’s an extra step most people never take, not because they don’t want to, but because arranging a second meeting requires effort that the app itself doesn’t structure for you. The dinner is the easy part. The follow-up is the part that takes a week, not an evening, and it’s the part with no app behind it.

So the time cost of Timeleft is deceptively small per session and genuinely uncertain in total. If you go once a month hoping something sticks, you may be waiting a long time. If you go weekly and put in the work of following up with one person each time, the total time cost is closer to four or five hours a month, plus the harder-to-quantify effort of chasing a second meeting.

The recurring-group model: higher weekly cost, less decision fatigue

Meetup works on the opposite principle. You join a group built around something specific — a running club, a board game night, a book group — and the same slot recurs on the calendar whether or not you feel like going. Meetup has been running long enough, and has enough scale, that recurring groups are the default format rather than a novelty.

The time cost here is more honest: you’re signing up for a weekly or biweekly commitment, typically an hour and a half to two hours including the low-grade friction of getting there. That’s a real chunk of a week. But it removes a different cost entirely — the cost of deciding, each time, whether to go somewhere and meet someone new. Once the slot exists, the question isn’t “should I make plans this week,” it’s just “am I going to the Tuesday thing.” That’s a much smaller decision, and it’s one you can make even when you’re tired, which is exactly when the bigger decision usually loses.

This is the case for structure over willpower: a fixed time and place doesn’t ask you to be motivated. It only asks you to show up to something that’s already scheduled. Ray Oldenburg’s writing on third places — the informal, low-stakes gathering spots that used to structure ordinary social life — makes a related point: what made those places work wasn’t that people loved going. It’s that going required no invitation, no RSVP, no planning. You just went, because it was Tuesday and the place was open. A recurring Meetup group is a clumsier, more scheduled version of that, but it borrows the same logic.

The library and free-route option: lowest cost, slowest payoff

Public libraries, community centres, and parks departments often run recurring free programmes — a knitting circle, a chess night, story hour for parents. Eric Klinenberg’s research on social infrastructure argues that these physical spaces measurably shape how much informal contact people have, independent of what’s programmed inside them. They cost nothing beyond your time, and the time cost is usually modest — an hour, no membership fee, no app to manage.

What they don’t offer is matching. Nobody is filtering for compatibility or interest overlap the way an app does. You show up and take your chances on who else is there. For some people that’s fine, even preferable — less pressure, less performance. For others it means several visits before a single conversation goes anywhere. The time cost is low per session but the number of sessions before payoff is harder to predict, and there’s no algorithm doing any of the introducing for you.

What actually fits in a week

None of these is free of a cost, and none of them works if treated as a single event. The American Enterprise Institute’s 2021 survey on friendship found that the share of Americans with no close friends had quadrupled since 1990 — from 3% to 12% — and the number of men reporting six or more close friends had roughly halved. That didn’t happen because people stopped wanting friends. It happened as the recurring, low-effort structures — the league, the club, the regular happy hour — disappeared, and nothing recurring replaced them.

So the practical answer, if you’re deciding what to try this week: pick one thing you can commit to at the same time, weekly, for six weeks before judging it. Not because six weeks is magic, but because a single try tells you almost nothing, and repetition is the only variable that reliably shows up in the research. If your week has one spare evening, spend it on the recurring group, not the one-off dinner — the group asks less of you after the fact. If you have more flexibility but want lower stakes, a free library or community programme costs you nothing to try twice. If you want built-in matching and don’t mind doing the follow-up yourself, the dinner model works, but budget the extra hour for the text you have to send afterward. That text is the actual hard part, and no app sends it for you.