Why Doing Something Beats Talking to Someone

Sit two strangers across a table and ask them to talk, and most people freeze. Put a paddle in their hands and a ball in the air, and they start talking without noticing they’re doing it. This is not a coincidence, and it is the whole case for meeting people through something you do rather than something you say.

The instinct when you’re lonely is to look for ways to meet people, which usually means looking for ways to talk to people. Dating-app-style friend apps, drinks with a stranger, a “conversation night” at a bar. These ask you to generate rapport out of nothing, on the spot, with someone who owes you nothing and you owe them nothing. It’s a hard way to build anything. A shared activity does something different: it gives you both a job. The job is the excuse for being there, the thing you’re looking at instead of each other, and — if it goes on for more than one session — the reason you’ll see each other again without having to invent one.

The research behind this, briefly

Sociologist Ray Oldenburg wrote about this before the internet made it feel like a new problem. He described “third places” — not home, not work, but the café, the bowling league, the barbershop, the pub with regulars — as the actual infrastructure of a social life. What made third places work wasn’t conversation for its own sake. It was that people showed up for a reason (a haircut, a game, a drink) on a schedule, and the talk happened around the edges of that.

Robert Putnam’s Bowling Alone traced the decline of exactly this kind of structure across the second half of the twentieth century — bowling leagues, union halls, civic clubs, all thinning out — and made a distinction worth keeping in mind: bonding social capital, the close ties within a group, tends to grow out of doing something together repeatedly, not out of one good conversation. Eric Klinenberg’s Palaces for the People pushes the same point further: libraries, parks, and public pools aren’t backdrops to social life, they’re the mechanism. The physical setup — a table, a court, a reading room — does work that willpower can’t.

There’s more recent, narrower evidence pointing the same way. A 2022 qualitative review of social prescribing — the practice of a doctor referring someone to a community activity rather than, or alongside, a treatment — found that structured, purposeful group activity produced more benefit than unstructured social contact on its own. People didn’t just feel less alone. They felt like they had a role again, something to contribute, which is a different and arguably deeper thing than having had a nice chat.

None of this says talking doesn’t matter. It says talking works better as a byproduct of doing than as the stated goal of the evening.

Meetup: the activity-first tool

Meetup is the obvious place to start, mostly because it’s the oldest and the largest — reportedly around 60 million members — and because its entire structure is built around interest, not chat. You don’t build a profile designed to be attractive to strangers. You join a hiking group, a board game night, a language exchange, a beginner’s pottery class, and the group exists whether or not you personally show up. That’s a genuinely different psychological position than a one-on-one app: you’re not the main event, the activity is, and that lowers the stakes considerably.

The free tier lets you join and create unlimited groups and events, which matters, because the alternative — a subscription gate before you know if there’s anyone in your area doing the thing you want to do — kills a lot of intent before it starts.

The downsides are real and worth naming plainly. Meetup groups vary wildly in quality and in whether they’re actually alive. A group with 400 members might have an organiser who hasn’t posted in eight months. Some groups skew heavily toward people already looking for something specific — dating, networking dressed up as “social” — which isn’t always obvious from the listing. And because anyone can start a group, there’s no floor on how well-run any given one is: some are warm and well-organised with the same dozen people every fortnight, others are a single burst of enthusiasm that fizzles after two meetups. You will have to try a few before one sticks, and trying a few means showing up to rooms full of strangers more than once, which is the actual cost of this approach and the reason people avoid it.

The non-app route: what a library or a running club actually offers

It’s worth saying that the most reliable version of this doesn’t require an app at all. A public library programme — a knitting circle, a chess night, a book club that meets in the same room every third Thursday — has almost everything a good Meetup group has, plus a physical location that isn’t going anywhere and doesn’t depend on one person’s enthusiasm holding up. Klinenberg’s argument about social infrastructure is really an argument for these places specifically: they’re subsidised, they’re free to use, and their whole point is repeat visits by the same people. A run club works on the same logic, with the added benefit that a shared physical task — pacing, breathing, finishing — produces the kind of low-stakes camaraderie that’s hard to fake in conversation.

The trade-off against an app is discoverability. Meetup will show you, sorted by distance and category, everything running near you this week. A library noticeboard or a local running shop’s group chat requires you to already be walking past the right place, or to ask around, which is exactly the kind of small social effort that feels disproportionately hard when you’re already isolated. Both routes work. One is easier to find; the other tends to be sturdier once found, because it isn’t dependent on a single volunteer organiser staying interested.

The thing that actually determines whether it works

Whichever route you pick, the finding that matters more than which tool you use is this: going once does very little. The value in an activity-based group isn’t the activity itself, it’s the fact that the same faces turn up week after week, which is the only real mechanism by which acquaintance becomes something more. A single hiking meetup with strangers is a pleasant Saturday. The fourth hiking meetup with the same eight people, where someone remembers you don’t like the steep trails and someone else always brings extra water, is the beginning of an actual social life.

This is where a lot of people go wrong with tools like Meetup: they treat it like a one-off event booking system, sampling a different group every week because no single one felt like “the one” immediately. Nothing feels like the one after one visit. Choose a group that runs on a fixed schedule — every Tuesday, not “we’ll post the date in the group chat” — and go four times before deciding anything about it. A fixed time and place removes a decision you’d otherwise have to make freshly, tired, on a Tuesday evening when staying in is easier.

This matters more than it sounds like it should. The share of Americans with no close friends at all has risen from 3% in 1990 to 12% now, and the drop has been sharpest among men, whose average number of close friends has fallen by nearly half over the same period. That didn’t happen because people stopped wanting friends or stopped knowing how to make small talk. It happened as the informal, recurring structures — leagues, clubs, regular haunts — thinned out from underneath everyone. An activity-based group, chosen for its schedule rather than its charm, and attended on a schedule rather than on a mood, is a small way of rebuilding exactly that structure, one Tuesday at a time.