Pick two routes, put them in the calendar as repeating events, and choose them by one test: would you still go if you knew in advance that nobody you fancied would be there? Routes that pass that test survive a bad month. Routes that fail it are dating apps with a longer commute.
Why "just go out more" does not work
It fails for three reasons and all three are practical rather than psychological.
It has no fixed time, so it competes with your sofa every single week and eventually loses. It requires you to be charming on demand, which is the exact skill that is depleted after a working day. And it is usually attempted as a single event, one Saturday, treated as an experiment that either works or does not.
Meeting people is not an event. It is repetition, and the thing being repeated matters less than the repetition.
The three routes that actually produce introductions
**Repeated proximity.** Same people, same place, same time, weekly. A running club, five-a-side, a choir, a climbing wall, a class that runs a term. The mechanism is that you get to skip the introduction stage entirely. By week six you are not meeting strangers, you are talking to people you already recognise, which is a completely different social task and a far cheaper one.
**A role.** Volunteering, organising the fixture list, running the sound desk, doing the tea. A role gives you something to be in a room, which removes the hardest part of turning up alone. It also produces conversations with a subject, and conversations with a subject are the only ones that work when you are tired.
**The second degree.** The most underused route, and the cheapest. Everyone you already know has a network you have no access to, and almost nobody activates it, because activating it requires making a specific request out loud and most of us never do.
The specific ask
"Let me know if you know anyone" produces nothing, because it gives your friend a vague search over their entire address book, with no criteria and no deadline. Nobody actions that.
Try this instead, to three or four people, individually, not in a group chat:
"I've come off the apps and I'd rather meet people through actual humans. If you're doing anything where you're bringing a few people, would you invite me? And if you can think of one person I'd get on with, I'd rather be introduced badly than not at all."
Two things make it work. It asks for an invitation rather than a match, which is a much lower bar. And it gives permission for the introduction to be awkward, which is the objection that stops most people from making one.
Choosing your two
Two, not eight. Eight is a resolution and resolutions collapse by week three.
Run each candidate through four questions. Is it at a fixed time each week or fortnight? Are the same people there each time? Does it give you something to do with your hands or your voice, so you are not standing about being interesting? And would you go if nobody attractive were there?
That last question is the whole filter. A club you would attend anyway will still be in your week in November. A singles night is an app with a bar.
Beyond that, the specific activity matters much less than people think. Sport, music, making things, a language class, a faith group if that is yours, a weekly quiz, a repair café, a shift at a food bank. Pick by what you can sustain in bad weather.
The first four times you go
Almost everyone quits a new thing in the first month, and almost always for the same reason, which is that the first four visits are the worst four visits and nobody warns you.
**Visit one.** You will not speak to anybody beyond the practicalities and you will feel conspicuous. This is normal and it is not a sign about the group.
**Visit two.** Someone will recognise you. That is the entire achievement of week two, and it is more than it sounds, because recognition is what everything else is built on.
**Visit three.** You will have one short conversation, probably about the activity itself, and it will be unremarkable. Unremarkable is correct. Nobody makes friends through a great conversation on week three.
**Visit four.** This is the one where people usually decide whether to keep going, and they decide it based on how the first three felt rather than on any evidence about the fourth.
Two practical rules. Commit to six before you judge it, in the diary, so week four is not a decision. And arrive five minutes early rather than on time, because the five minutes before something starts is where all the talking happens.
Give it eight weeks, and judge it on the wrong thing
Judge it on whether you know three people's names by week eight. Not on whether you met anyone, not on whether anyone flirted with you.
Three names means the room has become familiar, and familiarity is the raw material everything else is built from. If eight weeks pass and you know nobody's name, that is usually a signal about the format rather than about you. Formats where everyone faces the front and leaves immediately do not produce names. Formats with a tea break, a pub afterwards or a shared task do.
The honest limits
None of this is evidence-based. There is no trial showing that joining a choir produces relationships, and there will not be one. It is slower than the apps, and for the first month it will feel like it is producing nothing at all, because it is.
What can be said is narrower and still worth saying. The research on dating apps consistently finds users reporting ["significantly worse psychological health and well-being"](https://asu.elsevierpure.com/en/publications/dating-app-use-psychological-health-and-psychological-well-being-/) than non-users, in studies that cannot establish direction. And in a randomised study of a different platform, four weeks of deactivation raised subjective wellbeing and ["caused a large persistent reduction in post-experiment"](https://www.aeaweb.org/articles?id=10.1257/aer.20190658) use. Neither of those tells you that a running club will find you a partner. What they support is the smaller claim that getting the time back is worth something on its own, and that filling it is what makes the break last.
If the apps were also your only social contact
Then the first month is going to be quieter than you expect, and you should plan for that rather than be ambushed by it. [The first two weeks off dating apps](/guides/dating_apps/the-first-two-weeks-off-dating-apps) covers the shape of it, including the Sunday evening problem. And if you are still going back and forth on whether to come off them at all, [the audit](/guides/dating_apps/are-dating-apps-making-things-worse-for-you) is the piece to read first.
Common questions
**How do people meet without dating apps?** Mostly through repetition and through people they already know. Both are slower than an app and neither requires you to perform.
**I work from home and have no colleagues. Where do I start?** Start with the fixed weekly thing, because working from home removes the incidental repetition that used to come free. One evening a week with the same people replaces what an office was quietly providing.
**Is it weird to go to things alone?** Common, not weird, and a role fixes most of the discomfort. Turning up alone with a job to do is easier than turning up alone to socialise.
**How long before this produces anything?** Longer than the apps, and nobody can give you a number. Judge the first two months on names rather than on outcomes.
**Can I do both?** Yes, and plenty of people do. If you go back to the apps, do it with rules rather than by default. [Dating on purpose, if you go back](/guides/dating_apps/dating-on-purpose-if-you-go-back) covers what those rules need to contain.
Sources
- [Dating app use, psychological health, and psychological well-being: a systematic review and meta-analysis, Computers in Human Behavior](https://asu.elsevierpure.com/en/publications/dating-app-use-psychological-health-and-psychological-well-being-/). 23 studies, 26,068 people, worse reported psychological health among users than non-users.
- [The Welfare Effects of Social Media, American Economic Review 2020](https://www.aeaweb.org/articles?id=10.1257/aer.20190658). Four weeks of deactivation raised wellbeing and persistently reduced later use of the platform.
- [Problematic Online Dating, Journal of Medical Internet Research, 2025](https://pubmed.ncbi.nlm.nih.gov/40608479/). 32 studies, only 3 randomised, and the concept of online dating addiction remaining debated.
