The Coffee Shop Won't Fix What Remote Work Broke

A study published in Science last week put a number on something a lot of remote workers already suspected. Comparing 2011–2019 to 2022–2024, the researchers found that the shift to remote work explains roughly a third of the total increase in isolation and mental distress over that period. It shows up in mental health service use, in prescriptions, in time-use data. The starkest number: remote workers who live alone saw their days spent completely alone increase roughly tenfold compared with remote workers who live with someone else.

That last detail matters. It means the problem isn’t remote work in the abstract. It’s remote work plus an empty apartment. If you have a partner or a housemate, working from home just relocates your day. If you don’t, working from home can mean going from Monday morning to Friday evening without a single unplanned human interaction.

The standard advice for this is: go work from a coffee shop. It sounds like it should help. You’re around people. You get dressed. There’s a barista who might learn your order. So the question worth asking now is whether that actually does anything, or whether it just looks like doing something.

What a café gives you, and what it doesn’t

Ray Oldenburg, who coined the term “third place” back in 1989, was specific about what makes a café or bar function as one. It’s not proximity to other bodies. It’s regulars — the same people, recognising each other, building a low-stakes familiarity over repeated visits. A third place works because it accumulates a cast of characters over time, not because it’s full of strangers on any given afternoon.

Working from a café rarely gives you that. You sit with headphones on, facing your laptop, and the twelve other people around you are doing the same thing. You are physically present and socially absent — which, if anything, is a more efficient way of being alone in public than being alone at home. There’s ambient noise instead of silence. That’s a real thing; it changes how a day feels. It is not the same as contact.

This lines up with a distinction researchers have been drawing more carefully in recent years: workplace isolation and loneliness are not the same variable. A 2024 study of healthcare workers found they correlate but respond to different things — isolation is about the absence of interaction opportunities, loneliness is about the gap between the connection you have and the connection you want. Sitting in a busy café solves the first problem, superficially, without touching the second.

The office wasn’t actually about the office

It’s worth being honest about what remote work took away, because it wasn’t really the building. Gallup’s 2024 workplace data found that fully remote employees report loneliness at 25%, against 16% for people fully on-site — a real gap, but not an enormous one, and not one explained by square footage. What an office provided was structure you didn’t choose: the same faces, on a schedule you didn’t have to negotiate, doing something with mutual stakes. A 2025 study in the Journal of Affective Disorders found the loneliness association with remote work holds even accounting for job type and income, which suggests it’s the arrangement itself, not who ends up doing it.

A café gives you none of that structure. No one there has any reason to notice if you stop coming. That’s the actual test of whether a place is functioning as social infrastructure: does anyone register your absence? In a real third place, someone does, eventually. In most cafés, no one does, ever.

What would actually work, at the same low cost

None of this means cafés are useless — it means the version most people are doing wrong is the anonymous, headphones-in, different-café-every-time version. The fix isn’t more elaborate. It’s more repetitive.

Pick one place. Go at the same time, ideally two or three days a week, for at least a few weeks. Not a different café that’s slightly closer or slightly quieter — the same one. Sit in the same spot if you can. This isn’t about forcing conversation on day one. It’s about becoming a recognisable pattern in someone else’s day, which is the actual mechanism Oldenburg described and the thing a rotating cast of anonymous cafés can’t produce.

If a fixed café doesn’t do it, a coworking space with an assigned desk works on the same principle, with the advantage that the other regulars are also working alone and might be open to the same low-effort acknowledgement you are. Access to any of this has gotten harder — a 2025 study in Health & Place found third-place closures across every category between 2019 and 2021, concentrated in already under-resourced areas, so this isn’t purely a discipline problem. Where the options are thin, a scheduled recurring group does something a café can’t guarantee: it puts the same people in the same room on purpose. Meetup remains the largest free option built entirely around that model — join one group, go to the same meeting for a month, and see if it does more for you than four different coffee shops did.

The point of the Science study isn’t that remote work is unfixable. It’s that the fix has to replace what remote work removed — a predictable, repeated encounter with people who’d notice if you stopped showing up — and a laptop in a loud room, on its own, doesn’t do that.