The Friendship Recession: What the Data Says, and the Unglamorous Fix

August 10, 2026 · 8 min read · by Marvin Blase

A while ago I worked out how often I actually saw one of my closest friends. Not how often it felt like — how often it really was. The answer was four times in a year. We'd known each other for fifteen years, I'd have told anyone he was one of my best friends, and we managed roughly one evening a quarter.

That gap between how close a friendship feels and how much time it actually gets is what people mean by the friendship recession. It's a real, measured decline, and the numbers are less dramatic and more interesting than the headlines suggest. I went and read the primary sources, because most of what circulates online is a misquote of a misquote. Here's what they actually say — and what I think the honest response is.

What the friendship recession actually is

The clearest measurement comes from the American Time Use Survey, a US government survey that asks people to account for their day. A 2023 analysis of it by Kannan and Veazie, quoted in the Surgeon General's advisory on loneliness, is the source of most of the numbers you've seen:

Two of those deserve a second look.

In-person time with friends fell by two thirds — from an hour a day to twenty minutes. That's the 20-hours-a-month figure the advisory highlights, and it's the one I find hardest to argue with, because it isn't asking people how lonely they feel. It's asking them what they did yesterday.

The share of Americans with no close friends quadrupled, from 3% to 12%. Meanwhile the share reporting three or fewer close friends went from about a quarter to about half. The middle of the distribution didn't shift slightly — it emptied out.

The most recent data point is the newest: in the 2025 ATUS results, released in June 2026, 30% of American adults socialised or communicated on an average day, down from 38% in 2015. Average time spent on it fell from 41 minutes to 35. Over the same decade, time spent on games and leisure computer use rose from 25 to 37 minutes a day. I'd be careful about reading that as a straight trade — the survey doesn't establish that one caused the other — but the two lines are going in opposite directions and it's hard not to notice.

It started long before the pandemic

This is the part that gets lost. It's tempting to file the friendship recession under "things COVID did," and the data doesn't support that. The decline runs from 2003, steadily, through the whole period. Time spent alone had already climbed from 285 to 309 minutes a day between 2003 and 2019 — before 2020 added another 24 minutes on top.

The pandemic accelerated an existing trend. It didn't start it. Which matters, because if you believe it was a lockdown effect you'd expect it to reverse on its own, and there's no particular reason to think it will.

Researchers treat this as a health question rather than a lifestyle one, and the reason is blunt: the Surgeon General's advisory puts the increased risk of premature death from social isolation at 29%, and describes lacking social connection as comparable in risk to smoking up to 15 cigarettes a day. I'm not going to pretend an app addresses that. I mention it because it's the reason serious people study this at all.

Why remembering people got harder

Here's the mechanism I think explains the everyday experience of it, and it's the part I care about most.

When you see someone every week, you don't have to remember anything about them. You don't need to recall that their mother was having tests done, because you'll see them on Thursday and they'll tell you the result unprompted. The friendship stays current by itself. Frequency does the work.

At four meetings a year, that stops being true. You arrive with a nine-month gap to cover, and the details you were carrying have gone soft — was it her sister or her cousin who moved to Lisbon? Did the job thing happen? You end up doing the conversational equivalent of a status meeting, and the specific, slightly odd, actually-personal stuff never surfaces, because you can't remember enough of it to ask.

So the friendship recession isn't only a decline in hours. It's the loss of the ambient context those hours used to generate for free. And that's a much smaller problem, mechanically, than "everyone should see their friends more." You can't easily manufacture six extra hours a month. You can stop losing the details.

The unglamorous fix: fewer people, more detail

The advice this data usually generates is "reach out to a friend today," which is fine and lasts about a day. What's actually worked for me is narrower and duller:

1. Pick a small number of people. Not your contact list — the handful whose lives you genuinely want to keep track of. For most people it's somewhere between five and fifteen. Trying to maintain everything is how you end up maintaining nothing.
2. Write down what they tell you, within about a minute. Not impressions — specifics. "Her dad's surgery is on the 14th" survives nine months. "Had a good chat with Kim" does not. I've written a whole post on what's actually worth writing down about people, because getting this wrong is the most common failure.
3. Follow up on exactly one thing. The single highest-return move in a long-distance friendship is asking about the specific thing they mentioned last time. It costs nothing and it's unmistakable evidence that you were listening.
4. Protect the dates. Birthdays are the one recurring, zero-effort excuse to make contact, and they're also the easiest thing to lose when your old system (usually Facebook) quietly stops running. Here's the system I use.

None of that increases the hours. It increases what each hour is worth, which is the variable you can still control.

I'll be transparent about my bias: I build Remember Them, an iOS app for keeping exactly these details, and I built it because of the four-times-a-year problem above. But this genuinely doesn't require an app. A note per person in whatever notes app you already have does the same job — the format matters far more than the tool, and I'd rather you do it badly in Apple Notes than not at all. If you want the longer version of the technique, it's in how to remember details about people.

What this doesn't fix

Being honest about the limits, because a data post that ends in a neat solution is usually lying:

• Detail is not company. Remembering someone's news is not the same as being in the room. If you're lonely, better notes are not the answer — more contact is, and that's harder.
• It's asymmetric. You can only control your side. Some friendships fade because the other person has stopped putting anything in, and no system fixes that.
• The structural causes are structural. Longer commutes, fewer third places, work hours, moving for jobs — the ATUS analysis flags hours worked as a genuine constraint on social connection. Individual habits don't touch that.

What remains true is that the cheapest act of care available to most of us is remembering something specific and asking about it later. It doesn't scale, it isn't impressive, and it works.

Frequently asked questions

What is the friendship recession?

It's the measured, long-run decline in time spent with friends and in the number of close friendships people report. In the American Time Use Survey, in-person time with friends fell from about 30 hours a month in 2003 to about 10 hours a month in 2020, while the share of Americans reporting no close friends rose from 3% in 1990 to 12% in 2021.

How much time do people actually spend with friends now?

Around 10 hours a month of in-person time as of 2020, down from 30 hours in 2003. In the most recent ATUS results, for 2025, only 30% of American adults socialised or communicated at all on an average day, averaging 35 minutes — down from 38% and 41 minutes in 2015.

Did the pandemic cause the friendship recession?

No — it accelerated it. The decline in time spent with friends and the rise in time spent alone are both visible in the data from 2003 onwards, well before 2020. The steepest drop in time with friends happened in the years before the pandemic, not during it.

Which age group has been hit hardest?

Young people. For ages 15–24, in-person time with friends fell by close to 70% over roughly two decades — from about 150 minutes a day in 2003 to 40 minutes in 2020. That's a far steeper decline than for the adult population as a whole.

Can an app fix the friendship recession?

Not really, and be suspicious of anything that claims otherwise. What a notes app or a personal CRM can do is stop you losing the details between increasingly rare meetings, so that the time you do get is spent on real conversation rather than reconstructing the last nine months. That's a genuine but modest benefit.

Sources

All figures checked against the primary sources on 10 August 2026.

• Office of the U.S. Surgeon General, Our Epidemic of Loneliness and Isolation: The U.S. Surgeon General's Advisory on the Healing Effects of Social Connection and Community, 2023 — HHS. Time-with-friends, time-alone, close-friend and mortality figures.
• Kannan, V. D. & Veazie, P. J., "US trends in social isolation, social engagement, and companionship — nationally and by age, sex, race/ethnicity, family income, and work hours, 2003–2020", SSM – Population Health, vol. 21, 2023 — doi:10.1016/j.ssmph.2022.101331. The underlying ATUS analysis the advisory cites.
• U.S. Bureau of Labor Statistics, American Time Use Survey — 2025 Results, USDL-26-1022, released 25 June 2026 — bls.gov. The 2015-vs-2025 socialising figures.
• Cox, D. A., The State of American Friendship: Change, Challenges, and Loss, Survey Center on American Life, June 2021 — americansurveycenter.org. Close-friend counts; survey of 2,019 US adults, fielded 14–23 May 2021.