There is now a substantial and growing body of research on why Americans have fewer friends than at any point in modern history. There is a public health advisory. There are economists analyzing government time-use data. There are books, podcasts, think tanks, and a term — the "friendship recession" — specific enough to have its own Wikipedia-adjacent cottage industry of trackers and statistics pages.
Almost none of it mentions dating.
This is a strange omission, and worth sitting with for a moment. The loneliness epidemic and the difficulty of modern dating are, at the level of root cause, substantially the same story — the erosion of the social infrastructure that used to put people in front of each other. But they are covered as two separate problems by two separate sets of experts, in two separate genres of article, and the throughline between them rarely gets drawn.
It should be.
What the loneliness data actually says
In 2023, the U.S. Surgeon General issued a formal advisory declaring loneliness a public health epidemic, comparing its mortality risk to smoking roughly fifteen cigarettes a day. Roughly half of U.S. adults report measurable loneliness. This was not a pandemic-era spike that has since resolved — the underlying trend predates 2020 by close to a decade and has continued past it.
The specifics are worth sitting with. Time spent alone by the average American rose from 285 minutes a day in 2003 to 333 minutes a day in 2020 — the equivalent of gaining an entire extra day of solitude every month. Time spent specifically with friends fell by roughly 37% between 2014 and 2023, according to an analysis of federal time-use data by economist Bryce Ward. The share of Americans who report having no close friends at all has risen sharply since 1990, and the shift shows no sign of reversing.
Younger adults, not the elderly, are consistently the loneliest age group in every recent major survey — a finding that inverts the popular assumption that loneliness is primarily a problem of old age and social isolation late in life. Roughly a quarter of adults under thirty report feeling lonely most or all of the time, compared to a small fraction of adults over sixty-five. And the gap is not primarily about gender: recent research finds men and women report loneliness at nearly identical rates, though they tend to respond to it differently.
The infrastructure argument
The explanations offered for the friendship recession are consistent across researchers, even when the emphasis differs: fewer third places — the cafés, community centers, clubs, and shared public spaces where casual social contact used to happen by default. Longer commutes and more time spent working. The replacement of in-person contact with digital substitutes that resemble connection without reliably producing it. A broad, structural decline in the number of settings where meeting new people is incidental to the activity rather than the entire point.
This is precisely the same infrastructure that used to produce romantic connections.
The dinner party, the friend-of-a-friend introduction, the shared hobby group, the workplace before remote work thinned it out, the neighborhood before suburban sprawl and shrinking third places thinned that out too — these were never purely platonic systems. They were the ambient machinery through which people used to meet romantic partners, generally without trying to, as a byproduct of maintaining an ordinary social life. When that machinery declines, it does not decline selectively. It takes the friendships and the romantic introductions down together, because they were always running on the same infrastructure.
Dating apps were built, more or less explicitly, to replace this machinery with something engineered. For a while, they did — Match Group and its competitors built genuinely large businesses on exactly that premise. But an app can replace the transaction of introduction. It cannot replace the context that used to make an introduction trustworthy, low-stakes, and socially embedded — the fact that you met someone through a friend, at an event, in a room where your mutual social fabric provided some baseline information and some baseline accountability before either of you had said a word.
Why this matters more for the people who seem least likely to be affected
The instinctive assumption is that this is a problem for people with thin social lives to begin with — the isolated, the socially anxious, the people the loneliness research usually centers.
The more counterintuitive finding, both in the loneliness literature and in a decade of watching people date in person, is that the erosion of social infrastructure hits busy, high-functioning, professionally embedded people just as hard, for a different reason. It is not that they lack social skills. It is that their remaining social bandwidth gets allocated almost entirely to maintaining the relationships they already have — the demanding job, the existing friendships that require active upkeep to survive the general decline, the family obligations that don't disappear because a calendar is full. What gets crowded out first is exactly the category of loosely structured, low-stakes social contact that used to produce new connections, romantic ones included, without anyone having to schedule it on purpose.
This produces a specific and slightly absurd outcome: people who are, by any reasonable measure, functioning extremely well — socially capable, professionally successful, embedded in real friendships and real communities — who nonetheless find that the specific channel required to meet a romantic partner has quietly narrowed to almost nothing. Not because something is wrong with them. Because the ambient machinery that used to do this work for free, as a side effect of an ordinary social life, has been declining for over a decade, and nothing has organically replaced it.
What almost none of the loneliness coverage says out loud
Most loneliness research and reporting treats "romantic partner" as one category of relationship among several — friends, family, community, partner — worth mentioning but not worth isolating. This is defensible from a public-health standpoint, where the goal is aggregate social connection rather than any specific relationship type. It is less useful from the standpoint of a single professional trying to understand why meeting someone has gotten harder, because it obscures a genuinely different feature of the romantic category: unlike friendship, which can be maintained and even rebuilt through effort within an existing social circle, a romantic partner is disproportionately more likely to come from outside that circle. New introductions do a specific kind of work here that maintenance cannot substitute for.
This is why the friendship recession, correctly understood, is not adjacent to the dating difficulty story. It is largely the same mechanism, applied to a category of relationship that depends even more heavily on new introductions than friendship does — and gets almost none of the same public attention as a result.
What actually replaces the infrastructure that declined
The honest answer is not "go outside more" or "put yourself out there," advice that is true and close to useless, for reasons covered elsewhere on this site. Ambient social infrastructure does not get rebuilt by individual willpower. It gets rebuilt, when it gets rebuilt at all, by something deliberately standing in the gap that the third place, the friend group, and the neighborhood used to fill without anyone noticing they were doing it.
This is the specific thing a structured social evening is designed to be: not a replacement for a full social life, but a deliberately engineered substitute for the exact piece of social infrastructure that has eroded the fastest — the low-stakes, socially embedded, new-person encounter that used to happen by accident and now, for a large number of otherwise well-connected adults, essentially doesn't happen at all unless someone builds the room for it on purpose.
The loneliness researchers are right that the solution to a friendship recession is more community infrastructure, not more individual effort. The same logic applies, with more specificity, to the romantic version of the same problem. It was never really a dating problem. It was an infrastructure problem that happened to be easier to notice once it started showing up in people's love lives.
Relish hosts structured social evenings for driven professionals across 50+ cities in the US, UK, Canada and Australia since 2014. Find an evening near you →