The University of Kansas communication researcher Jeffrey Hall set out to answer a deceptively simple question in 2018: how many hours, in concrete terms, does it actually take to make a friend? The question had been asked before in informal contexts but had never been pinned down to numbers with any methodological rigour. Hall designed two complementary studies. The first, an online survey of 355 adults who had relocated to a new city within the previous six months, asked participants to think of someone they had met in their new location and to estimate the number of hours they had spent together since meeting. The participants then categorised the relationship on a friendship-closeness scale ranging from acquaintance through casual friend to friend, good friend, and close or best friend. The second study followed 112 first-year University of Kansas students who had recently moved to Lawrence, tracking the development of their new relationships across nine weeks through three successive surveys.
The results, published in the Journal of Social and Personal Relationships in 2019, produced numbers cleaner than Hall had expected. According to the University of Kansas news announcement accompanying the paper, Hall found that the probability of two adults transitioning from acquaintance to casual friend exceeded 50 percent at roughly 50 hours of time spent together. The transition from casual friend to friend required approximately 90 hours. The transition from friend to good or close friend required approximately 200 or more hours, with the regression analysis pinpointing a 50-percent transition probability at 219 hours.
What “time together” actually means
The hours that Hall counted were not all alike. His analysis treated different categories of interaction as contributing to friendship in different ways. According to the full text of the 2019 paper, leisure activities — hanging out, joking, sharing meals, playing games, going to events — contributed substantially to relationship closeness, while time spent working together produced much weaker friendship growth even when the hours were extensive. Many participants reported spending hundreds of hours alongside colleagues without those colleagues ever crossing the threshold into actual friendship. The implication is that workplace proximity is not, in itself, a sufficient condition for friendship formation, even though it is one of the most common contexts in which adults are with other adults.
The kind of conversation that occurred during the time together also predicted relationship growth. Hall distinguished between “small talk” (greetings, weather, work logistics), “catching up” (asking what the other person has been up to), “meaningful talk” (sharing personal information, opinions, feelings), “joking around,” and several other categories. Meaningful talk and joking around were the two strongest predictors of friendship development over time. Small talk, while ubiquitous, did relatively little. The combination of sufficient hours and the right kind of conversation during those hours produced friendships; the same hours filled with small talk and logistics did not.
Why adults find friendship harder
The numbers help explain a pattern that has been observed informally in social science research and popular commentary for decades: making close friends becomes substantially harder after college. According to a Psychology Today summary of the Hall research, the implication is structural rather than psychological. College students are in a near-uniquely time-rich environment for friendship formation. They live in close proximity to large numbers of similarly-aged peers, share an institutional routine that puts them in regular contact, have substantial unscheduled leisure time, and are at a life stage where openness to new relationships is high. Hall’s second study, which followed students through their first nine weeks at Kansas, found that good and close friendships could form within that nine-week window precisely because the students were spending many hours per week together — sometimes a third of all waking hours per month with a single eventual close friend.
Adults working full-time jobs, raising children, maintaining households, and managing professional and family obligations do not have anywhere near this kind of time available for new relationships. A typical adult might spend 5 to 10 hours per week with any single new acquaintance, even one they actively like and want to befriend. At that rate, reaching 50 hours of total contact takes 5 to 10 weeks. Reaching 90 hours takes about 3 months. Reaching 200 hours takes 6 to 12 months or longer, and that is assuming the contact is sustained and consistent across the whole period. Most adult acquaintanceships do not survive that kind of sustained low-frequency contact without something concrete to anchor them — a shared workplace, a children’s school, a shared hobby with regular meetings, a fitness routine.
What the Dunbar framework adds
Hall’s research builds on a broader framework developed by the Oxford evolutionary psychologist Robin Dunbar. According to the Chronicle of Evidence-Based Mentoring’s review of the Hall study, Hall extrapolated his work from Dunbar’s earlier findings on social network limits — the well-known “Dunbar number” suggesting that humans can maintain stable social relationships with approximately 150 people, organised into successive concentric circles of intimacy. Dunbar’s typical breakdown places about 5 people in an individual’s innermost circle (closest intimates), about 15 in the next layer (close friends), about 50 in the next (friends), and about 150 in the outermost layer (acquaintances). Each layer requires progressively less time investment, but the inner layers require substantial cumulative time to maintain.
The Hall findings essentially provide empirical hour-counts for the transitions between Dunbar’s layers. Moving someone from the outer “acquaintance” layer into the “casual friend” layer takes about 50 hours. Moving them into the “friend” layer takes about 90. Moving them into the inner “close friend” layer takes more than 200. The inner-most “intimate” layer, which Hall did not directly measure, presumably takes substantially more again. The cumulative time investment required to maintain a full Dunbar network — five intimates, fifteen close friends, fifty friends, 150 acquaintances — is large enough that most adults do not, in practice, maintain the full set. They keep some inner layers full at the expense of others, drop the outer layers entirely, or live with sparser social networks than the evolutionary architecture of the brain is designed for.
What this implies for new adult friendships
Hall has been careful in subsequent interviews not to make the hour-counts prescriptive. The 50/90/200 figures are statistical thresholds derived from population-level analysis, not contractual requirements that every individual friendship must meet. Some friendships consolidate faster, particularly when both parties have strong shared interests, are in a high-intensity context together (a wilderness trek, a startup, a long-distance move), or simply find each other unusually compatible. Others take longer. The figures describe the average trajectory rather than a fixed path.
What the figures clearly establish is that friendship is not a matter of compatibility alone. It is a matter of compatibility plus cumulative time. Two adults who like each other immediately and would, with sufficient time, become close friends will not actually become close friends if they cannot find 200 hours to be together over a reasonable period. The compatibility is necessary but not sufficient. The time is necessary too, and the time is what most adult lives are short on. The Hall study is in this sense more diagnostic than prescriptive — an explanation of why adult friendship is hard rather than a recipe for making it easier. The recipe, to the extent the data supports one, is straightforward and difficult: find people you like, then spend an unreasonably large number of hours with them, doing things you both enjoy, over a period of months. Hall’s own summary in the original 2018 announcement was succinct. “We have to put that time in,” he said. “You can’t snap your fingers and make a friend.”
