The maintenance math
Friendships are built in hours, not intentions — but how many hours, and what keeps each layer alive once it exists? This page walks through the famous "200-hour study," the kinship discount, and why your social time budget is fixed even when your calendar feels full. The parent topic names the architecture; this is the cost schedule.
What the evidence supports
- In the origin study, acquaintance-to-friend took roughly 50 hours together, and close friendship took 200-plus (Journal of Social and Personal Relationships, 2019).
- Time spent together — and contact frequency — is the strongest measured predictor of emotional closeness (Personal Relationships, 2011).
- Everyone concentrates attention on a small inner circle of around 15 people regardless of total network size (Social Networks, 2013).
What remains uncertain
- The hour thresholds are central estimates from two U.S. samples; individual pairs varied widely around them.
- Whether the quality of hours — deep conversation versus parallel activity — shifts the cost has been less studied.
- How much digital contact substitutes for in-person hours remains unsettled.
Evidence last reviewed: August 15, 2026. Conclusions may change as new research is published.
friendships are built in hours
The 200-Hour Finding, Walked Through
In 2019, communication researcher Jeffrey Hall published the study behind the number everyone quotes. He tracked two samples over time: 355 adults who had relocated within the previous six months, and 112 university freshmen — people whose social slates were recently wiped, so new friendships could be watched forming from zero. Participants logged hours together and rated closeness at intervals. The pattern that emerged: moving from acquaintance to casual friend took roughly 50 hours of shared time; casual friend to friend, roughly 90 hours; and friend to close friend, more than 200 (Journal of Social and Personal Relationships, 2019). Two caveats keep this honest. First, the spread was wide — some pairs converged quickly, others logged a year of contact and stayed casual, so hours are necessary but not sufficient. Second, this is one research group's estimate from two U.S. samples. Treat the numbers as calibration, not law: friendships are a function of time, and the coefficient is large. Put differently: if you and a new acquaintance share one standing evening together each week, the arithmetic says casual friendship arrives around the one-year mark and close friendship somewhere in year four — which matches the lived experience of book clubs, running crews, and office cohorts almost exactly.
Why Hours Predict Closeness
The mechanism is bluntly behavioral: emotional closeness tracks time invested and contact frequency, in self-report and in communication records (Personal Relationships, 2011). The same analysis found the fascinating exception that explains a lot of family life: kin run on a maintenance discount. Relatives reach the same closeness with less ongoing contact than friends do, because the relationship's history — decades of shared context — is already banked. Friendship gets no such credit. A friend you haven't seen in three years is not automatically your close friend anymore; a cousin you haven't seen in three years usually still is. The practical reading: friendship must be re-earned on a schedule, which is exactly why the calendar matters more than affection. Affection is abundant; scheduled hours are scarce. Which is why the most predictive detail in this research is unglamorous: the people who stay close are the people who share a recurring calendar event, whether they feel like it or not.
The Time Budget Is Fixed
The second piece of the math is a constraint, not a number to hit. Phone-record analyses show that people cap their active attention at a small inner circle — around 15 people — no matter how many contacts sit in the phonebook (Social Networks, 2013). The budget, not the address book, sets the network. Even heavy social-media users show the same ceiling in their meaningful networks (Royal Society Open Science, 2016). The implication for anyone trying to "have it all": depth is zero-sum across the network. Adding a new close friendship means the hours come from somewhere — usually an existing middle-ring tie that quietly fades. This is the portfolio view the parent topic calls rebalancing: your 30s and 40s don't shrink your network by malice; they shrink it by budget. The good news inside the constraint: because the budget is fixed, maintenance gets easier as you standardize it — the person who keeps five standing events on the calendar outsources the budget problem to the calendar itself, and the hours arrive without negotiation.
Your Social Signature Has Slots
A striking follow-up: when the cast of your network changes — a new job, a new city — the shape of it doesn't. The "social signature," the distribution of your calls across people, stays remarkably stable over years even as specific names rotate through (PNAS, 2014). People who gave most of their attention to five others keep doing so; the five just become new people. The honest reading has two edges. The hopeful edge: a faded friendship leaves a slot that will refill — connection-seeking is the default, not the exception. The cautionary edge: the slot refills with whoever is present, not necessarily whoever you'd choose. Deliberate maintenance is the difference between choosing your slots and inheriting them, which is the entire thesis of Building Your Tribe. The slots are also why the "we should catch up" friendship deserves more respect than it gets: it is a slot in waiting, one message from reactivation — the reconnecting page's entire subject.
What Maintenance Actually Costs, Cadence by Cadence
Formation is hours; maintenance is cadence. The maintenance literature identifies four behaviors that keep friendships alive — positivity, supportiveness, openness, and interaction (Journal of Social and Clinical Psychology, 2004) — and the cadence question maps onto the Dunbar layers surprisingly cleanly:
- 🫂 Inner five: near-weekly touch — a call, a meal, a shared run. This ring runs on responsiveness, and silence here registers within weeks.
- 😢 The fifteen: contact every two to four weeks keeps warmth intact; monthly is the floor, quarterly starts to cool.
- 👥 The fifty: a few meaningful touches a year — the birthday call, the annual visit, the "we should catch up" that actually gets scheduled.
- 🌐 The 150: once a year still works; holiday cards and milestone congratulations keep ties revivable indefinitely.
- ☁️ The cloud: no individual maintenance — recurring venues (the gym, the café, the dog park) do the work for you, as the weak-ties page explains.
| Layer | Maintenance cadence that keeps it alive | Evidence |
|---|---|---|
| 🫂 Support clique (~5) | Near-weekly contact; responsiveness is the whole job | Strong |
| 😢 Sympathy group (~15) | Monthly-ish contact; container-based contact works well here | Moderate |
| 👥 Friendship circle (~50) | A few meaningful touches a year; annual rituals do real work | Moderate |
| 🌐 Active network (~150) | Annual contact keeps ties revivable | Moderate |
| ☁️ Weak-tie cloud (500+) | No individual maintenance; recurring venues suffice | Limited |
📅 The calendar is the currency
Intentions do not convert to hours — schedules do. The single highest-leverage move in this whole topic is containerizing contact: a standing call, a league night, a monthly supper club. A recurring container delivers the 50 or 200 hours automatically, without anyone having to feel like they're "working on the friendship." The Habit Formation protocol's implementation-intentions page has the mechanics; this page supplies the arithmetic.
Questions, Answered Briefly
- ⏱️ Do online hours count toward the 200? Partially. Hall's data tracked in-person time; digital contact sustains a tie but converts to closeness less efficiently — the parent topic's digital-layer section has the honest mapping.
- 🕰️ How long until a friendship fades? There's no fixed shelf-life — drift is slow and usually reversible for years, and dormant ties revive surprisingly well, as the reconnecting page shows.
- 👨👩👧 Do family ties need the same hours? No — the kinship discount means relatives maintain closeness with less contact, because shared history is pre-banked.
- 📆 I genuinely have no time. What's the floor? Containerized contact: one standing event beats five intentions, because the event delivers hours without willpower. Protect one; build from there.
The Bottom Line
- Friendship is a function of hours. Roughly 50 moves an acquaintance to casual friend, 90 to friend, 200-plus to close — averages with wide spread.
- Family gets a discount; friends don't. Kinship is pre-paid by shared history, which is why friendships need a schedule.
- The budget is fixed. Attention concentrates on an inner circle of about 15 no matter your phonebook size — depth is zero-sum.
- Cadence scales with layer. Weekly for the inner five, monthly for the fifteen, annual for the 150 — and containers beat willpower at every level.
Related Topics
- Hall, "How many hours does it take to make a friend?" Journal of Social and Personal Relationships (2019)
- Roberts & Dunbar, "Communication in social networks: effects of kinship, network size, and emotional closeness," Personal Relationships (2011)
- Miritello, Lara, Cebrián & Moro, "Time as a limited resource: communication strategy in mobile phone networks," Social Networks (2013)
- Saramäki et al., "Persistence of social signatures in human communication," PNAS (2014)
- Dunbar, "Do online social media cut through the constraints that limit the size of offline social networks?" Royal Society Open Science (2016)
- Oswald, Clark & Kelly, "Friendship maintenance: an analysis of individual and dyad behaviors," Journal of Social and Clinical Psychology (2004)