What is Dunbar's number?
Dunbar's number is the proposed limit on how many stable social relationships one person can realistically maintain at once, usually cited as approximately 150. It's named after British anthropologist Robin Dunbar, who proposed it in the early 1990s after finding a correlation between the size of a primate species' neocortex, the part of the brain handling complex thought, and the typical size of that species' social group. Extrapolating the human neocortex size onto that same relationship produced a predicted group size of around 150.
That number was never meant to describe people you vaguely recognize or exchange small talk with. It describes relationships stable enough that you'd know who someone is, roughly where they fit in your life, and could resume a conversation with them without reintroducing yourself, the kind of network historically reflected in things like the size of hunter-gatherer clans or a realistic holiday card list. It's also directly relevant to a question I get asked a lot: how many people actually belong in a friendship journal.
The circles inside Dunbar's number
Dunbar's number isn't one flat number. It's a nested set of layers, each roughly three times the size of the one before it, with each smaller circle sitting entirely inside the larger ones.
| Circle | Approx. size | Who's in it |
|---|---|---|
| Support clique | 5 | People you'd call at 3am in a real crisis, no hesitation |
| Sympathy group | 15 | Close friends you confide in and see regularly |
| Affinity group | 50 | Good friends, people you'd invite to a big gathering |
| Dunbar's number | 150 | Stable relationships you could pick back up without awkwardness |
| Acquaintances | 500 | People you recognize and could loosely place |
| Recognizable faces | 1,500 | Faces and names you could match, without a real relationship behind it |
The exact figures shift slightly depending on which study you read, but the shape is consistent: a small handful of people who need constant attention, widening out to hundreds who need almost none.
Is Dunbar's number actually real?
It's a genuinely influential idea, but the specific figure of 150 is more contested than its popularity suggests. The original correlation was drawn from a small dataset of primate species and extrapolated to humans with wide statistical uncertainty built in from the start. A 2021 re-analysis of that same data by researchers Patrik Lindenfors, Andreas Wärtel, and Johan Lind redid the statistics and argued the confidence interval for the human number could plausibly run anywhere from about 2 to well over 500, which would make 150 a rough midpoint rather than a precise ceiling.
What's held up better across separate lines of research, phone-call records, social network analyses, actual holiday card lists, is the layered structure itself: people really do seem to organize relationships into a small inner circle, a somewhat larger group of close friends, and a much bigger outer ring, regardless of exactly where the outer ring's edge falls.
What this actually means for staying in touch
The practical takeaway isn't the number 150. It's the shape. The smaller the circle, the more attention it needs, and for most people the real problem isn't running out of room somewhere near 150, it's letting the innermost 5 to 15 people slide while spreading attention thin across a much bigger list.
That's a different problem than the one most contact-management tools are built to solve. A large network you want software to track automatically calls for auto-sync and enrichment, since manually logging hundreds of people isn't realistic. But the 5 to 15 people who actually make up your support clique and sympathy group don't need automation. They need someone to actually remember the surgery date they mentioned in passing, or notice it's been three months since you last talked. That's a habit problem, not a data problem, and it's solved by making the noticing and the logging fast enough that you actually keep doing it.
Quick answer: Dunbar's number is usually cited as about 150 stable relationships, nested inside smaller circles of roughly 5, 15, and 50. The exact ceiling is disputed. What's worth acting on is the layered structure itself, and the fact that your closest 5 to 15 people need far more deliberate attention than the other 135.
FAQ
What is Dunbar's number?
The proposed limit on how many stable relationships one person can maintain at once, usually cited as about 150. It comes from anthropologist Robin Dunbar's work correlating neocortex size with social group size across primates, extrapolated to humans.
Is Dunbar's number scientifically proven?
Not proven outright. The original estimate carries wide statistical uncertainty, and a 2021 re-analysis argued the true range could run from about 2 to several hundred. The layered structure of nested circles is better supported than the specific number 150.
How many close friends should a person have?
Dunbar's model puts the innermost support circle at about 5 people and the wider circle of good friends at about 15. There's no universal right number, but most people's real limitation is attention to that inner 5 to 15, not the 150 ceiling.
Why do people say you can only have 5 close friends?
That 5 comes from Dunbar's innermost layer, the support clique: people you'd call in a genuine crisis without hesitation. It's the smallest, most demanding circle in a model that also includes wider circles of 15, 50, and 150.
Where BuddyLog fits
If Dunbar's model is right about anything, it's that the people worth tracking closely are a short list, not a long one. BuddyLog is built around that inner circle: the 5, the 15, maybe the 50, logged by hand because you decided they were worth remembering, not auto-synced because your inbox happened to mention them. You don't need software to manage 150 relationships. You need something that makes it easy to actually keep up with the dozen or so that matter most. For more on what that looks like in practice, see the full guide to keeping a friendship journal.
