You Can Only Have So Many Real Friends. Here’s Why That Matters.

150 is the max

We are cognitively limited to about 150 relationships. Most of us are squandering the ones that matter most and focusing on the wrong things in life, leading to worse outcomes and happiness.

Robin Dunbar did not set out to produce one of the most widely cited numbers in social science. He was a primatologist studying the relationship between neocortex size and social group size across different species of primates, and he noticed a consistent pattern: the larger the neocortex relative to total brain volume, the larger the stable social group the species could maintain.

When he applied the relationship to humans the answer was approximately 150. That is, on average, the number of people with whom a human being can maintain stable, meaningful social relationships. These are the people you know personally, whose relationship to you you understand, and whom you would recognise and greet if you encountered them unexpectedly.

This number has held up remarkably well across a wide range of independent sources of evidence. Hunter-gatherer bands average around 150 members. Neolithic villages appear to have clustered around the same size. Military units from the Roman legions through to modern armies have independently settled on the company as their primary unit of cohesion (typically around 100 to 150 soldiers). The Hutterite communities that practice radical social equality split their settlements at around 150, on the explicit grounds that above that size the community loses coherence. Christmas card lists typically average around 150 names. The convergence is striking, and it points to something real about the cognitive architecture of human sociality.

The Layers of Friendship

Dunbar’s number of 150 is better understood not as a single limit but as the outer boundary of a set of concentric circles, each defined by a different quality and frequency of relationship.

• The innermost circle: contains approximately five people: your intimates, the people you are in daily or near-daily contact with, who you would turn to first in a crisis, and with whom you share the deepest mutual knowledge and trust. This is the support clique, and most people, when they think honestly about their lives, find that five is approximately right.
• The next circle: expands to about 15 people in total. The addition people here are your close friends. These are people you see roughly weekly, who you would drop plans to help, and whose lives are genuinely familiar to you in detail.
• The penultimate circle: expands to include about 50 people in total. The additional people here are people you would describe as friends rather than acquaintances. These people you are pleased to see at a party, who you have specific warm memories with, and who you are in monthly contact with.
• The outermost circle: takes your total connections to about 150. The additional folks here are your full sympathy group. These are addition people whose deaths would genuinely affect you, who you would help without expecting anything in return, and whom you know as real individuals rather than as names.

These numbers are averages with genuine variation around them. Extroverts tend to have slightly larger circles; introverts slightly smaller. There is evidence of consistent sex differences in the architecture of the inner circles, with women tending to maintain a smaller number of very intense intimate relationships and men maintaining slightly more but somewhat less intense ones. But the order of magnitude is consistent, and the structural logic appears to hold across populations and cultures.

Why the Numbers Are Surprisingly Small

The numbers feel small against the background of contemporary social life, in which most people have hundreds of contacts in their phones, thousands of connections on professional networks, and social media feeds that create the ambient impression of a vast and active social world. The contrast is not incidental. It is one of the central tensions of the social media age: the inflation of apparent social connection against the maintenance of actual social relationships.

Dunbar’s circles are not about familiarity or recognition. They are about the cognitive and emotional investment required to maintain a genuine social relationship, one in which you know the person’s current life circumstances, their relationships, their recent experiences, and the history you share. That kind of knowledge is expensive. It requires time, attention, and regular contact to keep fresh. It decays when contact lapses. And its maintenance competes with every other cognitive demand in a life that is already saturated with information, obligation, and distraction. (We’ve written elsewhere about why it gets harder to make and maintain friends as you get older, and this is part of the reason that is the case).

The cognitive limit is not simply a function of how many people you can remember. It is a function of how much relationship-maintaining mental labour your brain can sustain. Every close relationship requires ongoing social attention: tracking where the person is in their life, updating your model of them as they change, maintaining the reciprocity of contact and care that signals that the relationship is valued. Above approximately 150 relationships, the maintenance demands exceed what human cognitive architecture can sustain. Beyond that number, relationships start to feel thin, and the investment required to thicken them competes directly with the investment already going into the relationships within your active circles.

The Investment Relationship Requires

One of the less comfortable implications of Dunbar’s work is the finding that close relationships decay without active investment, and that the rate of decay is faster than most people assume. Research on friendship maintenance suggests that the threshold for maintaining a friendship at the close-friend level (the circle of 15) is approximately monthly in-person contact. Miss a month and the relationship begins to drift towards the outer circle. Miss six months without any significant contact and a formerly close friend may have effectively migrated to the acquaintance circle, requiring considerable reinvestment to restore.

This has direct implications for the way modern life is structured. Geographic mobility moves people away from the friends they have built relationships with, and the activation energy required to maintain those relationships at a distance is high. Career transitions change the professional social context in which many adult friendships are embedded. Life stage transitions (parenthood especially) create periods of severe time pressure in which the maintenance of friendships outside the immediate family is the first thing to be sacrificed. Each of these transitions tends to shrink the inner circles, and restoring them requires conscious effort that most people, most of the time, do not make.

The implication is that friendship is not something that simply persists if neglected. It is something that requires regular, active maintenance. The common experience of drifting away from people you were once close to is not a failure of character or affection but a predictable consequence of the maintenance demands of close relationships combined with the pressures of contemporary life. This has a corollary that is both useful and slightly dispiriting: if you want to have five genuine intimates at sixty, you need to have been actively investing in those relationships throughout your forties and fifties. The relationships do not preserve themselves.

Friendship as Health Behaviour

Dunbar’s 2021 book Friends: Understanding the Power of Our Most Important Relationships makes the health case for friendship with a directness that the underlying research fully supports. The book synthesises decades of evidence on the relationship between social connection and physical health, drawing on both Dunbar’s own research and the wider epidemiological literature on social determinants of health.

The health finding that Dunbar emphasises most strongly is the comparison between friendlessness and heavy smoking. Both are associated with significantly elevated mortality risk, with multiple meta-analyses suggesting that the effect size of social isolation on mortality is comparable to or larger than the effect of smoking fifteen cigarettes a day. This comparison is not rhetorical. It is based on the same kinds of prospective, longitudinal population studies that established the mortality risks of smoking, and it suggests that the failure to maintain close relationships carries health costs that most people do not reckon with when they are too busy to see their friends.

The mechanisms are partly the direct biological effects of social contact (the oxytocin and cortisol regulation that face-to-face interaction provides) and partly the indirect effects of having a social support network that helps people manage stress, maintain healthy behaviours, and access care when they need it. Close friends notice when you are ill. They encourage you to seek help. They provide practical and emotional support during health crises. They maintain and reinforce health behavioural norms like healthy eating, regular activity, moderation. The health effects of friendship operate through multiple pathways simultaneously, which is part of why the overall effect size is so large.

Social Media and the Inflation Problem

The relationship between social media and Dunbar’s numbers is one of the most practically important aspects of his recent work. The question he has investigated directly is whether the apparent explosion in social connectivity produced by platforms like Facebook, Instagram, and LinkedIn has increased the actual quality and quantity of the close relationships that matter for health, or whether it has simply inflated the apparent size of people’s social networks without adding genuine relationship depth.

The answer is largely the latter. Dunbar’s research on social media users found that the number of close and intimate relationships did not increase with the size of a person’s online network. People with 500 Facebook friends had approximately the same number of people they could call on in a crisis as people with 50 Facebook friends. The outer circles had expanded, but the inner circles, which require real investment to maintain, had not grown. Social media had added width to the network without adding depth.

This is not trivial, because the relationships that matter for health are concentrated in the inner circles. The 500 acquaintances on a social media platform are not providing the cortisol regulation, the oxytocin release, or the practical and emotional support that close relationships provide. They are providing an ambient social presence which is not nothing, but which is also not the health-protective relationship that Dunbar’s framework identifies as fundamental. Social media has given people the experience of connection without the biology of it, and the substitution has been so smooth and so comfortable that most people have not noticed what they are missing.

What This Means for Teams

The organisational implications of Dunbar’s framework are direct and insufficiently appreciated. The number 150 has an obvious resonance with the research on team and unit size: organisations that keep their operational units below approximately 150 people tend to function with more coherence, more trust, and less need for formal rule-enforcement than those that exceed it. This is not a management formula but an observation: at below 150 people, it is possible for everyone to know everyone else personally, and the social dynamics of mutual accountability and genuine familiarity can operate. Above 150, they cannot.

More specifically, the conditions under which close relationships form and are maintained at work have direct implications for whether people’s inner circles are being enriched or depleted by their working lives. A stable team, working together regularly and in close physical proximity, over a period of months and years, is a context in which inner-circle relationships can form. People who work in those conditions often find that some of their deepest friendships emerge from it. A team that meets virtually, turns over its membership frequently, and operates in a context where the implicit norm is professional rather than personal contact is unlikely to produce those relationships, not because the people are different but because the structural conditions for close relationship formation are absent.

Contact frequency and stability are particularly important. Dunbar’s research confirms that close relationships require regular contact to maintain their depth. Team members who are in frequent, genuine contact, including the informal and personal contact that happens in physical proximity, are investing in relationships. Team members who meet on a scheduled video call twice a week to review project status are not investing in relationships in the same sense. The investment requires something more than professional interaction: it requires the kind of spontaneous, unscheduled, personal conversation that happens in corridors and kitchens and over lunch, not on calendar invites.

For leaders, this suggests that decisions about team size, team stability, contact frequency, and working patterns are not merely operational. They are decisions about the social conditions of people’s lives, and their consequences extend into the health and wellbeing of the workforce.

Learning More

To go deeper, explore these resources on people-shift.com:

• Wellbeing
• Subjective Wellbeing
• Employee Engagement
• Bowling Alone
• Dunbar’s Number

People Shift View: One of the reasons we think we find leadership development work both rewarding and so draining is that every time you enter a new room of people, you need to exert energy to try and build the connections that help people feel more within your circle of 150. Obviously you don’t need to do that to provide training well, but if you want to really create connections and change in a room, it helps to do so. And doing this, particularly quickly, is something we find fairly effortful. Enough on that though!

We like this model overall. We know that, like everything, it has its critics, and that’s fine. We find it intuitively pretty agreeable and think many other people do too. We also find it validating that it calls out some of the challenges we see with social media as well.

As we say in many places, we’re not sure it’s the role of work to help people build connections. But we do think that any such connections that are built are a great positive externality. 

And from our own careers, we look back most fondly on the roles and teams we’ve been in which have a human-scale number of relationships to manage.