Relationships · Social

The Three Tribes: Why You Need 5 Close, 15 Trusted, 50 Active

Robin Dunbar spent four decades studying why human social groups top out at specific numbers, not random ones. Here's the layered structure behind the famous 150, and how to actually invest your time across it.

https://taskcoach.ai/blog/three-tribes-dunbar-number/

The social layers you actually run on

Robin Dunbar, an evolutionary psychologist at the University of Oxford, spent four decades studying something most of us never think to question: why human social groups keep landing on the same handful of numbers, over and over, across wildly different cultures.

The number most people know is 150, "Dunbar's number." But that's really just the outer edge of something bigger. Dunbar's work, replicated across modern societies, hunter-gatherer tribes, military units, and online networks, points to four nested layers:

  • 5 intimate relationships (your closest support)
  • 15 trusted friends (regular emotional engagement)
  • 50 active acquaintances (weekly to monthly contact)
  • 150 stable relationships (people you genuinely know)

Each layer serves a different function, costs a different amount of your attention, and runs on different rules. Most people manage all four badly, not because they're bad at relationships, but because nobody ever taught them to allocate attention by layer instead of by whoever happens to be standing in front of them. Fixing that allocation pays off more than it sounds like it should.

Layered structure. Different costs. Different functions. Different rules.


Why there's a hard limit at all

Dunbar's explanation traces back to neocortex size and the sheer cognitive cost of tracking another person in real detail: their history, their preferences, what's going on in their life right now, what they're worried about. Holding all of that in working memory is expensive, and your brain only has so much of that specific kind of bandwidth to spend.

The layers map onto that cost. The innermost 5 eat up something like 40% of your total social bandwidth. The next 10 (positions 6 through 15) take another 20%. The next 35 (16 through 50) take another 20%. The remaining 100 (51 through 150) split what's left.

That's the whole reason most adults can name around 150 people they genuinely know, why only about 5 of those feel like "real" friends, and why trying to widen the inner circle just dilutes it. It also explains something that trips people up about social media: it expanded the outer layer of weak ties enormously without touching the inner one at all. You can follow ten thousand people. Your actual inner five is still five.


What each layer is for

The 5: who you call in a crisis. This is 2am-phone-call territory. A parent dying, a marriage ending, a medical emergency. These people absorb the real emotional weight. Most adults, if they're honest, only have two or three people who genuinely belong here, even if they'd list more on paper.

The 15: the people whose life you actually track. Regular emotional engagement, contact every week or two, real shared history. You remember their birthday without a reminder and you notice when something's off. A meaningful chunk of the wellbeing benefit found in the long-running Harvard Study of Adult Development lives in this layer.

The 50: your active social fabric. Monthly or quarterly contact. Coworkers you actually like, gym regulars, neighbors, extended family. They add texture and opportunity to life even though you'd never call them at 2am. This is the layer the pandemic hit hardest, and plenty of people still haven't rebuilt it.

The 150: your village. People whose face and name you'd recognize at a community event, who'd do the same for you. Not close, but not strangers either.


Auditing what you really have

Name the five. Name the next ten. Calendar-audit the last thirty days. The gap between intention and reality is the data.

Three quick exercises that show you where you actually stand, not where you assume you stand.

Exercise 1: name the 5

Without overthinking it, write down the five people you'd call in a real 2am crisis. Most people can name two or three without hesitating, then stall. That stall is data.

Exercise 2: name the 15

Now add the next ten: people whose ongoing life you genuinely follow. This one's harder than it sounds. A lot of people discover that what they've been calling "close friends" actually belong a layer further out.

Exercise 3: audit the calendar

Look back at the last 30 days. Who did you actually see, call, or message in a way that meant something? Compare that against your list of 15. The gap between who you say matters and who you actually showed up for is usually bigger than people expect.


How to reallocate your time

50% into the inner five. 30% into the next ten. 5% into the outer layers.

The Harvard Study's core finding, that relationship quality predicts how well your later life goes, doesn't apply evenly. It applies layer by layer, and the inner ones carry almost all the weight. That points to a specific, slightly uncomfortable reallocation:

Put about 50% of your social time into the 5. Put 30% into the next 10. Save 15% for positions 16 through 50, and 5% for everyone else.

Most people do the opposite by default: they spread attention evenly across whoever happens to be around, which quietly starves the inner layers that matter most. It feels backward at first, but it's what the longitudinal data actually supports.

Some concrete targets:

  • Weekly contact with each of the 5. A call, a video chat, in person. Texting alone doesn't count.
  • Monthly real contact with the 15. Not just a birthday message, actual engagement with what's happening in their life.
  • Quarterly contact with the 50. Brief is fine. A meal, a check-in, a planned catch-up.
  • A yearly touch for the 150. A card, a note, showing up to the thing.

Investing in the inner layers tends to compound outward too, since the people closest to you are usually the ones who introduce you to the people in your next layer out.


Why this matters more now, not less

Optimizing for followers trades high-impact relationships for low-impact ones. The math says no.

The modern trap is optimizing for the outer layers, followers, network size, reach, at the direct expense of the inner ones. The math on that trade is bad. You're swapping high-impact relationships for low-impact ones and calling it growth.

The fix isn't comfortable. It means saying no to plenty of surface-level social opportunities to protect time for the 5 and 15. It means treating small, deep relationships as worth defending, even when a bigger, shinier network is right there tempting you.

Run the Harvard data forward and the person with strong inner layers and a thin outer network comes out ahead of the person with the opposite setup, consistently.


Where TaskCoach fits in

The Social pillar in TaskCoach.AI can turn the Dunbar layers into actual habits: weekly check-ins with the 5, monthly touches with the 15, quarterly ones with the 50. The dashboard keeps that maintenance visible instead of letting it quietly slide, and protects your inner-layer time from getting eaten by whatever outer-layer noise shows up that day.

It won't build the relationships for you. It just makes them a lot harder to forget.

The bottom line

The 5/15/50/150 structure isn't arbitrary. It's a real cognitive limit with four genuinely different layers, each doing a different job, and your time should reflect that.

Name your 5. Take stock of your 15. Move your time toward the inner layers on purpose, even when it feels like you're neglecting the rest. Decades of Harvard data and decades of Dunbar's research point at the same conclusion from two completely different directions.

Frequently asked questions

What is Dunbar's number?

About 150, the cognitive ceiling on how many stable relationships one person can realistically maintain. Robin Dunbar worked it out from the ratio of neocortex size to group size in primates, then found the same number showing up in hunter-gatherer clans, military units, and even online networks. It's really just the outer layer of a bigger, four-layer structure.

How many close friends should I have?

Dunbar's research points to roughly 5 truly intimate relationships (near-daily contact, your closest support) and 15 trusted friends you engage with regularly. Try to maintain more than that at real depth and something gives: the extra relationships just slide down into a shallower layer.

Are Dunbar's numbers scientifically valid?

The 150 figure has held up across hunter-gatherer clan sizes, effective military unit sizes, and analysis of online social networks. The inner layers (5, 15, 50) are also supported, though with more individual variation. Treat these as central tendencies, not hard caps everyone hits exactly.