What the data actually shows
Anthropologist Robin Dunbar's research on social networks finds that friendships are not self-sustaining: they decay without contact. In his work, the strength of a friendship is closely tied to how much time and communication is invested in it, and bonds weaken measurably when that investment drops — for example, after one person moves away. Friendship, in this view, is a maintained relationship rather than a permanent state.
Dunbar's research also describes a layered structure to our social world — a small inner circle of a handful of closest friends (often summarised as around five), a wider layer of close friends (on the order of roughly fifteen), and successively larger, looser layers. Keeping someone in an inner layer takes more frequent contact; when life reduces that contact, people tend to slide outward into looser layers rather than vanish entirely.
A long line of social-psychology research, going back to studies of how proximity shapes friendship, finds that we disproportionately befriend the people we repeatedly and effortlessly encounter — neighbours, classmates, colleagues. That helps explain why life transitions are so disruptive: a move, a job change, or a new schedule removes the unplanned, repeated contact that the friendship was quietly running on, and few friendships survive purely on intention.
Real numbers in context
The numbers here are about social structure, and they should be read as rough patterns rather than precise rules. Dunbar's research describes an inner circle of roughly five closest people and a close-friend layer of around fifteen, nested inside much larger, looser layers — with the broad outer limit on stable relationships often cited as around 150. These are averages with wide individual variation, not targets.
What the figures really illustrate is scarcity and decay: close ties are few and they require upkeep, so when a life change cuts the contact a friendship was living on, it tends to slip outward in that structure rather than hold its place. Surveys also point to a wider context in which adults report relatively few close friends and friendship-making gets harder with age — so drift after transitions is happening against an already thin baseline, which is worth keeping in view without treating it as a personal failing.
The same figures as bars. Close ties are few and require upkeep; the looser outer layers are much larger. When contact drops, a friendship tends to slip toward the wider layers rather than hold its place.
Nested layers of decreasing closeness. These are rough averages with wide individual variation, not targets. Keeping someone in an inner layer takes more frequent contact; when a life change cuts that contact, people tend to slide outward rather than disappear.
| Layer | Approx. number of people |
|---|---|
| Innermost circle (closest friends) | ≈ 5 |
| Close-friend layer | ≈ 15 |
| Outer limit on stable relationships ('Dunbar's number') | ≈ 150 |
Why this feels different from how it actually is
Drift feels personal because the cause is invisible. You do not see 'we no longer share a commute, a workplace, or a free Tuesday night' — you just notice that you talk less, and the mind tends to fill that gap with a story about feelings, fault, or fading affection. The structural explanation is usually more accurate and far less common as a first guess.
It also feels asymmetric and a little shameful because the friendships that drift are often ones nobody decided to end. There is no breakup, no event to point to, so it can feel like a quiet failure on someone's part. In the research, this kind of fade without rupture is the normal way most friendships change over a life — closeness flows toward whoever you are sharing daily life with at the time.
And life transitions tend to consume exactly the resource friendships run on: time and attention. New jobs, new partners, and especially new children absorb the discretionary hours that used to go to friends. The friendship has not been judged and rejected; it has been outcompeted for a scarce resource, which feels different from the inside than it is.
Drift after a life change is one of the most common and least personal things that happens to friendships.
What the research says to do about it
If the goal is to keep a specific friendship through a transition, the research points clearly to maintaining contact, because that is the variable bonds actually depend on. Since proximity is gone, the contact has to become deliberate — regular check-ins, a standing call, a recurring plan — to substitute for the unplanned encounters the friendship used to get for free.
Lowering the bar for contact tends to help more than saving up for big reunions. Frequent, low-effort touchpoints map onto how friendships are sustained better than rare, high-production catch-ups, because what decays is the rhythm of contact rather than the depth of any single conversation. A short message often does more maintenance work than its size suggests.
It also helps to accept that you cannot keep every friendship in the inner layer, and to be intentional about which few you actively maintain. The layered structure in Dunbar's research implies a real constraint on close ties, so choosing where to spend limited social energy — and letting some bonds settle honourably into looser layers — is realistic rather than cold.
What the research says does not help
Reading drift as a verdict on the friendship or on yourself usually does not fit the evidence and adds unnecessary hurt. Most fading is a side effect of changed circumstances, not a judgement, so interpreting it as rejection tends to be both inaccurate and discouraging from reaching back out.
Waiting for the friendship to revive on its own rarely works, because the research is fairly consistent that friendships decay without contact. Assuming a strong past bond will survive on stored goodwill while contact sits at zero is the most common way good friendships quietly lapse.
Trying to maintain a large number of close friendships at full intensity through a busy life transition tends to fail under the same constraint everyone faces — there are only so many hours and only so much room in the inner layers. Spreading effort too thin often means no friendship gets the regular contact it needs, whereas concentrating on a few is more sustainable.
The friendship has not been judged and rejected; it has been outcompeted for a scarce resource.
What this looks like in real life
The friend who moved away
Nobody decided to end it. You simply no longer share a commute, a workplace, or a free Tuesday night — so the unplanned, repeated contact the friendship was quietly running on is gone, and you talk less. The mind fills that gap with a story about feelings or fault, but the more accurate explanation is structural: the scaffolding disappeared, not the affection.
Outcompeted by a new baby
A new child absorbs the discretionary hours that used to go to friends. The friendship hasn't been judged and rejected — it's been outcompeted for a scarce resource, time and attention. Keeping a specific one alive through the transition means making contact deliberate: a standing call or frequent short messages that substitute for the encounters it used to get for free.