Skip to content
Fraily

Friendship & Health

How Many Close Friends Is Optimal? Research Says: Five

Five. Not fifty, not fifteen — five close friends is the optimum for health and well-being. This number doesn’t come from a lifestyle magazine but from meta-analyses with over 310,000 participants. The surprise: fewer measurably harms your health. Too many does too.

By Fraily EditorialReading time approx. 8 minutes

Why Exactly Five?

Social networks are organized in concentric layers: 1–2 closest confidants, 5 close friends, 15 good friends, 50 acquaintances, 150 contacts — the well-known Dunbar number. Each layer requires a different level of time and emotional investment. The innermost layer of five — the group you reach for when something happens at night — holds a special significance.

A prospective study by Santini et al. (2021) with N = 38,000 painted a nuanced picture: the probability of depressive symptoms followed a U-shaped curve. Too few friends were problematic, but so were too many. Life satisfaction showed a parallel inverted U-shape — with a clear optimum at five.

A meta-analysis by Holt-Lunstad, Smith & Layton (2010) with N = 310,000confirmed: close friendships are the strongest predictor of survival — stronger than diet, exercise, or BMI. Only tobacco use reaches a comparable effect size.

What Too Few Friends Do to Your Body

Loneliness acts not only emotionally but physiologically. Dunbar (2025) describes the β-endorphin mechanism: social touch, shared laughter, singing, storytelling — all of these release endogenous opiates that are 20 to 100 times more potent than morphine. These endorphins regulate the immune system and mood. Without this input, a building block of physiological health is missing — not just psychological confidence.

The numbers for Germany are alarming: 46% of 16- to 30-year-oldsfeel lonely (Bertelsmann Stiftung, 2024). Before the pandemic, the figure was 14–17%. In 2020, it jumped to 41% — and only dropped to 36% by 2022/23. Social loneliness (39%) is more common than emotional loneliness (29%). The five-friend rule is not academic — it affects nearly every other young adult in Germany.

Why More Friends Aren’t Better

The uncomfortable part: if you try to maintain ten or fifteen close friendships, you dilute the time per relationship. Each one becomes shallower. Santini et al. found increased depressive symptoms and lower life satisfaction for this group — statistically significant and stable across multiple measurement points.

Five is therefore not a floor you should try to exceed. It is an equilibrium between depth and breadth. The limiting factor is time: every person has only limited hours for emotional investment. If you try to be close to everyone at once, you end up truly close to no one.

How Often Should You Contact Your Closest Friends?

Your closest five friends need contact at least once a week— not as an average but as a minimum. If you fall below that, emotional closeness starts to measurably decline within a few months (Dunbar, 2025).

Surprisingly, the form of contact does not matter. Whether phone call, text message, or in-person meeting — what counts is frequency, not the medium, as an analysis by Saramäki et al. (2014) shows. This contradicts the popular assumption that only “real meetings” count.

Another finding that underscores the urgency: in the innermost layer of five, only about one person per decadeis replaced (Dunbar, 2025). Once someone drops out, they rarely come back. Maintaining these relationships is not an optional leisure activity — it is one of the most effective investments in your health.

Can Volunteering Replace a Friend?

Partly, yes. Three close friends plus one volunteering activity were just as effective as five close friends alone in the Santini study. Organized groups provide a portion of what close friendships offer — regular contact, shared activity, a sense of belonging.

However: the effects are not additive. Five friends plus three volunteering commitments led to worse outcomes than five friends alone. The limiting factor remains time, not contact.

ConstellationEffect on Well-Being
0–2 close friendsSignificantly elevated risk for depressive symptoms
3 friends + 1 volunteer roleComparable to 5 friends
5 close friendsOptimum for life satisfaction
5 friends + 3 volunteer rolesWorse than 5 friends alone
> 8 close friendsQuality per relationship drops, effect reverses

After Santini et al. (2021), simplified.

Who Does the Five Apply To — and Who Not?

Five is a population average. Women and extroverted people statistically benefit from slightly more close friendships, men and introverted people from slightly fewer. A study of over 7,500 adolescents (ABCD cohort, published in eLife) confirmed: for cognitive abilities, mental health, and academic performance too, the optimum is five.

The practical takeaway: if you have three close friendships and feel good, you don’t need to push to five. If you have seven and notice each becoming superficial, that is not a luxury problem — it is the curve.

What You Can Do Today

Write down the names of your closest friends — the people you would call if something happened at night. Next to each name, note when you last contacted them. If more than seven days have passed, that relationship is starting to erode.

  1. Open your contact list and write to someone you haven’t contacted in over two weeks
  2. Set a weekly reminder for your closest layer of five — Sunday evening, five minutes
  3. Or let Fraily track that in the background for you — with a FriendshipValue that shows you where you stand and who you could reach out to again

Research Limitations

The “magic five” is robust but not a biological constant. Different studies measure “close friends” differently, making comparisons difficult. The direction of causality is not fully resolved: are healthier people more capable of maintaining five relationships — or do five relationships make you healthier? Most studies control for covariates like income or education, but a residual risk of confounding variables remains.

The findings come predominantly from Western, individualized societies. Whether the five applies equally in collectivist cultures is not sufficiently researched. What the data does support, however, is this: If you don’t have five close friendships, that is not a harmless detail of your life. It is a health factor on the scale of smoking.

Nurture the five — don’t just count them

Fraily is built for exactly this. No matching app, no ranking — a quiet rhythm that reminds you of the people who are already there. The FriendshipValue shows you who you’re close to and who you’re not anymore — so friendships don’t die in silence.

Frequently Asked Questions

How many close friends are ideal?
Across multiple large studies, the optimum is around five close friendships. This number corresponds to the innermost layer of the human social network and is associated with the best scores for life satisfaction and mental health (Santini et al., 2021).
Do online friendships count as close friends?
Surprisingly, the form of contact does not matter. Whether phone call, text message, or in-person meeting — what counts is frequency, not the medium (Saramäki et al., 2014). An online friendship can belong to the closest layer if contact is weekly and personal.
How often should I contact my closest friends?
At least once a week — not as an average but as a minimum. If you fall below that, emotional closeness starts to measurably decline within a few months (Dunbar, 2025).
Can volunteering replace a friend?
Partly. Three close friends plus one volunteering activity showed similar scores to five close friends. However, five friends plus multiple volunteering commitments led to worse outcomes — there simply isn’t enough time (Santini et al., 2021).
What can I do if I have fewer than five close friends?
Deepen existing relationships rather than seeking new ones. Often there are people in your wider circle who have the potential for more closeness. Regular group activities — volunteering, clubs, sports groups — create contexts where friendships grow naturally.

Sources

  1. Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7(7), e1000316.
  2. Santini, Z. I., et al. (2021). The moderating role of social network size in the temporal association between formal social participation and mental health. Social Psychiatry and Psychiatric Epidemiology, 56, 417–428.
  3. Dunbar, R. I. M. (2025). Why friendship and loneliness affect our health. Annals of the New York Academy of Sciences, 1545, 52–65.
  4. Saramäki, J., et al. (2014). Persistence of social signatures in human communication. Proceedings of the National Academy of Sciences, 111(3), 942–947.
  5. Bertelsmann Stiftung (2024). Wie einsam sind junge Erwachsene im Jahr 2024? Gütersloh.