How to build a support network, one relationship at a time
Science · by Simone Rainieri · 9 min read read · Published
A support network is the small set of specific people you could actually ask for a specific kind of help, and you build it by making individual one-to-one relationships deep enough that asking is not awkward, not by knowing more people. The research is unusually clear on this. What protects people under stress is largely perceived support, the belief that help would be there, rather than received support, the tally of help actually delivered. That single distinction changes what you should spend your effort on.
It also explains a common and confusing experience: someone with a busy calendar, a full group chat and forty colleagues who nonetheless could not name a person to call at 11pm on a Sunday. Volume of contact and availability of support are almost separate quantities. You can have a great deal of one and none of the other.
The four kinds of support, and why you need different people for each
Sheldon Cohen and Thomas Wills, in their 1985 review in Psychological Bulletin that set the terms for the field, distinguished four functions: emotional support, informational support (guidance for a decision), instrumental support (concrete practical help), and appraisal or esteem support (feedback that you are not being unreasonable). Their central claim, the buffering hypothesis, is that support protects people against the effects of stressful events specifically when the support available matches the need the event creates.
That last clause does the work. A friend who is superb at sitting with you while you are upset is the wrong person for a decision about whether to take the job, and the person who will drive to the airport at 5am may be useless at either. A support network is not a group of interchangeable good people. It is a small portfolio with different functions covered.
Why perceived support beats received support, and one very strange finding
There is a result here that seems wrong until you sit with it. Niall Bolger, Adam Zuckerman and Ronald Kessler ran a daily-diary study of couples during a period of acute stress and found that support the recipient noticed often failed to help, while support the recipient did not notice did. People whose partners reported providing a great deal of support, but who did not themselves report receiving much, showed lower anxiety and depressed mood than those who reported receiving a lot.
The explanation is that being visibly helped carries a cost. It confirms you could not manage, it creates a debt, it makes you the person who needs things. Which means the friend quietly moving the deadline for you is doing something better than the friend announcing that they are here for you, and it means a network built on conspicuous mutual rescue is more fragile than one built on ordinary presence.
The practical translation: the most useful support network is made of people who are already in your week for reasons unrelated to your problems. Someone you walk with every Sunday can help you without the help becoming an event. Someone you only contact in a crisis cannot.
How to actually build one, in order
- Audit rather than aspire. Write the four functions down and put real names against each. Emotional, informational, practical, esteem. Most people find two functions well covered and two empty. The empty ones are your brief.
- Fix the practical one first, because it is the cheapest to build and the most reciprocal. A neighbour with your spare key, a colleague who covers, a cousin twenty minutes away. Practical support is transactional in a good way: it can be repaid, so it does not accumulate debt.
- Turn one acquaintance into a one-to-one relationship. Not five. One. A support network of three real people outperforms a circle of twenty acquaintances, and the whole thing is built pair by pair.
- Add recurrence before you add depth. A fortnightly fixed slot with one person does more than an intense conversation every six months, because availability is built out of repetition rather than intensity.
- Be the support before you need it. Not strategically. It is simply that the people you have shown up for are the people it stops being awkward to ask, and the awkwardness is the only real barrier for most adults.
- Give the network somewhere to notice you. Tell one person the small true version of what is going on, monthly. Perceived support is built by evidence, and evidence requires disclosure.
The one-on-one bias, and why it is correct
Group settings are efficient for meeting people and poor at producing support. In a group of six, nobody is responsible for you, and diffusion of responsibility is a real effect rather than a metaphor. In a pair, there is nowhere for the responsibility to go.
So if you are starting from very little, the highest-return move is to take the group you are already in and extract one person from it into a repeating one-to-one. A walk, a Thursday coffee, a phone call on the commute. This is unglamorous and it is genuinely how most adult support networks get built. The comparison between the two formats is set out in one-to-one friendship apps vs group events.
What a functioning network looks like, versus what people imagine
| Imagined | Actual |
|---|---|
| A large circle of close friends | Three to five people, each covering a different function |
| Everyone knows everyone | Mostly separate relationships that never meet |
| Deep conversations regularly | Ordinary recurring contact, occasional depth |
| You call people when in trouble | People are already in your week, so the call is not an event |
| Built through big social effort | Built pair by pair over one to two years |
If you are starting from close to zero
Then the honest timeline is one to two years to build three real relationships, and the first six months will feel like nothing is happening because acquaintance to friend is not a visible transition. That is not failure, it is the shape of it. How to deal with loneliness as an adult covers the interim, and how to make friends as an adult covers the mechanics of the first step.
If a life transition is what emptied the network, the rebuilding is more targeted, and friendship platforms for major life transitions is the piece for that case. Vairi fits this particular problem narrowly: one match at a time is deliberately a one-to-one product rather than a room, which suits network-building better than a feed does, and the anonymity until mutual opt-in lowers the cost of saying what is actually going on. It is also small, new, and only in London and New York, so for most people reading this it will be one component at best, not a solution.
Key takeaways
- Cohen and Wills 1985 buffering model: support protects against stress when its type matches the demand. Four functions, emotional, informational, instrumental and appraisal, and they usually need different people.
- Perceived availability of support predicts outcomes better than support actually received. Build for believability, not for frequency of rescue.
- Bolger and colleagues found support the recipient noticed often did not help, while unnoticed support did. Quiet ordinary presence beats visible rescue.
- Eurostat EU-SILC data put 13.8 percent of the EU population at poor social support in 2019, and 18.2 percent among those aged 65 and over.
- Three to five people covering different functions is a working network. Twenty acquaintances is not.
- Build pair by pair, add recurrence before depth, and expect one to two years from a standing start.
What is a support network?
The specific people you could ask for a specific kind of help: emotional, informational, practical, or reassurance that your judgement is sound. It is defined by what those people would do rather than by how many of them there are.
How many people should be in a support network?
Three to five who cover different functions is enough for most people, and is more useful than a much larger circle where every relationship serves the same purpose. The functions matter more than the count.
What is the difference between perceived and received social support?
Perceived support is your belief that help would be available if you needed it. Received support is help actually given. Research has consistently found the perceived kind is the stronger predictor of wellbeing under stress, which is why building credible availability matters more than accumulating instances of help.
How long does it take to build a support network from scratch?
Expect one to two years for three real relationships if you are starting with almost nobody, with the first six months feeling flat because acquaintance-to-friend has no visible moment. Recurrence is the accelerator: a fixed fortnightly slot with one person compounds faster than occasional intense contact with several.
Related reading
- How to deal with loneliness as an adult
- How to make friends as an adult
- One-to-one friendship apps vs group events
- Friendship platforms for major life transitions
- How to tell if a friendship is one-sided
Sources
- Cohen and Wills (1985), Stress, social support, and the buffering hypothesis, Psychological Bulletin 98(2), 310-357. https://www.semanticscholar.org/paper/Stress,-social-support,-and-the-buffering-Cohen-Wills/523fb3964458ea60541137a955371ceda95e29c0
- Bolger, Zuckerman and Kessler (2000), Invisible support and adjustment to stress, Journal of Personality and Social Psychology. https://pubmed.ncbi.nlm.nih.gov/11138764/
- Eurostat, Mental well-being and social support statistics (Oslo-3 social support scale, EU-SILC). https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Mental_well-being_and_social_support_statistics
Read next: Wink app alternatives for adults who want friends, not dates
Meet people from your city, from your sofa.
Friday Aperitivo is one relaxed hour online with a few new people, every Friday at 8pm UK time. Tables of up to six, voice with camera optional, and Vairi hosts with question cards. Free.
Proudly listed on Launchpadly Startup Directory