What is social capital and why is it important?

Updated from 2022.

A few years ago, when I was working at HM Land Registry, their head of design Laura Yarrow asked me to write about how I build social capital.

I hadn't come across the term. After talking to Laura it was clear I'd been doing it for years without a name for it. But I got the definition wrong at the time, and it's worth correcting, because the mistake is a common one.

I described social capital as building trust, skills and capabilities. Skills and capabilities are something else. They're human capital, and they live inside a person. Social capital doesn't live in a person at all. It lives in the space between people. It's the trust, the relationships and the quiet obligations that let you get something done that you couldn't do on your own.

Put simply: human capital is knowing how to do the work. Social capital is being able to get the work done.

Why it matters

Most projects I've seen struggle didn't struggle for technical reasons. The team knew what to build. What they lacked was a route into the parts of the organisation that mattered, or enough credibility with the people holding the budget, or a relationship with the service owner who could really change something.

This is especially true in health, where I work now. Products fail because nobody who mattered was in the room early enough, because the clinical champion moved on or because the evidence answered a question the commissioner wasn't asking. Those are relationship failures dressed up as delivery failures.

You can be brilliant and stuck but social capital can be what unsticks you.

Three kinds, not one

It helps to separate social capital into three types, because they're built differently and teams tend to be strong in one and weak in the others.

Trust inside the team. The bonds that let people disagree without it becoming personal, admit they're lost, and cover for each other.

Reach across the organisation. The loose connections into other teams, other directorates, other professions. These are usually weak ties - people you don't know well - and that's exactly why they're valuable. Your close colleagues know what you know. New information comes from further out.

Traction up and down. The relationships that cross a power gradient: to executives, funders, commissioners, regulators. The people who control permission and money.

Most teams are decent at the first, patchy at the second, and neglect the third until they urgently need it - which is precisely too late.

What builds it

Working in the open

Showing the work, letting people ask questions, letting them challenge us. Done consistently, it spreads into corners of the organisation you'd never have reached deliberately.

The mechanism matters here. Working in the open makes your work legible to people outside your team. Legible work is something a stranger can attach themselves to. That's how you accumulate the loose connections across silos that you'll need later.

How: Regular show and tells, open channels on Slack or Teams, sharing work in progress rather than polished conclusions.

One caveat. A show and tell only builds social capital if the people watching can influence something. If nothing they say ever changes the work, it isn't openness, it's broadcast - and people can tell the difference. Invite a challenge, then visibly act on some of it.

At HMLR we ran show and tells that were open to the whole organisation and shared the recordings. People used them to test our thinking, and other teams picked up the ways of working for themselves.

Understand what people are trying to achieve

Leadership can come from anyone at any time. Mostly it's about understanding what people are trying to get to and helping them get there.

The underlying mechanism is reciprocity. Social capital is built by doing things for people before you need anything from them. Not transactionally - you rarely know which relationship will matter. But a network of people who've been helped by you is a fundamentally different asset from a network of people who've merely met you.

How: Regular one-to-ones. Kicking off projects by getting everyone's goals on the table. A User Manual of Me so people know how you work and how to work well with you. Retrospectives that reflect on how the team is working, not just what it shipped.

Define outcomes together, early

Agreeing outcomes as a group at the start does more than keep you focused. It gives you something to build relationships around.

A shared outcomes map is a useful object for people who have no other reason to talk to each other. A clinician, a data engineer and a finance lead can all argue about the same diagram. That argument is the relationship forming.

How: Run outcomes mapping at the start and return to it. I usually use theory of change, which connects long-term outcomes to the short-term things you're actually able to do.

At HMLR we did this with people connected to other parts of the programme who'd been working on outcomes separately. Bringing it together showed how the different project outcomes connected, and it unlocked outcomes for another project that had been stuck.

Spend it

Capital that never gets spent isn't capital. Make the introduction. Lend your credibility to someone else's proposal. Put your name on an ask that isn't yours.

Unusually for an asset, this one grows when you use it and decays when you hoard it.

The parts people don't say

Social capital is not automatically a good thing, and it's worth being straight about it.

It's exclusionary by default. Networks reproduce themselves. Left alone, you'll build a strong, trusting network of people who look and think like you, and you'll mistake the comfort of that for effectiveness. Widening who's in the room has to be a deliberate act, repeated.

It gets cosy. A tight group that trusts each other completely will eventually stop hearing anything new. Trust and challenge have to be held together or you get consensus dressed up as alignment.

It's invisible and unrewarded. None of this appears in a delivery plan. It's slow, it accrues over months, and it's almost impossible to attribute. Which is why it's the first thing cut under pressure and the last thing anyone gets credit for.

It's portable. It leaves when people leave. If an organisation depends on a handful of individuals holding relationships together, it's fragile, and it doesn't know it yet. The useful question is not "who do I know" but "what would stop working if I left tomorrow?" - and then, how do you make less of it depend on you.

At the scale of a city

I now work across a health innovation programme in the Liverpool City Region, and the thing I've noticed is that everything above is amplified when you leave a single organisation behind.

The people who need to work together - universities, NHS trusts, founders, investors - don't alwreport to each other. Nobody can instruct anybody. There's often no shared delivery plan, no common incentive and often no shared vocabulary.

In that context, trust isn't a nice-to-have that helps delivery. It is the delivery mechanism. The job is largely creating the conditions where a founder, a clinician and an investor will sit in the same room, take each other seriously, and keep turning up.

That's the thing about social capital. It looks soft right up until you try to work without it.