Fund the Role, Not the Founder

A funder asks a reasonable question before renewing a grant: what happens to this walking group, this men’s shed, this parent-and-baby circle, if the person currently running it leaves? Too often the honest answer is: it stops. That is not a failure of the activity. It is a failure of role design, and it is fixable before you write the next cheque.

The founder cliff

Call it the founder cliff. A group starts because one person — a link worker, a volunteer, a particularly organised participant — holds every function at once: booking the room, texting reminders, greeting newcomers, managing the WhatsApp group, deciding what happens if someone gets upset. The group runs well. Attendance is good. The funder is pleased. Then that person changes jobs, has a baby, or burns out, and within a few weeks the group is gone.

This is not a hypothetical risk you’re managing. It is the default outcome of funding a person rather than a structure. Grant reports rarely capture it because the group usually collapses after the reporting period, when nobody is measuring anymore.

What a group actually needs, broken into roles

The qualitative literature on social prescribing offers a clue here that funders don’t often act on. A 2022 meta-synthesis of participant accounts found that benefit came not from social contact alone but from structured, purposeful activity — people describing a restored sense of participation and role, not just company. A systematic review the year before found the same pattern: self-esteem and confidence gains were tied to having something to do and a place in doing it, not to loose social mixing. If a group’s value comes partly from participants having a defined role, that logic should apply to the group’s own operation, not just its activity content.

In practice, most sustainable groups — whether anyone designed them this way or not — separate four functions:

Function	What it actually involves	Can it rotate?
Convening	Booking space, setting the schedule, confirming who’s coming	Yes, with a checklist
Welcoming	Greeting newcomers, making the first visit not awkward	Best kept with 2-3 people, not one
Continuity	Holding the group’s history — why it started, who used to come, what’s been tried	Hard to rotate; needs deliberate handover
Escalation	Knowing what to do if someone discloses risk, illness, or crisis	No — must sit with a named, trained person

Most funded groups fund one person to do all four and call it “coordination.” That’s the design flaw. Ray Oldenburg’s account of the third place — the café, the barbershop, the pub — described these places as working partly because the roles inside them were dispersed across regulars, not concentrated in an owner. Robert Putnam’s account of civic decline similarly points to the loss of the layered volunteer structures — the treasurer, the newsletter editor, the person who always brings the coffee — that used to distribute responsibility across a membership rather than a founder.

What this means in practice

Fund a structure, not a person. Before renewing or approving a grant for a group, ask to see who holds each of the four functions above, by name. If one person’s name appears in all four boxes, that is the single biggest sustainability risk in the proposal — bigger than budget, venue, or attendance projections. Ask the applicant to name at least one other person for welcoming and continuity before you sign off, and require a written handover note for continuity, even a one-page one, as a condition of the grant.

What to build into the grant agreement

1. Require two names, not one, for any role above “convening.” A group with a single point of contact is a group you are funding to fail at year two.
2. Ask for a continuity document, not a full manual — a page covering why the group started, what’s been tried and dropped, and who the regulars are. This is cheap to produce and expensive to reconstruct after someone leaves.
3. Separate escalation from welcoming. The person best at making newcomers comfortable is rarely the person who should be handling a disclosure of self-harm or abuse. Fund training for whoever holds escalation, specifically.
4. Budget for a deliberate handover period, not an abrupt one. Two months of overlap, even part-time, costs less than restarting a group from zero and losing a year of attendance data.
5. Track role occupancy, not just attendance, in your monitoring. A dashboard that shows “12 sessions run, 140 attendances” but nothing about who ran them tells you nothing about whether the group survives the next twelve months.

Where the evidence runs out

None of this is drawn from a study of role structure in community groups specifically — no such evidence base exists in the sources available here. The reasoning is inferred from adjacent literature: what participants say they get out of structured activity, and older accounts of civic associations and third places that happened to distribute roles this way. Treat it as a design principle worth testing against your own attrition data, not a proven intervention.

What this does not solve

This addresses one specific failure mode — collapse after a key person leaves — and nothing else. It does not address why people don’t show up to begin with, which is a separate and larger problem: the American Enterprise Institute’s 2021 survey found the share of men reporting no close friends had risen fivefold since 1990, and that decline in social ties happens well upstream of any group you might fund. A well-structured group still only reaches people who found their way to the door. Fixing the founder cliff makes a good programme durable. It does not make a programme reach the people who never hear about it.