Measuring Volunteer Recruitment, Training, and Retention as Leading Indicators

If your programme depends on volunteers — befrienders, peer supporters, workplace connector roles, community hosts — you are probably measuring the wrong thing first. Most evaluation plans start with beneficiary outcomes: did loneliness scores drop, did people report more social contact. Those matter. But by the time a beneficiary outcome moves in the wrong direction, the actual failure happened months earlier, inside the volunteer pipeline, where almost nobody is looking.

Volunteer recruitment, training completion, and retention are leading indicators. They tell you a programme is about to fail before a single participant outcome does. Funders rarely ask for them. You should report them anyway, and you should report them before you report anything else.

Why the pipeline is the real dependent variable

Programmes built on volunteer labour to deliver connection — matching, facilitating, hosting — are structurally fragile in a way that clinical interventions are not. A therapist’s caseload doesn’t evaporate because three volunteers had a bad month. A volunteer-run befriending scheme’s caseload does.

This fragility sits on top of a longer-term trend. Robert Putnam’s Bowling Alone documented a decades-long decline in the kind of associational membership that used to supply volunteers as a byproduct — church groups, unions, bowling leagues, PTAs. People who belonged to those things volunteered for other things almost automatically. That supply has thinned. The 2021 Survey Center on American Life report on friendship found that the share of men reporting six or more close friends fell from 55% in 1990 to 27% in 2021, and 15% of men now report no close friends at all — a fivefold increase since 1990. Fewer dense friendship networks means fewer people getting asked to volunteer by someone they already know, which was always the main recruitment channel for informal civic roles.

None of this means volunteers are unavailable. It means you cannot assume replacement supply the way programmes could twenty years ago. You have to build recruitment as a continuous function, not a one-off launch activity.

What to measure, and when

Most programmes measure recruitment (how many signed up) and stop there. That single number tells you almost nothing about programme health six months out.

Stage	What to track	Why it matters
Recruitment	Applications per cohort; source of applicant (referral vs. cold)	Referral-heavy pipelines collapse faster when the referring cohort leaves
Training	Completion rate; time from sign-up to first shift	A long gap between sign-up and first activity is where interested people quietly disengage
Early tenure	Attrition in first 90 days	This is where most volunteer loss happens, and it is almost never reported separately from overall attrition
Sustained tenure	Active volunteers at 6 and 12 months; hours contributed per volunteer	The number that actually predicts whether beneficiary-facing capacity holds
Exit	Reason for leaving, self-reported	Distinguishes burnout, mismatch, and life circumstance — each needs a different fix

Report the 90-day figure separately from annual attrition. Aggregated attrition numbers hide the fact that most loss happens fast, at the point where enthusiasm meets the actual, less glamorous work of showing up.

The second cohort problem

Name this one, because it is the pattern that kills programmes quietly. A programme launches with a strong founding cohort of volunteers — motivated, often recruited through personal networks, willing to absorb the friction of a new process. Everything looks healthy for six to nine months. Then the founding cohort starts leaving, for entirely ordinary reasons — new jobs, moves, burnout — and there is no second cohort in the pipeline behind them, because recruitment was treated as a launch task rather than a standing operation.

The programme doesn’t fail visibly. It shrinks. Caseloads per remaining volunteer creep up, quality drops, and the beneficiary outcomes you’re tracking degrade for reasons that have nothing to do with the intervention design and everything to do with capacity you stopped watching.

What this means in practice: budget and staff recruitment as an ongoing function with its own target — a rolling pipeline, not a launch campaign — and report volunteer attrition to funders at the same cadence you report beneficiary outcomes. If you only report the second, you will not see the first collapse coming.

Evidence status

Claim	Evidence status
Social connection interventions improve self-esteem and confidence	Reasonably supported; a 2021 systematic review of social prescribing found consistent gains, though with heterogeneous designs
Social prescribing reduces service use (GP, emergency, inpatient)	Weakly supported; a 2021 systematic review found three of nine included studies reported reductions, not a majority, and none were randomised at scale
Loneliness has direct costs to employers	Supported by survey data; Cigna’s 2020 report found lonely workers miss work roughly twice as often for illness and five times as often for stress-related reasons
Volunteer-delivered connection programmes need routine capacity monitoring	Not directly tested in any cited study, but implied by the general call in the National Academies’ 2020 report for systems to routinely assess social connection rather than treat it as a one-off
Shared physical spaces (libraries, parks) support informal volunteer-adjacent contact	Supported descriptively; Eric Klinenberg’s account of social infrastructure argues shared space shapes contact rates, though this is observational rather than experimental

The systematic reviews on social prescribing are honest about their own limits: heterogeneous programme designs, few controlled trials, and outcomes that are mostly self-reported. Use them to justify running a programme, not to promise a funder a specific effect size.

What this does not solve

None of this fixes the underlying supply problem — fewer people with the dense friendship networks that used to generate volunteers as a byproduct of ordinary social life. A better recruitment pipeline gets you more applicants from a shrinking pool; it does not grow the pool. And volunteer-delivered programmes, however well staffed, still mainly reach people willing to accept a volunteer’s help in the first place. The hardest-to-reach isolated people — the ones with no network to hear about the programme through — are underrepresented in every volunteer-run scheme’s caseload, and better retention metrics will not change that on their own.