The tactics that keep a new volunteer stop working later

There is a volunteer in most programs who has been there six years. She knows where the spare keys are. She trained half the current roster. And every quarter she gets the same welcome newsletter as the person who signed up in March, plus an invitation to the same refresher session she could teach.

Nothing is wrong. That is the problem. She is being managed with a set of tools designed for somebody in their eighth week, and those tools have stopped doing anything for her.

Three stages, and they need different things

The most useful piece of theory in this whole subject is a 2007 paper by Chacón, Vecina and Dávila in Social Behavior and Personality, which proposed that volunteer duration is predicted by different factors at different tenures.

Short term, meaning the first months, the predictor is satisfaction. Medium term, it becomes organizational commitment. Long term, past roughly a year, it becomes role identity: whether the person has come to think of being a volunteer of yours as part of who they are.

Three stages, three different levers, and almost every volunteer program I have looked at operates only the first one. We survey satisfaction, we improve the shift experience, we run a nice welcome, and then we apply that same machinery to somebody in year four and wonder why she seems flat.

The finding that makes this uncomfortable

Vecina and Chacón went further in a 2013 paper in Anales de Psicología. They studied 257 volunteers across five nonprofits, measuring engagement, satisfaction and organizational commitment against intention to remain over the following year.

Organizational commitment was the only predictor of intention to remain. Engagement and satisfaction had no significant direct path to it at all.

That is a strange result to sit with if you run a volunteer program, because satisfaction is what we measure. It is the question on the form. And here is a study saying that among volunteers who are already established, how satisfied they are tells you very little about whether they will be here next year.

The same paper found satisfaction and engagement were doing other work. Engagement predicted psychological well-being. Satisfaction predicted how volunteers rated their own physical health. Those are good outcomes and worth having. They are not retention.

Two caveats before anyone rewrites a strategy on this. It is a single study of 257 people in one country, and it is correlational, so intention to remain is not the same as remaining. But it points the same direction as the stage model, and the stage model has held up.

What organizational commitment is made of

Commitment is not enthusiasm. It is the sense that this organization is one you belong to and are willing to be inconvenienced by.

NCVO’s Time Well Spent 2023, based on YouGov fieldwork with 7,006 UK adults, asked continuing volunteers why they kept going. The group or organization itself came out at 46 percent, tied with the difference they felt they were making, also 46 percent. Commitment to the cause came third at 41 percent.

Read that carefully. The organization scored as highly as the impact and higher than the cause. People do not stay attached to homelessness in the abstract. They stay attached to the Tuesday crew at the shelter on Bank Street and the woman who runs it.

Which gives you something to work with, because organizational belonging is built by very ordinary things. Being told about the funding decision before it goes out publicly. Being asked what you think before the decision, not after. Being told when something has gone badly, in a way that assumes you can handle it. Every one of those is a small statement that the volunteer is inside rather than outside, and the cumulative effect of a hundred of them is the difference between a helper and a member.

The reverse is also cheap and constant. A long-serving volunteer who learns from the newsletter that the program she has run for four years is being restructured has just been told exactly where she stands.

Role identity means she says the words herself

The long-term predictor is the hardest to talk about without sounding vague, so let me put it in concrete terms. Role identity has arrived when the volunteer, asked at a family dinner what she does, mentions you unprompted. When she says “I’m one of the coaches at the club” rather than “I help out sometimes.”

That shift is not sentimental. It is the strongest available protection against the ordinary drift that ends most volunteering. Somebody whose Thursday is a task will drop it when work gets busy. Somebody for whom Thursday is part of who she is will move the work.

You can build this on purpose. Three things do most of it.

Give the role a name and let her use it. Not “volunteer” but “duty officer,” “safeguarding lead,” “Tuesday shift lead.” A name is a claim a person can make about herself. A generic label is not.

Let her speak for you. The volunteer who represents the organization to somebody outside it has to describe the organization as hers to do the job at all. Send her to the school assembly. Let her do the induction talk. Put her on the stand at the county fair.

Ask her to bring people in. Hager and Brudney’s 2004 study for the Urban Institute found that recruitment of volunteers by other volunteers was one of four practices positively associated with retention, alongside recognition, training and matching people to tasks. I suspect part of why it works is that it runs in both directions: you gain the recruit, and the recruiter has just publicly identified herself with you.

Most programs never get to stage three

The 2023 Volunteer Management Progress Report from Tobi Johnson & Associates, with 1,247 completed responses across 36 countries, put average volunteer tenure at 20 to 30 months. It also found 33.8 percent of organizations keep their volunteers less than ten months.

Set that against the stage model and you get an unpleasant piece of arithmetic. A third of programs lose people before organizational commitment has finished forming. The average program has a volunteer for long enough to reach the identity stage and mostly does nothing to build it, because the whole apparatus is still pointed at the new arrival.

I have watched this from the other end. The international camps I have been involved with run to more than five thousand participants and are staffed nearly entirely by volunteers, and the thing that makes that possible is not recruitment. It is that a large number of people arrive already describing themselves by the job they hold there. They are the sound crew. They have been the sound crew for eleven years. Nobody has to persuade them in March; the calendar is already defended. Building that took decades of handing people named roles and then leaving them alone inside them.

What to stop doing to your veterans

Two habits are worth dropping.

The first is treating extra contact as care. Weekly check-ins are appropriate for somebody in week three and faintly insulting to somebody in year five, and Hager and Brudney found supervision and communication were negatively associated with retention, which is a correlation rather than a cause but not one to ignore. If you would not do it to a colleague of fifteen years’ standing, do not do it to her.

The second is spending your recognition budget on things that cost her time. Phillips and Phillips found in 2010 that experienced volunteers rated time-consuming rewards such as trips and conferences significantly lower than inexperienced volunteers did. An awards dinner forty minutes away on a Thursday is a bill, not a gift, and I have written more about what recognition actually lands.

The honest limitation

Stage models are tidy and people are not. Nobody crosses from satisfaction to commitment on a particular Tuesday, and plenty of long-serving volunteers stay for reasons the model does not capture, including that their friends are there and it is a fixed point in a week that needs one. The research base here is also smaller than it should be: a handful of studies, mostly outside the United States, mostly correlational.

What survives all that hedging is the practical instruction, and it is a narrow one. Whatever you are doing for volunteers in their first ninety days, and there is a good deal worth doing there, do not assume it is the right thing to keep doing in year three. Ask a six-year volunteer what she would want that she is not currently offered. She will tell you, and it will almost never be more training.