Recognition works. The annual dinner mostly doesn't.

A volunteer who gives you three hours on a Saturday morning drives forty minutes on a Thursday evening to a hotel function room, eats a chicken dinner, sits through eleven minutes of speeches and a slideshow, collects a certificate with her name spelled slightly wrong, and gets home at half past ten.

She has now given you roughly four hours in order to be thanked for the forty she gave over the year. The certificate goes in a drawer.

I am not against thanking volunteers. I am against this particular way of doing it, and there is enough evidence to argue about it properly.

Recognition itself is not the problem

Start with the case in favor, because it is real.

Hager and Brudney’s 2004 study for the Urban Institute found recognition activities were one of four management practices positively associated with volunteer retention, alongside training, screening and matching people to tasks, and recruitment by other volunteers. And there is one decent measured example of awards specifically. The Indiana University Public Policy Institute, working with the Boy Scouts of America Crossroads of America Council and publishing in March 2025, surveyed 4,943 volunteers and found that those who had received an award were 1.43 times more likely to still be volunteering. Turnover in that population was 13 percent across 2016 to 2017.

That is a meaningful difference, not a rounding error, and it comes from a real program rather than a lab.

It is also correlational, which cuts in an obvious direction. Awards go to people who did more, and people who did more were probably going to stay anyway. The finding could be measuring commitment that already existed rather than commitment the award created. The authors also flag an unexplained result from the same survey that should make everyone cautious about reading too much into any single number here: women who received training were 1.41 times more likely to leave than men who received training, and nobody has a good explanation. They call for further research. That is the right response.

So: recognize people. The question is what form of it earns the effort.

The ranking that should change your budget

In 2010 Phillips and Phillips published a study in the Journal of Business and Retail Management Research in which 148 respondents, 95 percent of whom had volunteer experience, rated 31 possible volunteer rewards.

The four top-rated rewards were meeting the people they served, a free meal at the volunteer site, thank-you notes from the people served, and thank-you notes from the organization.

Look at what those have in common. Every one is cheap. Three of the four connect the volunteer to the people the work was for. The single highest-rated item is not a reward in any normal sense of the word: meeting the people you served is just evidence that the work happened.

The study has limits worth stating. It surveyed university students, so it skews young and it measures stated preference rather than observed behavior. But the pattern is coherent with a much larger dataset. NCVO’s Time Well Spent 2023, from YouGov fieldwork with 7,006 UK adults, asked continuing volunteers why they carry on. The organization or group itself came out at 46 percent, tied with the difference they felt they were making. Commitment to the cause was 41 percent.

Awards do not appear on that list. The difference they are making does, at the top.

Evidence beats celebration

Here is the distinction I would build a recognition program on. Most recognition celebrates the volunteer. Better recognition evidences the work.

Celebration says: you are wonderful, here is a certificate. It is about the volunteer’s qualities, it is awarded by someone above them, and it is at least partly about the organization performing gratitude in public.

Evidence says: here is what happened because you were there. It is about the work, it is verifiable, and it does not require the volunteer to accept a compliment, which a surprising number of people find uncomfortable.

Evidence is also usually cheaper and always more specific. “Thanks for everything you do” costs nothing and lands as nothing. Telling a driver how many households her own Tuesday run reached last month, and that the two she added in March are both still on the list, costs one email and cannot be said to anybody else.

Some concrete forms that work.

Numbers attached to their shift, not the program. Program-level totals belong in the annual report. What a volunteer wants is the number their own hands produced.

Words from the people served, passed on unedited. A forwarded message with a name on it beats anything printed on card. This is literally two of the four top-rated rewards in the Phillips study.

Telling them what happened after they left. The family got housed. The kid came back the next week. Volunteers are usually removed from outcomes by the shape of the work, and closing that loop is free.

Putting them in the room with the people they helped, where it is appropriate and consented to. The highest-rated reward in the study, and most programs treat it as a logistical inconvenience.

The dinner, specifically

I have picked on the annual appreciation dinner because it is the default and because it fails on its own terms.

Phillips and Phillips found that experienced volunteers rated time-consuming rewards such as trips and conferences significantly lower than inexperienced volunteers did. That is the finding to write on the wall of your planning meeting. The people who have given you the most time are the people least interested in a reward that costs them more of it. A gift that takes an evening from someone whose gift to you is evenings is not a gift, it is an invoice with a nice tablecloth.

Run the arithmetic the way you would for any other program cost. Forty volunteers, three hours each including travel, is 120 hours. At the value of volunteer time published by Independent Sector with the Do Good Institute for calendar year 2025, $36.14 an hour, that is about $4,300 of donated labor, on top of what you spent on the room and the chicken. Nobody ever puts that on the budget line, because volunteer time is treated as free right up until someone calculates it.

The objection I hear is that some volunteers love the dinner, and that is true. Clary and colleagues identified a social function in volunteer motivation, the desire for relationships and for being valued by a group you belong to, and for people running high on that, an evening together is the point. I would not cancel it. I would do three things to it: make attendance obviously optional with no reputational cost, stop making anyone sit through a program in order to receive a piece of paper, and stop calling it the recognition strategy. It is a party. Parties are good. It is not a substitute for telling people what their work did.

You do not need to profile anyone

One more finding from Phillips and Phillips that saves you work. Volunteers with very different motivational profiles turned out to have remarkably similar reward preferences, and the authors concluded the cost of profiling volunteers to tailor rewards may not be worth it.

I find that quietly liberating. There is an industry of advice telling you to segment your volunteers and personalize their recognition, which sounds sophisticated and in practice means a spreadsheet nobody maintains. The available evidence says most people want roughly the same four cheap things. Do those four for everybody and put the saved effort into making the work itself worth doing.

What scale taught me about this

The organizations I have worked in involve nearly ten thousand volunteers across around eighty countries. At that size an event-based thank-you is arithmetically impossible, and finding that out was useful, because it forced a different question: what actually travels?

What travels is the result. A volunteer on a camera position for a live production going out to around eighty countries in more than twenty languages can see the output. She knows what she contributed and she can point at it. That is why those positions hold people for years, and it has nothing to do with anyone thanking her.

The lesson generalizes downward. Anywhere a volunteer can see the finished thing, recognition is mostly taking care of itself. Anywhere they cannot, someone has to carry the evidence back to them, and that job is more important than the certificate budget.

Where this evidence is thin

I would rather you knew the state of the research than took my word for it. The recognition literature is much weaker than the retention literature generally. Phillips and Phillips is a student sample rating hypothetical rewards. The Boy Scouts study is one council, with a 13.7 percent response rate, and correlational. Hager and Brudney’s association between recognition and retention is twenty years old and also correlational.

What I would say with confidence is narrower than “do these things.” It is this: nothing in the evidence supports spending your recognition budget on expensive, time-costly, ceremony-shaped events, and several separate lines of evidence point at cheap, specific, impact-carrying messages instead. If your program is doing the first and not the second, you are spending money and your volunteers’ evenings to get a worse result than an email would produce. That is worth fixing before the next dinner, particularly for the long-serving people the dinner costs the most.