When 100% Should Have Said 100%: What a Box Truck Full of Furniture Taught Me About Participatory Evaluation
Good evaluation starts before anyone agrees on the questions.
I was doing my initial research for a new client, an organization that provides an entire houseful of furniture to families experiencing hardship, when I found something that didn't add up.
I was reading through their past grant applications, the way I always do before I write a word of anything new, and I ran into a table of outcomes data. Sixty-eight percent of families reported that receiving furniture relieved financial pressure. Fifty-seven percent reported it allowed them to experience a new beginning. Seventy-nine percent said they'd been sleeping on the floor before the furniture arrived. Eighty percent had no table to eat at. Sixty-seven percent had no usable furniture at all.
I stopped and read it again. Something was off.
Think about what's actually happening here. These families didn't just get a couch. They got an entire household of furniture, delivered by box truck, that they themselves had gotten to choose walking through a gently-used furniture store. If someone did that for my family, I can tell you with total confidence that 100% of me would report relief from financial pressure. Every single time. So why was the number 68%? And what does "experience a new beginning" even mean as a stand-alone data point, and why wouldn't that also land at 100%?
So I did what I always tell my students to do when a number doesn't sit right. I went looking for where it actually came from.
A form can only ever measure what it thought to ask.
It took some digging, but I found the source. It was a short survey, handed to families right after they'd finished picking out their furniture, while they were still standing in the store, before the truck had even been loaded. The question read something like: Which of the following is true for your family? Then a list of options: relieves financial pressure. Allows us to experience a new beginning. And so on.
Here's the problem. Nothing on that survey said "check all that apply." So families did what any reasonable person does when a form doesn't say otherwise. They picked one option, maybe two, and moved on with their day. Which meant the resulting statistics didn't actually mean only 68% of families felt financial relief. They meant 68% of families happened to check that particular box, out of several true statements they could have chosen from if the form had let them.
The data wasn't dishonest. It wasn't even really wrong, exactly. It was just measuring the wrong thing, because the tool that produced it had been built by someone who never had to fill it out.
Table of Contents
What Participatory Evaluation Is (and Isn't)
That last sentence is really the whole argument, so let's sit with it for a second. No one who filled out the survey helped write it. If they had, someone would have said, out loud, in the room: "wait, can I only pick one?" And the flaw would have been caught before it ever became a data point in a grant report.
That's the gap participatory evaluation exists to close. At its simplest, participatory evaluation means the people closest to a program, the ones actually receiving services, doing the work, or living the outcomes, are involved in designing how that program gets measured. Not just supplying the data afterward. Designing the tool itself.
That distinction matters, because it's easy to mistake participatory evaluation for something it's not. It isn't a focus group you run after your logic model is already finalized so you can say you "got community input." It isn't a satisfaction survey, however well-intentioned. Those are forms of listening, and listening is good, but listening to people about a tool you already built is fundamentally different from asking them to help build the tool.
The flaws you catch in the room are the ones that never make it into the data.
Participatory evaluation says: before we decide what "success" looks like on paper, let's ask the people it's supposed to describe whether that's actually what success looks like to them. Sometimes that means co-designing survey questions with former or current program participants. Sometimes it means an advisory group of community members who review evaluation plans before they go into a proposal. Sometimes it's as simple as pilot-testing your measurement tool with three or four real people before you send it to three hundred.
And here's the part that matters most for how you write proposals, not just how you write evaluation sections: this isn't only an evaluation-section fix. It's a whole-proposal fix. When your evaluation plan is built around what the community itself calls meaningful change, that shift ripples backward through the whole narrative. You stop measuring outcomes on the organization's terms and start centering community conditions instead. A proposal that opens by describing what the community says it needs, and closes by measuring change in the community's own terms, is a fundamentally different document than one where the organization defines the need, delivers the service, and measures its own success on its own terms from start to finish. Participatory evaluation, done well, isn't a section you add. It's a lens that changes how the whole story gets told.
Why It Matters for Empowerment
Go back to the furniture families for a second. Imagine a different version of that survey. One where a few former recipients sat down with the organization beforehand and said, "Here's what we actually felt, and here's what a form should ask if it wants to capture that." Maybe they'd have said the choices shouldn't be either/or. Maybe they'd have said financial relief and a fresh start aren't competing feelings, they're the same feeling, described two different ways, and a good survey should let people say both. Maybe they'd have pointed out that the real story wasn't captured in any of those five options at all.
That conversation would have produced better data. But it also would have done something else, something that matters just as much as the data. It would have told those families their read on their own experience counted for something. Not as a nice gesture. As actual input that shaped what got measured and, by extension, what got reported to funders as meaningful.
That's the empowerment piece, and it's worth being precise about what it means, because the word gets used loosely. Empowerment in evaluation isn't a feeling you generate by asking people how they feel. It's a transfer of authority. It's the difference between an organization deciding what counts as success and then checking in with the community to see how they're doing against that standard, versus a community having a hand in defining what success even means in the first place. The first version keeps all the power exactly where it already was. The second one moves some of it.
What gets delivered is easy to count. What it means rarely fits in a number.
This is where a lot of well-meaning evaluation plans quietly fall short. An organization can genuinely care about the people it serves and still build every measurement tool from its own vantage point, because that's simply who's in the room when the tool gets built. Participatory evaluation is a deliberate correction for that. It doesn't ask you to trust the community more in the abstract. It asks you to put them in the room before the survey exists, not after.
And once you've done that, you can't fully go back to the old way. Once you've seen how a form written by the people who'll actually answer it looks different from a form written for them, the flaws in the old approach become obvious in a way they weren't before. That furniture survey didn't fail because anyone was careless. It failed because it never occurred to anyone that the people filling it out might have caught the problem in about ten seconds, if only they'd been asked.
When Compelling and Exploitative Look the Same From the Outside
That same furniture bank organization's data holds one more lesson, and it's worth sitting with. Look again at those original numbers. Seventy-nine percent of families said they'd been sleeping on the floor before the furniture arrived. Eighty percent had no table to eat at. Sixty-seven percent had no usable furniture in their home at all.
If families helped shape the survey that produced those numbers, and agreed that questions like these were fair and respectful to ask, then the resulting statistics rest on solid ground. The community had a hand in creating that data, knowing it would be used to describe the program's impact. That's participatory design working as intended.
Care given well deserves to be described well.
Where it gets more interesting is what a grant writer does with that statistic next. It's easy, and often effective, to build a scene around a number like seventy-nine percent: a child sleeping in a real bed for the first time, tucked in under sheets patterned with a favorite cartoon train. I'll admit where that detail actually comes from when I write scenes like this. It's not pulled from any family's file. It's pulled from my own life, from watching my son pick that exact pattern off a shelf when he was little. A well-built composite draws on real texture like that, generic and common enough that it isn't tied to any one traceable person, in order to help a reader feel what a statistic means rather than just register it as a percentage. That's not a liberty taken with a client's story. It's a craft skill, and one worth teaching, because a composite like this is often what makes a number compelling enough to move a funder to act.
Here's the test worth applying, and it isn't "could this ever resemble somebody real." Almost any ordinary detail could, given enough families. If a real family happens to have picked those same sheets, in that same shade of blue, that's coincidence, not appropriation, precisely because the detail was common and personal to begin with, not lifted from their specific circumstances without asking. The question that actually matters is where the detail came from. Did it come from a shared, ordinary pattern of childhood, the kind any parent might recognize? Or did it come from one particular family's private account, specific enough that they, or people who know them, would recognize their own story being used without permission?
That's why the loop back to participants matters here too, alongside co-designing the survey itself. An ongoing relationship with the families a program serves is what makes it possible to ask, when a real, identifiable story is being considered rather than a composite drawn from common life, whether that family is comfortable with how their circumstances might appear in writing. A composite doesn't need that permission, because it represents a pattern, not a person. A real, identifiable story does.
Why Funders Increasingly Want to See This
Funders have been quietly shifting expectations for a while now, and if you've written proposals for even a few years, you've likely felt it without necessarily naming it. There's more language in RFPs about community voice, co-design, and lived experience. There are more evaluation sections that ask not just what you'll measure, but who was involved in deciding what to measure. This isn't a passing trend. It's part of a broader move toward trust-based philanthropy, which starts from the premise that the people closest to a problem usually understand it better than the people writing checks to solve it.
From a funder's chair, this makes sense. A funder reading two evaluation plans, one built entirely by program staff and one built with input from the people the program actually serves, is reading two very different claims about how well an organization understands its own impact. The second plan signals that the organization isn't just delivering a service and assuming it works. It's checking, with the people who'd actually know, whether it works the way they think it does.
Somewhere, someone is reading closely enough to notice the difference.
There's also a more practical reason funders like this. Participatory evaluation tends to catch problems earlier and cheaper than the alternative. The furniture organization's flawed survey didn't get caught by an external evaluator or a funder's site visit. It got caught by a grant writer reading through old reports and noticing the numbers didn't make sense. A participatory design process catches that kind of flaw before it ever produces a bad statistic, which means fewer awkward conversations later about why last year's numbers don't match this year's story.
None of this means you should frame participatory evaluation as a box to check because funders want to see it. That's the wrong reason, and reviewers can usually tell the difference between a genuine practice and a phrase borrowed from an RFP. But it's worth knowing that when you build a proposal this way because it's the right way to build it, you're also building exactly the kind of evaluation plan that's earning more attention and more trust from the funders paying closest attention right now.
How to Build It Into a Proposal From the Start
Here's where the abstract becomes practical. If you're convinced participatory evaluation belongs in your next proposal, here's what that actually looks like on the page and in the process that produces it.
Start earlier than you think you need to. Participatory evaluation doesn't work if it's bolted onto a nearly-finished evaluation section. It has to happen while the logic model is still being built, ideally before you've written a draft of what success looks like. If your client organization already has a program advisory board, a client council, or even a handful of former participants they stay in touch with, that's your starting point. If they don't, part of your job may be helping them build that relationship before the proposal deadline, not just for this application, but as ongoing practice.
Ask a narrow, specific question rather than a broad one. "What do you think we should measure" is hard for anyone to answer cold. "Here are three ways we're thinking about asking whether this program helped. Which of these actually captures what changed for you?" is something a real person can respond to in fifteen minutes. Bring drafts, not blank pages.
Write the co-design into the narrative itself, plainly. Funders can't give you credit for a process they can't see. A sentence like "the evaluation questions below were developed with input from six program alumni during a two-hour design session in March" does real work. It's specific, it's verifiable, and it shows exactly the kind of practice this whole article has been building toward.
Build in the feedback loop as a stated part of the plan, not an afterthought. If your evaluation section describes how you'll collect data, it should also describe how results get reported back to participants and how their response to those results feeds into program adjustments. That's the difference between an evaluation plan and a genuinely participatory one.
And finally, budget for it. Participatory design takes staff time, sometimes stipends for participant advisors, sometimes translation or accessibility support to make sure the people you're asking can actually participate fully. If it's worth doing, it's worth a line item, not an unpaid extra task piled onto an already stretched program team.
What This Looks Like in Practice
Participatory evaluation is the umbrella. Underneath it, a few specific methods are especially good at putting the philosophy into action, and they're worth knowing by name since they show up more and more in funder-friendly evaluation plans.
Photovoice asks participants to document their own experience through photography, then discuss what those images mean in their own words, often in a group setting. It's especially effective when the people involved might not feel as comfortable expressing something in a written survey as they would pointing a camera at what matters to them and explaining why.
Photovoice: participants document their own experience through photography, then explain in their own words what the image means.
Ripple Effect Mapping brings a group together to visually trace the effects of a program, including the unexpected ones, through facilitated group reflection. Instead of an evaluator deciding in advance what outcomes to look for, the group maps out what actually happened, including the effects nobody predicted, which often turn out to be some of the most compelling parts of a funder report.
Both of these pair naturally with the broader evaluation toolbox we covered in Beyond the Survey: 16 Evaluation Tools Every Grant Writer Should Have in Their Back Pocket, where we set participatory evaluation aside specifically to give it this fuller treatment. If you haven't read that piece yet, it's a good next stop for rounding out your evaluation toolkit.
The common thread across all of these methods is the same one that's run through this entire article. The people living the outcome help shape how the outcome gets seen.
Common Pitfalls
A few ways this goes wrong are worth naming plainly, because good intentions don't automatically prevent them.
Tokenism is the most common. This is the single focus group held once, after the real decisions are already made, so the organization can honestly say community members were consulted. Real participatory evaluation shapes the tool. Tokenism just decorates a tool that was already finished.
Extractive consultation is a close cousin. This is when participants are asked to give their time, their stories, and their insight, and receive nothing in return, not payment, not results, not any say in what happens next. If the relationship only runs one direction, from the community to the organization, it isn't participatory. It's data collection wearing a participatory label.
Overcorrecting into paralysis is a subtler pitfall. Some organizations, once they understand the stakes, become so worried about getting community input wrong that they stop asking altogether, or ask so cautiously that the questions become meaningless. Imperfect, genuine participation beats a perfectly-worded process that never actually happens.
And finally, treating the loop as optional. It's easy to build a good participatory design process for the survey itself and skip the harder, ongoing work of reporting results back and adjusting based on what participants say. That loop isn't a nice extra. It's the part that turns a one-time consultation into an actual, ongoing relationship of trust, which is the entire foundation the rest of this depends on.
FAQ
What's the difference between participatory evaluation and just getting community feedback
Feedback usually happens after a program or tool is already built, and it's optional for the organization to act on. Participatory evaluation means the people affected help design the measurement tool itself, before it's finalized, and stay involved in interpreting and responding to what it finds.
Does participatory evaluation take more time than a standard evaluation plan?
Usually yes, especially upfront. Co-design sessions, relationship-building, and closing the feedback loop all take real time and often real budget. Most organizations find it pays for itself by catching flawed measurement tools before they produce years of bad data.
Can I use participatory evaluation with a program that's already running?
Yes. It's easier to build in from the start, but an existing program can still bring participants in to review and improve its current evaluation tools, and to help interpret results going forward.
Is participatory evaluation only for direct-service programs, like the furniture example?
No. The same principle, that people closest to an outcome help define how it's measured, applies to advocacy work, systems change initiatives, and community organizing programs just as much as direct service.
How do I describe this in a proposal without it sounding like a buzzword?
Be specific. Name who was involved, how, and when, rather than using the phrase "community-informed" as a general claim. Specificity is what separates a genuine practice from borrowed language.
