There's a question that surfaced repeatedly in our research engagements this year, sometimes spoken aloud, more often hovering beneath the surface: Who is this data actually for?

It's a deceptively simple question. And in 2025, it finally started getting the attention it deserves.

For decades, the social sector has operated within an extractive research paradigm—one where communities are studied, findings are reported upward to funders, and the people whose lives generated the data rarely see it again in any usable form. This year, we watched that model buckle under the weight of its own contradictions.

The accountability inversion

Funders are facing a legitimacy question they can no longer avoid: if the goal is community wellbeing, why does the evidence pipeline flow almost exclusively toward institutional decision-makers?

We've seen early movement toward what some are calling downward accountability—the idea that organizations owe their data to the communities it describes, not just the funders who paid for its collection. In practice, this means rethinking everything from consent processes to reporting formats to who gets to interpret findings.

It's messy. It slows things down. And it's long overdue.

Several of our engagements this year involved redesigning evaluation frameworks with community validation built into the process—not as a checkbox, but as a structural requirement. The findings don't leave our hands until the people they represent have had a chance to contest, contextualize, or correct them.

The quiet crisis in organizational data fluency

Here's an uncomfortable truth: most mission-driven organizations are sitting on more data than they know what to do with. Client intake forms, program attendance logs, survey responses, anecdotal feedback—it accumulates in spreadsheets and filing cabinets and intake software, largely untouched.

The problem isn't access. It's capacity. Teams are stretched thin, and the skills required to move from raw information to actionable insight—data cleaning, analysis, visualization, interpretation—are rarely funded or prioritized.

This year, we've shifted a significant portion of our work toward what we're calling data fluency building: helping organizations develop internal muscles for working with their own evidence. Not outsourcing the analysis, but scaffolding the capability to do it themselves.

The goal isn't to turn program managers into statisticians. It's to demystify the process enough that data becomes a tool teams actually use, rather than a compliance burden they endure.

Needs assessments got honest

The community needs assessment has long been a genre unto itself—a ritualized document that tells funders what they expect to hear while telling communities very little they don't already know.

In 2025, we saw growing pushback against this performance. Organizations are asking harder questions before launching into data collection: What do we actually not know? What decisions will this research inform? And perhaps most importantly—has someone already done this work?

That last question matters more than it might seem. The duplication of research in the social sector is staggering. Communities are surveyed repeatedly by different organizations asking nearly identical questions, each producing reports that rarely reference each other. The burden falls on the same populations, over and over, with diminishing returns for everyone involved.

We've started building evidence inventories into our discovery phases—mapping what's already known before designing new collection. Sometimes the most valuable research finding is that the research has already been done.

Qualitative data is having a moment

The sector's historical preference for quantitative metrics—numbers that can be aggregated, compared, and dropped into dashboards—is softening. Not disappearing, but making room.

There's growing recognition that the most important changes often resist quantification. How do you measure a shift in someone's sense of belonging? The restoration of trust between a community and an institution that harmed it? The slow, nonlinear process of healing?

This year, we've seen increased appetite for narrative-based evidence: case studies, journey maps, story banking, and participatory sense-making processes that honor complexity rather than flattening it. Funders are beginning to accept that "we interviewed twelve people and here's what they told us" can be as rigorous as a sample size in the hundreds—if the methodology is sound and the interpretation is honest.

The ethics of "good enough" evidence

Research purists will bristle at this, but it needs to be said: in community-facing work, methodological perfection is often the enemy of timely action.

Organizations are making decisions every day—about programs, resource allocation, strategic direction—whether or not they have peer-reviewed evidence to guide them. The question isn't whether to act with incomplete information; it's how to act responsibly with incomplete information.

We've been working with clients to develop what we call evidence thresholds: explicit agreements about what level of confidence is required for different types of decisions. A pilot program doesn't need the same evidentiary standard as a multi-year strategic investment. A community already telling you what it needs doesn't require a survey to confirm it.

This isn't anti-rigor. It's right-sized rigor. And it frees organizations to move at the speed their communities require.

What we're watching for next year

The sector is at an inflection point. The old research paradigm—extractive, upward-accountable, methodologically conservative—is losing its grip. But what replaces it remains contested.

We're paying attention to a few emerging threads: the growth of community-owned data infrastructure, the integration of Indigenous data sovereignty principles into mainstream practice, and the slow but meaningful shift in funder attitudes toward qualitative and participatory evidence.

The through-line in all of it is a redistribution of power—over what counts as knowledge, who gets to produce it, and whose interests it serves.

Research, at its best, is an act of care. A way of saying: your experience matters enough to document, to understand, to learn from. The sector is starting to remember that.