The Nonprofit Data Trap: How Digital Accountability Is Reproducing Inequality

When a small tenant-rights organization in Chicago applied for a foundation grant last year, the staff spent three weeks compiling their impact report. They counted hotline calls. They tracked “housing units stabilized.” They built a dashboard. What they could not count—and what the founder could not measure—was the moment a formerly undocumented woman finally trusted them enough to call after 18 months of silence. They did not get the grant.

This is not an isolated story. Research on nonprofit funding and evaluation has documented how smaller organizations can struggle with the demands of funder reporting. The Center for High Impact Philanthropy at the University of Pennsylvania notes that smaller nonprofits often have limited capacity for data collection and management, and that reporting different measures to multiple funders can waste resources and distract organizations from their missions.

Across the American nonprofit sector, a quiet shift has been underway for more than a decade. Foundations and philanthropic institutions increasingly require quantified evidence of success: dashboards, key performance indicators, third-party evaluations, data management systems, and sophisticated reporting infrastructure. These demands are framed as transparency, accountability, and evidence-based practice. In reality, they are restructuring the funding landscape in ways that systematically disadvantage the smallest and most community-rooted organizations, the ones that often do the most transformative work.

The result is what we might call the nonprofit data trap: a self-reinforcing cycle in which the organizations best positioned to secure funding are those with the staff capacity, technical infrastructure, and institutional fluency to translate grassroots work into funder-legible outputs, while community-based organizations without these resources are quietly screened out. The trap is not intentional. But it is structural. And it is deepening.

The Rise of Metrics Culture in Philanthropy

The compliance burden is not evenly distributed. It falls heaviest on those least able to bear it.The demand for quantifiable impact did not emerge from nowhere. It has roots in the philanthropic reform movements of the late 1990s and early 2000s, when a generation of tech-sector donors, flush with new wealth and enamored with the efficiency logic of Silicon Valley, began applying return-on-investment frameworks to charitable giving. The movement went by names such as venture philanthropy, effective altruism, strategic philanthropy. Its core premise was that rigorous measurement would separate programs that “worked” from those that did not, and that funding should flow accordingly.

There is a surface logic to this. Accountability matters. Waste matters. Organizations should be able to explain what they do and why it helps people. These are not unreasonable demands in principle.

But the operationalization of these principles has produced something more troubling. Sophisticated grantmakers now routinely require applicants to demonstrate theory-of-change documentation, logic models, baseline and endline data collection, and randomized or quasi-experimental evidence of impact. Some require applicants to use specific data platforms—Salesforce, Apricot, Exponent Case Management—before they can even apply. For large nonprofits with dedicated evaluation departments and data analysts on staff, these requirements are manageable, if costly. For a four-person immigrant services organization operating on a $200,000 budget, they can be existentially draining.

When funders insist on certain categories of evidence, they are not simply asking organizations to count things. They are determining which things count.

A 2023 survey by the National Council of Nonprofits found that small and medium-sized organizations with annual revenues under $1 million spend a disproportionate share of their administrative capacity on funder reporting. Executive directors of color reported that reporting requirements were among the top three barriers to organizational sustainability, alongside inadequate overhead funding and unpredictable grant timelines. The compliance burden is not evenly distributed, and falls heaviest on those least able to bear it.

What Gets Lost in Translation

The problem is not only logistical. It is epistemic. When funders insist on certain categories of evidence, they are not simply asking organizations to count things. They are determining which things count.

Consider the work of political education: teaching low-income residents about housing law, labor rights, or immigration policy. This is not a service that maps easily onto a metric. You cannot randomize it. You cannot run a control group of people who did not learn their rights. The “outcome” of political education may not manifest for months or years, and when it does, it will look like a tenant filing a complaint, a worker joining a union, or a community member running for city council. These effects are diffuse, cumulative, and deeply resistant to attribution. They also happen to be some of the most powerful forms of social change.

The same is true of trust. Organizations that work with communities traumatized by government surveillance, policing, or immigration enforcement know that trust is not a soft outcome—it is a precondition for everything else. A health clinic that serves undocumented clients must spend years building the kind of trust that allows a patient to mention, quietly, that they have been living with pain for years because they were afraid to seek care. That clinic’s impact report will record a medical encounter. It will not record the years of relationship building that made the encounter possible.

Care work, solidarity, and accompaniment are not easily measurable, but they are not therefore unmeasurable in the ways that matter most to the people they serve. The problem is that the people they serve rarely write the rubrics.

Power and the Definition of Impact

Lurking beneath the data demands is a question of power: Who gets to define what counts as impact, and whose knowledge is treated as credible?

In practice, the answer to both questions tends to be the funder. Foundation program officers, often highly educated, often working from offices in major metropolitan areas, often without lived experience of the communities they fund, develop the criteria that organizations must meet to access resources. This is not malice. It is structural. But the effect is that communities themselves are largely excluded from the process of defining what would constitute meaningful change in their own lives.

Participatory grantmaking—a set of practices in which communities have direct say in funding decisions—offers a partial corrective. Organizations like the Headwaters Foundation for Justice, the Wellspring Philanthropic Fund, and Brooklyn Org (formerly the Brooklyn Community Foundation) have experimented with models that shift decision-making authority to community members, trust-based grantmaking approaches that reduce reporting burdens, and multiyear general operating support that allows organizations to determine their own accountability structures.

The evidence from these experiments is promising. Grantees report increased organizational stability, reduced burnout among leaders, and greater capacity to do deep community work rather than funder compliance work.

But these approaches remain marginal within the broader philanthropic ecosystem. Most major foundations continue to operate within accountability frameworks that were designed by and for large, professionalized nonprofits, and that treat grassroots organizations as junior partners who must prove their worth using someone else’s definitions of worth.

Who Survives the Trap

The cumulative effect of these dynamics is a funding landscape shaped less by community need than by organizational capacity. The organizations that thrive under data-intensive accountability regimes are, predictably, the ones that have the resources to build data-intensive accountability capacity: large nonprofits with national footprints, organizations with foundation-funded evaluation positions, and newer entrants that were designed from the beginning around funder expectations.

The organizations that struggle are disproportionately Black-led, Indigenous-led, and immigrant-led grassroots groups that emerged from within communities rather than from the nonprofit-industrial complex, and that have developed trust-based, relationship-centered models of work that do not fit neatly into a logic model. Their work is not less rigorous. It is differently rigorous, rooted in community knowledge, responsive to local conditions, and accountable first to the people they serve rather than to external funders.

When these organizations close—and many do close, quietly—for lack of funding,   the losses are enormous and largely invisible. They lose not just programs but relationships, knowledge, networks, and the irreplaceable credibility that comes from years of presence in a community. No dashboard captures the impact of that loss.

Toward Accountability That Includes Everyone

Fixing the nonprofit data trap will require not just technical adjustments to reporting requirements but a willingness to redistribute epistemic authority.

None of this is an argument against accountability. It is an argument about whose accountability frameworks we use, and whether those frameworks are themselves accountable to the communities they purport to serve.

There are concrete reforms the philanthropic sector could pursue. Funders could implement tiered reporting requirements that scale with organizational size and capacity. They could fund data and evaluation infrastructure as legitimate overhead, rather than treating it as a sign of organizational bloat. They could adopt trust-based grantmaking principles—multiyear, unrestricted funding; streamlined applications; no unsolicited site visits—that have been tested and validated by a growing body of evidence.

Most fundamentally, funders could recognize that the current system does not simply measure impact neutrally. It produces a particular kind of impact: one that is legible to funders, manageable by professionalized nonprofits, and increasingly disconnected from the knowledge and priorities of the communities at the center of the work. Fixing the nonprofit data trap will require not just technical adjustments to reporting requirements but a willingness to redistribute epistemic authority—to take seriously the idea that communities themselves are the most qualified evaluators of their own wellbeing.

The tenant-rights organization in Chicago eventually found a funder willing to provide two years of unrestricted general operating support. They stopped building dashboards. They hired a community organizer. And the woman who had been afraid to call for 18 months? She did call. She organized her building. She won.

That is impact. It just does not fit in a spreadsheet.