What Funders Are Really Looking for When They Read Your Impact Report

A nonprofit submitted a renewal application with three years of clean outcome data, a professionally designed report, and a testimonial video. The renewal did not come through. When the executive director asked why, the program officer's answer was simple: the report told her what happened, but never what the organization learned from it.

Reports fail funders not because they lack information, but because they lack argument.

The Center for Effective Philanthropy found that 87 percent of foundation leaders reported increased demand for grant funding, even as government support for many grantees declined. CEP also found that most foundation leaders believe they understand grantee challenges, while only about half of nonprofits feel that same understanding. Better charts will not close that gap. Clearer thinking will.

A report's job is not to present evidence. It is to make an argument, with evidence as support. Data and stories are raw material. Interpretation is the finished product. A statistic tells a funder what happened. It does not tell them what that means for the theory of change or what the program should do next. A youth mentoring program reporting a 20 percent increase in school attendance has produced a finding. An organization that explains what changed in delivery to produce that number has produced meaning. Funders can pull numbers from a dashboard. They read a narrative report because they want the organization's expert read.

Recommendation: Before writing any results section, have program staff answer this: what does this tell us that we did not already know, and what will we do differently? If the answer is thin, the metric does not belong as a headline.

The same logic applies to storytelling. Testimonials are often treated as the emotional counterweight to data, present to humanize an otherwise clinical report. Warmth is not the same as persuasion. A story becomes persuasive when it works as evidence for a specific claim, not decoration around one. The difference between a moving anecdote and a useful one is whether the reader can say, afterward, exactly what it proved.

Recommendation: State the claim a story supports before it appears. If you cannot tie a narrative to a specific argument, it belongs in a newsletter, not an impact report.

Format follows the same principle. Program officers do not have time to reconstruct an argument from dense documentation. They are scanning for three things: what changed, what caused it, and what that says about organizational judgment. Reports that lead with data and save interpretation for the end ask the reader to do work the writer should have done first.

Recommendation: Lead every section with the interpretive claim, then support it with data.

This includes the parts that did not go well. An organization that names a metric that fell short, explains why, and describes its response signals something a report full of favorable numbers cannot: that the team is paying attention. Candor reads as competence. A flawless narrative reads as incomplete.

Recommendation: Include at least one section on an underperforming metric, framed around cause and response, not justification.

None of this happens at the writing stage. The most common mistake is handing raw program data to communications staff and asking for a polished narrative. Deciding what a finding means requires someone with real authority over the program. When that judgment arrives only at the editing stage, the report ends up well written and analytically thin.

Recommendation: Get a senior program leader involved before drafting begins, not after.

The impact report is not an annual task to get through. It is a test of whether an organization understands its own work. Teams that build interpretation into how they think about outcomes find that clarity long before the next report is due. Handled well, it tells a funder more about organizational judgment than any metric could.

What is one finding from your last report that deserved more interpretation than it got?