The tension nobody names
A funder asks what difference the money made. The organisation opens the form two weeks before the deadline and finds attendance registers, some photographs, a spreadsheet somebody started in March, and a folder of quotes with no dates. So the report describes activity in careful detail, adds two moving quotes at the end, and hopes the reader joins the dots.
That is not dishonesty. It is what happens when the evidence a report needs is a different kind of thing from the evidence delivery naturally produces. Understanding what the reader is looking for helps, because it turns out to be a shorter and more reasonable list than most people fear.
This applies well beyond grant funders. A commissioner reviewing a contract, a board reading an annual report, a members’ meeting, a local authority partner, all of them read with roughly the same questions in mind.
The five things being assessed
Nobody publishes this as a scoring grid, and every funder has its own emphasis. But across grant reports, contract reviews and trustee scrutiny, these five questions come up again and again.
Did the money go where it was supposed to?
The plainest question and the first one asked. Spend broadly matching the budget, with any variation explained rather than buried.
Did the delivery happen at the scale promised?
How much you said you would do, how much you did, and an honest note where the two differ.
Is there evidence of change, not just activity?
Something showing a difference for the people you work with, and a note of where that something came from.
Are you honest about what you don't know?
Named gaps, response rates, and the questions still open. This reads as competence, not as weakness.
Is there a credible plan for what's next?
What you learned, what you will change because of it, and what you will collect that you did not have this time.
What these have in common
None of them asks for a large number. Four of the five are about clarity rather than results: was the money used as agreed, did the work happen, do you know where your evidence is thin, and do you know what you would do differently. Only one asks for evidence of change, and even that one asks for evidence, not for a triumph.
A report that is modest and specific outperforms a report that is impressive and vague, almost every time.
The biggest credibility killer
Overclaiming. A confident figure with nothing solid behind it. “92% of participants improved their wellbeing” from eleven forms returned out of sixty people. “We lifted 200 families out of poverty.” A programme described as proven after two terms.
These claims usually survive the first read. The problem is the second one. A trustee with a research background, a funder’s programme officer who reads forty of these a year, or a monitoring visit that asks a simple follow-up: how many people did that percentage come from? Who asked them? When? What about the ones who stopped coming?
If the answer is thin, the damage is not confined to that figure. Everything else in the document is quietly reclassified as unverified. That is the real cost, not one disputed number, but a reader who no longer takes the rest at face value. And in a sector where the same people move between funders, that reputation travels.
What honesty sounds like
The alternative is not weak. It is specific. “We asked 60 participants and 22 responded. Of those, 18 described feeling more confident asking for help. We do not know how the 38 who did not respond would have answered, and we would expect them to be less positive on average.”
That paragraph is unglamorous and it is far more persuasive than 92%, because a reader can see exactly what it rests on. It also makes a statement like “we do not have enough evidence on this yet” a sign of a well-run organisation rather than an admission of failure. You are telling the reader you know the difference between what you can show and what you hope.
- Give the denominator every time. A percentage without the number behind it invites the question you least want.
- Separate what you delivered from what changed. Do not let one paragraph imply the other.
- Say 'contributed to', not 'caused', unless you have an evaluation design that supports the stronger word.
- Name one thing that did not work. It costs you nothing and buys a great deal of trust.
A checklist before you send
Run through this with the finished draft in front of you. It takes about ten minutes and catches most of what gets reports questioned.
- Have I distinguished clearly between what we did and what changed?
- Does every number say how many people it came from, not just how many took part?
- Have I given response rates, and said what I think the non-responders might mean?
- Can I point to the source behind each figure if someone asks tomorrow?
- Have I flagged anything I am not confident about, rather than smoothing over it?
- Have I avoided saying we caused something when we contributed to it?
- Does the spending picture match what we said we would do, with variations explained?
- Is there a specific plan for the evidence we did not manage to collect this time?
A printable one-pager. Use your browser's print or “save as PDF” option.
One further habit worth building: keep the working behind the report. If a figure came from 22 responses collected in June, note that next to the figure in your own records, not just in the submitted document. Questions arrive months later, usually when the person who did the counting has moved on.
Where to start if this feels like a lot
Do not rebuild your reporting in one go. Take the next report and change two things: add the denominator to every number, and add one honest paragraph about what you do not yet know. That alone puts you ahead of most of what lands on a funder’s desk.
For the underlying distinction, read outcomes vs outputs. To set out why you expect your work to produce a change in the first place, use the theory of change template.
ImpactProof keeps every figure in a report linked to the records it came from, so the follow-up question has an answer.
