Impact reporting a funder will actually read

Most impact reports are written in the wrong direction. They start from everything the organisation did, and work outward until the document is long enough to feel serious. The person receiving it reads the first page and the numbers, and forms a view.

A report built the other way round, from the three things a funder actually needs to decide, is shorter to write, easier to defend and considerably more likely to be read.

What is a funder deciding?

Whether the money moved anything, whether you know how you know, and whether you would notice if it had not.

The third one carries more weight than people expect. A funder reading a report of unbroken success is not reassured by it. They have read many of those, and what the good ones have in common is at least one place where the organisation says a thing did not work and explains what it did about it.

What should the report contain?

Four things, and they fit on three pages.

What you set out to change, stated as it was stated at the start rather than rewritten to match what happened. What you did, in quantities: how many, where, over what period. What changed, separated clearly into what you measured and what you believe. And what you would do differently, written plainly enough that it is obviously not a formality.

1. 01

   The aim

   What you set out to change, as stated at the start

2. 02

   The activity

   What you did, in quantities and dates

3. 03

   The result

   What changed, measured and believed kept apart

4. 04

   The learning

   What you would do differently, meant seriously

Three pages in this order beats thirty in any other. The section most organisations leave out is the last one, and it is the section that makes the rest of the document credible.

Where does the data come from?

Almost always from something you already collect and have never assembled.

Attendance registers. Enquiry emails. The sign-up form on your website. WhatsApp groups. A spreadsheet somebody keeps because they find it useful. Most organisations I have worked with are sitting on enough operational record-keeping to build a defensible report. What is missing is that nobody has ever put the pieces next to each other.

That is worth knowing before anyone sells you a monitoring system. The first version of impact reporting is a person spending two days with the records you already have.

What separates measured from believed?

A measured claim has a source you can point at and a method somebody else could repeat. A believed claim is a reasonable inference from what you have seen.

Both belong in a report. What damages an organisation is presenting the second as the first, because a funder who catches one inflated claim reasonably discounts every other number in the document.

So write them separately and say which is which. "We recorded 412 attendances across 14 sessions" is measured. "Facilitators consistently reported that the older participants set the tone" is believed, and it is genuinely useful, and it costs you nothing to label it honestly.

What about the numbers you cannot get?

Say so, briefly, in the report.

Long-term outcomes are the usual case. An organisation running a skills programme rarely knows what happened to a participant three years later, because tracking people over years is expensive and most grants do not fund it. Writing "we do not currently track outcomes beyond twelve months, and here is what we would need in order to" is a stronger sentence than an estimate dressed up as a finding.

It also does something useful: it tells a funder exactly what to fund next.

How often should this happen?

Less often than most organisations think, and more regularly than most manage.

One properly built annual report, plus a short quarterly note of a page, is enough for almost every funder relationship. What actually goes wrong is that the annual report gets written from scratch in a panic each year by whoever is least busy, which is why it takes three weeks and reads like it.

Building it from a record you keep as you go turns three weeks into two days.

What does good look like from the outside?

A funder should be able to read your report and describe your programme accurately to a colleague, from memory, five minutes later. That is the whole test, and most reports fail it because they are optimised for completeness rather than for being retained.

If you would like the reporting built from data you already collect rather than a new system to feed, that is the work.