Measuring programme impact is not counting what you did

Aug 29
3 min read

Updated: 1 day ago

Introduction

Asked about impact, most small organisations report activity: sessions delivered, people seen, meals served, calls answered. These are real achievements and they answer a different question from the one being asked.

Impact is what changed as a result. Funders increasingly distinguish between the two, and more importantly the organisation itself cannot tell whether its work is doing any good without making that distinction. Measuring it properly is more achievable for a small charity than the language around it suggests.

1. Measuring programme impact means separating outputs from outcomes

Get the vocabulary straight and the rest follows.

Outputs are what you delivered: forty sessions, two hundred visits. Outcomes are what altered for the people involved: they found housing, their confidence improved, they stopped needing the service. Report both and label them correctly.

2. Decide what change you are aiming for before you measure

You cannot measure a goal you have not stated.

Write down what should be different for the people you work with as a result of the programme. Vague aims produce vague measurement, and the exercise of writing it down often reveals that the team held different assumptions.

3. Choose a small number of indicators

Ambition here defeats itself.

Two or three things you can genuinely collect, consistently, without adding an hour to every session. A framework with fifteen indicators produces incomplete data on all of them and reliable data on none, and it quietly teaches staff that the measurement is a formality.

4. Collect a baseline

Without a starting point there is no change to report.

Ask the questions at the beginning as well as at the end, even briefly. Organisations frequently reach the end of a project and discover that they can describe where people are now but not where they started.

5. Use methods that fit the setting

Formal instruments are not always appropriate.

A short scale, a simple before-and-after question, structured staff observation, or a conversation recorded consistently. The best method is the one that people will actually complete in your context without it distorting the service.

6. Combine numbers with accounts

Neither is sufficient alone.

Figures show scale and pattern; individual accounts show what the change looked like in one life. A report carrying both is credible where numbers alone feel thin and stories alone feel selective, and it is also more useful internally when you are deciding what to change.

7. Be careful about claiming causation

The honest boundary.

If somebody's situation improved while attending your programme, you contributed; you may not have caused it alone. Saying that plainly is more credible than an attribution claim your data cannot support, and funders notice the difference.

8. Report what did not work

The strongest signal of a serious organisation.

Programmes that underperformed, groups you failed to reach, an approach you abandoned. This is uncomfortable and it makes everything else you report believable, which is worth more than the discomfort.

9. Use the findings internally, not only in reports

Measurement that only faces outward is wasted.

Look at your own data and change something: a session time, an eligibility rule, a referral route. Organisations that visibly act on their own measurement collect better data, because staff can see the point of gathering it.

Protect the data you collect about the people you serve. Consent, anonymisation where possible, secure storage and retention limits apply to outcome data exactly as they do to donor records, and often more sensitively.

Conclusion

Measure what changed rather than what you delivered, because activity counts answer a question nobody was asking.

State the intended change before choosing indicators, keep to two or three you can genuinely sustain, collect a baseline so there is something to compare against, choose methods that fit your setting rather than borrowed formal instruments, combine figures with individual accounts, describe your contribution honestly instead of claiming sole causation, include what underperformed, act on the findings internally so staff see the purpose, and handle outcome data with the same care as any other personal information.