Can we say we caused this?
- TimeNot established by the research. A rewrite of findings you already have.
- CostFree — a rewrite of findings already produced.
- Doing itYou can do this yourself
Almost never — say you contributed to it, which is both more defensible and what the sector's own good-practice guidance actually recommends. Attribution and contribution are different questions: proving attribution means knowing what would have happened without you, which a small charity almost never can. Other people and other services nearly always played a part. Naming them makes your claim more believable, not less.
Do this first
England and Wales: This page sets out the position in full. What you may honestly claim is a matter of good practice from named bodies, not of law. Scotland: the same position, with no difference in principle. Northern Ireland: the same position, with no difference in principle.
NPC’s Code of Good Impact Practice on contribution
On attribution and contribution, the sector’s own good-practice code is clear: claim contribution, not cause. Overclaiming is described as the most common failure in charity impact reports, and the most damaging to a charity’s relationship with its funders. The fix fits in one sentence. This step works on the summary you wrote when you analysed what you gathered.
It comes from the Code of Good Impact Practice, which NPC now hosts. Principle 6, “Be honest and open”, says, verbatim:
“Be honest when considering the contributions made by others to the actual impact achieved; don’t over- or under-claim. It’s more realistic to talk about the contribution you’ve made rather than to try to attribute impact to your work alone.”
That is NPC’s recommendation, and this guide follows it. It is good practice from a named body, not law, and it binds nobody. But if other people plausibly helped bring about a change, your report should say “we contributed to” rather than “we caused”. Other people nearly always helped.
What attribution and contribution mean
Attribution and contribution are different questions, and naming them is most of solving the problem.
Attribution is the attempt to prove that your programme directly caused a result, with every other cause stripped out. To do that properly you need to know what would have happened without you. A small charity almost never can.
If a funder’s evaluation plan asked whether a change is attributable to your project, this is where you answer it. The plan itself is set when you write the monitoring and evaluation section.
Contribution is the gentler claim: that your work made a plausible, credible contribution alongside other causes. It comes from John Mayne’s contribution analysis, in which outcomes come from a “causal package” of several factors working together. Mayne puts it this way, as quoted in the TASO and INTRAC summaries: “Contribution analysis explores attribution through assessing the contribution a programme is making to observed results.”
If attribution feels impossible, you are in good company. NPC calls it, verbatim, “the most difficult and contentious aspect of evaluation”. It is not a skill everyone else has mastered and you have not.
A note on the sources. Mayne’s own papers were read for this guide only through summaries by TASO and INTRAC. The NPC Code was read directly, and it carries the practical point on its own.
Four ways a claim inflates
The social return on investment method, SROI, names four ways a claim grows bigger than the truth. You do not need to run an SROI study to use them. They work as four plain questions about any result.
| The term | What it means | The question to ask |
|---|---|---|
| Deadweight | The outcome would have happened anyway. | Would some of these people have got there without us? |
| Displacement | The outcome moved from somewhere else rather than being new. | Did someone else lose what our people gained? |
| Attribution | Part of the outcome belongs to other organisations or people. | Who else helped — family, other services, an employer? |
| Drop-off | A lasting outcome fades, and less of it is yours each year. | How much of this will still be true next year? |
Most charity reports never ask these questions out loud. A sentence on each puts your report on firmer ground than a page of confident verbs.
The same care applies to the words themselves: outputs, outcomes and impact are different things.
Separate outputs from outcomes from impact sets out all four words in NPC’s definitions. Contribution language matters most at the outcome and impact end, because that is where other causes crowd in.
Social value ratios and the league-table mistake
The best-known form of this mistake is the social value ratio: so many pounds of social value for every pound spent. It invites comparison. Charity A’s ratio is higher than charity B’s, so charity A must be better.
The body that sets the standard for social return on investment says otherwise. Social Value UK’s published position, as consistently reported, is:
“It is not recommended that you use the ratio to compare different organisations, even if they work in the same sector — such as children’s services — because they will be working with different users with different needs and may have made different judgements in calculating their ratio.”
So you should not compare your ratio with anyone else’s, and you should not let anyone else compare theirs with yours. Two ratios only line up if the studies used the same scope, the same stakeholders and the same adjustments, which they rarely do.
Be clear about how firm that quotation is. The article it comes from no longer loads at its published address, because Social Value UK’s old website now redirects elsewhere. The wording appears the same way in several independent sources, but this guide could not read it on a live page from the body itself.
What was read directly is Social Value International’s list of principles. Principle 5 says, in as many words, “Do not overclaim.” That principle alone carries most of the argument.
If you do quote a ratio of your own, state its scope beside it and let it stand alone. The step on choosing a framework covers whether SROI is worth doing at all.
How to rewrite the findings
Take the draft sentences that make claims and put each one through the same five steps:
- Find the verb. “Got them into work”, “reduced isolation”, “transformed”. Each one is a causal claim.
- Ask the four questions. Deadweight, displacement, attribution, drop-off. One sentence each is enough.
- Name the others. Say in the report who else was involved.
- Change the verb. “Contributed to”, “helped”, “was one of the things that”. Keep the number, change the claim.
- Check the words. Decide whether the claim is an output, an outcome or impact, and use the word that fits.
The result reads more modestly. It also survives a sceptical reader, which the original did not. With the claims rewritten, the next step is to write your impact report around them.
Common mistake: claiming “we caused this”
Overclaiming is described as the most common failure in charity impact reports and the most damaging to funder relationships. Attribution is, in NPC’s words, “the most difficult and contentious aspect of evaluation”.
NPC’s Code of Good Impact Practice, Principle 6, gives the fix. “It’s more realistic to talk about the contribution you’ve made rather than to try to attribute impact to your work alone.” That is good practice from a named body, not a legal requirement.
Common mistake: not knowing the four ways a claim inflates
The vocabulary lives in the SROI guide, which most small charities never open. Deadweight is what would have happened anyway. Displacement is change shifted from elsewhere. Attribution is the share that belongs to someone else. Drop-off is the fading of an outcome over time. Naming those four gets you most of the way.
Common mistake: comparing your SROI ratio with another charity’s
“So many pounds of social value for every pound” invites a league table. By the standard-setter’s own reasoning the ratio is not comparable between organisations: different users, different needs, different judgements. The source article no longer loads at its published address, so treat the quotation as Social Value UK’s position as consistently reported. Principle 5 of the Social Value Principles, “Do not overclaim”, was read directly.
Worked example
Saltmarsh Trust runs an employment project. Its clients found work, and the draft report says the project “got them into jobs”.
The project lead runs the four questions:
- Deadweight. Some clients would probably have found work anyway, because the local jobs market was improving.
- Displacement. The jobs did not obviously come from somewhere else.
- Attribution. Others helped — the Jobcentre, families, and in two cases the employer’s own training scheme.
- Drop-off. The project does not yet know how long the jobs will last, and says so.
The sentence becomes: “The project contributed to clients finding work, alongside support from the Jobcentre, their families and employers.” The four adjustments get a sentence each underneath.
The draft also quoted a social value ratio next to a figure from another charity’s report. The comparison comes out. Saltmarsh’s own ratio stays, with its scope stated, and stands alone.
The outcome: findings rewritten in contribution language, with the other contributors named. This is good practice from named bodies — NPC, Social Value International, and Mayne through the summaries of his work — not a legal requirement.
What you should have at the end
Your findings rewritten in contribution language, with other contributors acknowledged.
The rewritten findings show that you asked the obvious sceptical questions before anyone else did. They are the raw material when you write your impact report.