by: Francisco Martinez | Principal and Co-Founder
Sovereign Good's thoughts on what works and what doesn't, and the entry points for both scenarios.
Every funder wants impact. Few want to do the unglamorous work of finding out whether they're getting it. In that gap lives a whole genre of activity that looks like impact, photographs like impact, and reports like impact, while changing very little about the problem it claims to address. The polite word for it is storytelling. A blunter word is impact-washing.
Impact-washing is not a program that tried something reasonable and didn't work. That's just philanthropy; most good ideas fail, and the failures teach you things. Impact-washing is the practice of presenting activity as if it were results: counting what's easy to count, framing what's flattering to frame, and quietly declining to ask the question that would settle the matter.
You can usually spot it by what gets measured. A workforce program reports that it trained four hundred people. Good. Did they get jobs? Did the jobs pay enough to live on? Were they still employed a year later, or did the program graduate them into the same dead end they started in? The training number is real, and it tells you almost nothing about impact. It's an output, a measure of effort. Impact is what changed in someone's life as a result, and it is harder to measure, which is exactly why so much reporting stops at the output and calls it a day.
What matters isn't whether numbers are present but whether the painful one is (in reality). Real measurement includes the possibility of bad news. When a funder or grantee only ever reports figures that flatter the work, you're not looking at evidence. You're looking at marketing wearing evidence's clothes.
This matters more in philanthropy than almost anywhere else, for a structural reason worth sitting with entirely. In business, a bad product eventually loses customers, and the market delivers the verdict whether you want it or not. In philanthropy, the people receiving the service rarely have that power. They didn't choose the program, they're not paying for it, and the funder, several steps removed, hears about results mostly through reports written by the people who need the funding to continue. Almost every incentive in the chain rewards a good story over an accurate one. Without someone deliberately insisting on the truth, the system drifts toward comfortable fiction, not because anyone is corrupt, but because nothing in the structure pushes back.
So the work of telling real impact from the performance of it falls to whoever is willing to do it. In our experience, that means a few specific habits. Ask what would count as failure before the money goes out, and write it down, so success can't be redefined after the fact to match whatever happened. Insist on at least one measure the grantee can't fully control; employment a year later, not attendance at the workshop. Talk to the people the program serves, directly, without the grantee in the room arranging the conversation. And treat a grantee who reports a real problem as more trustworthy than one who reports unbroken success, because unbroken success over several years is not a sign of excellence. It's a sign that no one is measuring honestly.
There's a temptation to read all of this as cynicism about the field. It isn't. The reason to care about the difference between impact and its performance is that real impact is possible, and settling for the performance is how you end up funding neither. Donors who insist on knowing the truth, even when the truth is inconvenient, are the ones taking the work most seriously. They want it to work, and you can't know whether it's working without being willing to find out that it isn't.
If you're a funder, the question to bring to your own portfolio is simple and slightly uncomfortable: for each thing you support, do you know what changed, or only what was done? If the honest answer is "what was done," that's not a failure. It's a starting point, and a better one than most. The next conversation is about what you'd need to know to give a different answer a year from now.