Our vision
A world where no one has to wonder what their money made possible.
Our method
One set of questions, asked of any organization.
We assess organizations whether or not we fund them, and what we publish is not limited to the ones we do.
The eight questions are built on the framework of the Impact Management Project 1, the sector’s five-year effort to agree what a complete impact claim has to answer.
Every assessment
Eight questions, asked before and after.
Eight questions, asked in the same order of every organization, and asked twice: once before the work, to say what we expect, and again after, to record what happened. Asking twice is what makes an answer checkable, because the first is a claim and the second is the result set against it.
Together they build one picture: what changes and for whom, how much of it, whether it would have happened anyway, how likely we are to be wrong, what it cost beyond money, and whether we said so in advance. The same discipline holds whether the work happens in your neighbourhood or across an ocean, and whether or not our money is in it.
Before the work
What do we expect this work to change, and is it a change a person would recognize as their own life getting better? We use one definition of “better” throughout, and it is not ours: human flourishing.
After it
What changed, in the same unit we predicted it in.
Before the work
Who will the work reach, and how far from help were they before it arrived?
After it
Who it reached, from the organization’s own records.
Before the work
How much change, how deep for each person, and for how long after we stop paying.
After it
How much, how deep, and whether it held once the grant ended.
Before the work
What is the specific alternative, including whether another funder would have stepped in had we not?
After it
Whether that alternative materialised, and what it does to the figure we published.
Before the work
How strong is the evidence behind each link in the chain, and how sure are we of the whole?
After it
Where we were wrong, and by how much.
Before the work
What could go wrong for the people this is meant to serve?
After it
What did go wrong. Harm is recorded on its own and never subtracted from benefit, because netting lets a large benefit absorb a serious harm without anyone seeing it. Where we do not know, the record says unknown, not none.
Before the work
Will we be able to verify any of this, and by what route?
After it
Whether we could, and what the checking found.
Before the work
The prediction is written and published before the work starts, and before a dollar moves where we are funding it.
After it
It is published against at the close of the edition, misses included.
Our research
How we measure, in plain terms.
For every cause area we study, we build the measures around the work itself.
A food bank’s model follows the dollar from meals served through to families food-secure.
The chain in brief. A full model values every link, each anchored to published research and graded by the strength of its evidence.
See one model opened up: inside our malaria net analysis →
One number
Human flourishing per dollar.
Everything the Good Rises Fund grants to is measured in one unit, so that grants in different fields can be compared at all.
The unit is built to add up: a grant, an edition, and in time everything you have given across human, environmental, and animal work, in one figure. Nothing in the sector shares a unit that lets giving be added together, which is why this one is worth the difficulty it causes.
And a common unit is one half of a decision. It tells you how much good a dollar buys; it cannot tell you which good you want. That half is yours, and the unit works inside whatever you choose.
Every organization doing good work moves some part of the world toward better. But “better” needs a definition sturdy enough to measure against, and we did not invent one. We use the definition developed at Harvard’s Human Flourishing Program 2: human flourishing is a state in which all aspects of a person’s life are good. It is among the most rigorously studied definitions of a good life in the literature, tracked across more than 200,000 people in 22 countries.
Five domains, sustained by two conditions. The five domains and the first condition are Harvard’s. The second we added ourselves.
Why the second condition exists. A stable climate, clean water, and the living world around us are the ground beneath health and material stability, and that dependence is what our models can price today. Caring for them is also part of what a good life consists of rather than a means to one, which is stewardship. That is declared and not yet built, so no grant is credited for it, and the shortfall is ours rather than the work’s.
Where a benefit will not honestly fit human-centred numbers, we do not force it. We do not claim to measure the worth of a river, a forest, or an animal in itself. The frame is human flourishing, deliberately so. Where the model stops, we say so beside the figure.
Every model we build has the same destination, which is why we can use the phrase literally: flourishing is our unit of account.
Built with AI, checked by people
Experts, working with always-on AI, assess organizations on what they actually achieve – and where the evidence is thin, we say so. AI is how analysis of this depth gets built and kept current across this many cause areas at a cost we can sustain. It is never the last word: economists and subject-matter experts review what the machines produce, every model is anchored to published evidence, and every assumption is stated.
We publish so that people can give with confidence and for-impact organizations can show what they know is true.
We do not do this alone. Our models stand on published academic research. Our infrastructure comes through technology partners, used at fair market value. And we work with corporations and funders who share the conviction that good deserves evidence.
Want new findings as we publish them?
Sources
1. The eight questions are built on the framework of the Impact Management Project, whose norms are now held by Impact Frontiers. We use it as a completeness test. We are not accredited by it, affiliated with it, or endorsed by it, and it makes no claim about our work.
2. The five domains of flourishing are from Tyler J. VanderWeele’s Flourish Index, Human Flourishing Program, Harvard University. The five domains and the first sustaining condition are his. The second sustaining condition, the living world, is ours, and is marked as such wherever it appears.
Questions about our numbers: we answer them plainly →