Size and urgency of the problem
How many people or ecosystems are affected, and how severe is the harm?
The score is our own opinion, from 0 to 100, using seven published criteria. These scores are a first internal assessment, dated 16 November 2026. Each value still needs a second review and approval before any public use. Missing figures are left open; they are not estimated here. The list is reviewed regularly.
Bands: 80–100 Recommended (strong evidence or a well-reasoned expected value); 60–79 Promising (higher uncertainty); under 60 is not shown on You.
How many people or ecosystems are affected, and how severe is the harm?
How well is it shown that the intervention works, and how uncertain is the transfer into everyday use?
How much effect is to be expected per euro (expected value)?
Will the extra euro actually move something, or is the programme already fully funded?
Can the euro be followed, and what share reaches the programme or the people it is for?
Does the project visibly contribute to one of the five shifts?
Serious leadership, local partners, safeguarding and accountability?
Formula:
Score = 0.15·S + 0.25·E + 0.20·K + 0.10·A + 0.10·T + 0.10·P + 0.10·G
“How much of 100 arrives” is a transparency figure, not proof of impact. Scores across shifts are only roughly comparable: well-evaluated health and poverty programmes tend to score higher than policy, research or conservation work, where randomised trials are rare.
The Impact Score is our own assessment. It is not a certification, a seal or an audit, and it is not a guarantee of impact. Please read the full disclaimer before acting on anything you find here.
We draw on published data from the organisations themselves and from independent evaluators, including GiveWell, Giving Green, Founders Pledge, The Life You Can Save and Charity Navigator. Sources and dates are shown on each project card. We do not copy another evaluator’s ranking as our own, and Charity Navigator is used as a hygiene filter, not as a ranking of impact.
Every project card on You shows its sources and dates. Our weights and score bands are public — you can read the seven criteria on this page.
These scores are first assessments, dated 16 November 2026, and the list is reviewed regularly. We do not copy another evaluator’s ranking as our own. Charity Navigator is used as a hygiene filter, not as a ranking of impact.
If something looks off, write to us at [email protected]. Corrections are welcome.
We look for organisations led in or rooted in the countries they serve, across all five shifts. On You you will find, for example, CEDOVIP, a Ugandan organisation, led locally; Educate Girls, working in India; Lively Minds, with programmes in Ghana and Uganda; and CAMFED, supporting girls in rural Africa.
An honest limit: evaluator coverage is uneven, so strong local groups without formal evaluations may be missing from the list. If you know one, suggest it via Join.
Before you give, these questions are a good place to start — anyone can ask them:
Independent evaluators who publish on questions like these include GiveWell, Giving Green, Founders Pledge, The Life You Can Save and Charity Navigator.
Humanity 2050 receives no money from the listed organisations and collects no donations itself. People give directly on the organisations’ own sites.
Humanity 2050 has one founding partner: PranaSomatics, the brand of Pranasomatics SL, a company owned and run by Kai Hill, the publisher of this website. PranaSomatics offers paid training courses and supports Humanity 2050 by offering 2 to 3 free scholarship places per cohort. Because of this relationship, Kai Hill has a personal and commercial interest in PranaSomatics. We want you to know this openly:
No other organisation, sponsor or advertiser pays for or influences the content of this website. Read the full note in the disclaimer.
A score is orientation, not a verdict. It cannot capture everything that matters: dignity, culture, long-term or systemic effects, or high-risk research whose outcomes are still uncertain.
Some of the best work is hard to fit into a 0–100 number. Please read the cards, the sources and the disclaimer, not only the score.
See a project we should look at, or a score you think we have wrong? Suggest it or challenge it via Join, or write to [email protected].
You can also try your own weights on the 25 projects on You. This ranking is for exploring only. It is not a new official score, and it is not a recommendation to give or not to give.
Educate Girls works to get out-of-school girls into school and to improve learning, in India. The published Impact Score is 80.8 (Recommended), a first assessment dated 16 November 2026. Here is how the evidence led to each criterion, and from there to the total.
Millions of girls in India are out of school, and there is also a learning crisis among children already in school.
An IDinsight randomised trial of the first Development Impact Bond (2015–18) found, among other results, learning gains of +0.31 standard deviations, or 28% higher than the comparison group, 768 girls enrolled, and targets exceeded. A later scale-up was also evaluated by randomised trial (IDinsight report). The Life You Can Save recommended the organisation in 2026. IDinsight project page · The Life You Can Save.
The learning gains were independently measured, and The Life You Can Save recommends the organisation. We did not take over a GiveWell-style cost-effectiveness figure of our own.
Scale-up is under way, and the need is plausible.
Charity Navigator 93%, 4 of 4 Beacons. How much of 100 arrives: unknown — scores.md records no comparable arrival figure. The official donate page at educategirls.us was unreachable (503) on 16 November 2026; a working gift route via The Life You Can Save / Donorbox was checked.
This is exactly the intended focus: education for girls.
The Development Impact Bond included independent outcome measurement, plus The Life You Can Save and Charity Navigator Four-Star as further checks.
Score = 0.15·80 + 0.25·85 + 0.20·75 + 0.10·70 + 0.10·80 + 0.10·95 + 0.10·80 = 12 + 21.25 + 15 + 7 + 8 + 9.5 + 8 = 80.75, published as 80.8.