These are the gifts and campaigns where the result came out unmistakably good, and where we can show you why we think so. We judged them blind: not by the size of the cheque, the fame of the donor or the beauty of the story, but by what changed for people afterwards and how sure we can be. Each one has its evidence and the lesson we'd pass on.
Judged blind. We look at the result, not the label: what changed for people, how sure we can be, and whether it would have happened anyway. Size, fame and story don’t earn a place. There are no scores here, just the evidence and a verdict in words.
The entries
It worked
Smallpox eradication
World Health Organization, with funding, vaccine and staff from many countries · 1967–1980 · Global campaign
Tasting noteThe benchmark. A clear target, relentless counting, and a finish line anyone could check.
What they did
In 1967 WHO launched a ten-year Intensified Smallpox Eradication Programme aimed at the countries where the disease was still endemic. Teams used surveillance, case finding, contact tracing and ring vaccination, alongside public information campaigns.1
Why it counts
The last natural infection was recorded in Merka, Somalia, on 26 October 1977. In May 1980 the World Health Assembly declared smallpox eradicated, and it remains the only human disease ever wiped out.123
The evidence
Eradication is the rare outcome that can be verified directly: an independent commission searched, found no cases, and certified it. WHO estimates the programme cost about US$300 million in total, against savings to the world economy it puts at about US$1 billion a year. That ratio is WHO's own estimate, not ours.1
The lesson
Steer by the outcome itself. The campaign won by counting cases and chasing each one down, not by counting vaccinations. When you can see the result directly, nothing else should be your scoreboard.
It worked
The hookworm campaign in the American South
Rockefeller Sanitary Commission, funded by John D. Rockefeller · 1909–1914 · Foundation
Tasting noteMeasure first, treat second. A century later, that first survey let economists measure the payoff.
What they did
The Commission surveyed eleven Southern states, found that on average about 40 percent of school-aged children were infected with hookworm, then funded travelling dispensaries that gave treatment and taught local doctors and families how to prevent reinfection.45
Why it counts
Hookworm was quietly draining children's energy and schooling across a whole region. The campaign cut infection quickly and, as the next section shows, the benefits showed up in classrooms and, later, in pay packets.
The evidence
Economist Hoyt Bleakley compared areas by how badly infected they were before the campaign. Places with more hookworm saw bigger gains in school enrolment, attendance and literacy afterwards; adults, who had been infected far less, showed no such change; and children exposed to the campaign went on to earn more as adults.4
The lesson
Survey before you spend. Knowing the baseline is what makes a result knowable. Most charities still skip this step.
It worked
The Rosenwald schools
Julius Rosenwald · 1914–1931 · Individual donor
Tasting noteBricks and mortar that measurably changed lives, aimed exactly where the need was deepest.
What they did
The Rosenwald Rural Schools Initiative built nearly 5,000 school buildings for Black children in the rural South across the rural South.6
Why it counts
The schools served roughly 36 percent of rural Black school-aged children in the South, and they arrived as the Black-White gap in schooling began to narrow sharply.7
The evidence
Using census data and Second World War enlistment records, Daniel Aaronson and Bhashkar Mazumder found significant gains in attendance, literacy, years of schooling, cognitive test scores and later migration north. The gains were largest in the most disadvantaged counties.67
The lesson
Give where the gap is widest. The biggest effects came where opportunity had been lowest, which is exactly where a gift's marginal dollar goes furthest.
It worked
The March of Dimes and the polio vaccine
National Foundation for Infantile Paralysis (now March of Dimes), funded by millions of small gifts · 1938–1955 · Mass small-donor giving
Tasting noteDimes from millions of people, spent on science and then tested in public. Thrilling, and rigorous.
What they did
Founded in 1938, the Foundation raised money through a nationwide network of volunteer chapters, paid for patient care and funded virus research, including Jonas Salk's work on a vaccine.89
Why it counts
Polio once paralysed 13,000 to 20,000 people a year in the United States. The Foundation backed field studies involving about 1.8 million children, and on 12 April 1955 the vaccine was announced as "safe, effective, and potent". The last case of polio acquired from wild virus inside the United States was detected in 1979.9
The evidence
The vaccine was not taken on faith: it was tested in field studies on a scale that left little room for doubt before it was used widely.9 The Foundation also supported Albert Sabin's oral vaccine, licensed in 1961.9
The lesson
Small donors can fund big science, if someone insists on a proper test. And when you win, find the next fight: the Foundation moved on to birth defects in 1958 instead of fading away.8
It worked
The Green Revolution
Rockefeller Foundation's Mexican Agricultural Program, Norman Borlaug and many national partners · 1943 onward · Foundation
Tasting noteEnormous gains with real side effects. Still one of the best investments any donor ever made.
What they did
In 1943 the Rockefeller Foundation and the Mexican government started an agricultural research and training programme. Norman Borlaug, part of it, bred robust dwarf wheat; by 1956 Mexico was self-sufficient in wheat, and in the mid-1960s his varieties spread to India and Pakistan.1011
Why it counts
Borlaug received the 1970 Nobel Peace Prize "for having given a well-founded hope - the green revolution", and always said he was part of a team.11
The evidence
A 2021 study estimates that high-yielding varieties raised food-crop yields by 44 percent between 1965 and 2010, and that a ten-year delay would have cost 17 percent of developing-world GDP per person by 2010.12 The costs were real too: overuse of water, fertiliser and pesticide damaged soils and water in places, which a major review blames largely on subsidies that encouraged excess.13
The lesson
Fund the research nobody else will, then follow it into the field. And count the side effects honestly: a clear win can still leave a bill.
It worked
Oral rehydration, taught mother by mother
Researchers in Dhaka and elsewhere; BRAC's Oral Therapy Education Programme in Bangladesh · 1970s–1990 · Science, then delivery
Tasting noteSalt, sugar, water, and a house call. Cheap enough to reach everyone, simple enough to remember.
What they did
Scientists showed that a simple solution of salt and sugar lets the gut absorb water during diarrhoea. In 1980 BRAC began a ten-year effort to teach every mother in Bangladesh to make and use it at home, with a home-made recipe tested at the Cholera Research Laboratory in Dhaka.14
Why it counts
In 1978 The Lancet called the underlying discovery "potentially the most important medical advance this century".15 BRAC's workers visited about 12 million households, and over 90 percent of mothers knew of and could prepare the solution.14
The evidence
Strong on reach and knowledge, harder on deaths: a 2021 review of the programme notes that measuring mortality from diarrhoea is difficult, so evaluations leaned on knowledge and use. We'd flag that as a gap, not a flaw.14
The lesson
Put the treatment in the hands of the person who'll be there at 2 a.m. The best intervention is the one that actually gets used.
It worked
Gavi, the Vaccine Alliance
An alliance launched with a $750 million pledge from the Gates Foundation · 2000– · Pooled fund
Tasting noteNot glamorous, just enormous: vaccines bought together, delivered at scale, and paid for jointly with countries.
What they did
The Gates Foundation's 1999 pledge of $750 million helped launch a global alliance to speed up vaccine delivery to the poorest countries.16 Gavi works to shape vaccine markets and co-finances vaccines with the countries it supports.17
Why it counts
Gavi reports that more than 1.3 billion children were vaccinated through routine programmes it supported between 2000 and 2025, and estimates more than 23 million future deaths averted.17
The evidence
Those are Gavi's own modeled estimates, built on vaccine-effectiveness evidence; we show them as published, modeled, not guaranteed. A healthy sign: the number of manufacturers supplying the vaccines it buys rose from 5 in 2001 to 19 in 2025, and countries now pay a growing share themselves.17
The lesson
Fix the market, not just the moment. Pooling buyers changed prices and supply for everyone, which is a kind of impact no single donation can buy.
It worked
The Global Fund
Governments, foundations and companies, pooled · 2002– · Pooled fund
Tasting noteA pooled fund that publishes results every year and says plainly when progress is at risk.
What they did
Since 2002 the Global Fund has financed programmes against AIDS, tuberculosis and malaria; in 2024 alone it invested in health systems and disease surveillance in more than 100 countries.18
Why it counts
Its 2025 Results Report estimates 70 million lives saved since 2002, and a 63 percent fall in the combined death rate from the three diseases in the countries where it invests.18
The evidence
These are the Fund's own modeled figures, and the credit is shared with governments and other funders. We'd like to see more independent attribution. But the direction is not in doubt, and the Fund publishes enough for others to check.18
The lesson
Publish results every year, especially when they're inconvenient. A funder that tells you progress is at risk is one you can trust when it says things went well.
It worked
Against Malaria Foundation
A small charity that buys and distributes insecticide-treated bed nets · Charity we track
Tasting notePlain, proven, relentlessly checked. The quiet classic of modern giving.
What they did
AMF funds long-lasting insecticide-treated nets for households where malaria is common, and is one of GiveWell's Top Charities.19 Its current published figures are on our AMF page.
Why it counts
Nets are one of the best-tested tools in global health. A Cochrane review of randomised trials found insecticide-treated nets reduced deaths from all causes among children by 17 percent compared with no nets.20
The evidence
Trial evidence for the tool, plus GiveWell's detailed public analysis of the charity. Any cost-per-life figure is GiveWell's estimate: modeled, not guaranteed.1920
The lesson
Boring is a compliment. When the tool is proven, the remaining questions are delivery and room for more money, and those can be checked.
It worked
GiveDirectly
A charity that sends cash directly to people living in poverty · Charity we track
Tasting noteThe benchmark every other programme should have to beat. Respectful, measurable, and studied to the bone.
What they did
GiveDirectly makes large, unconditional cash transfers by mobile money, and its transfers have been studied in several randomised trials. Its profile is on our GiveDirectly page.21
Why it counts
In a randomised trial in Kenya, households that received transfers spent more on consumption, gained assets and reported large improvements in psychological wellbeing.22 A larger study found big gains for recipients, spillovers to neighbours and local firms, and minimal inflation, with a local multiplier of about 2.5.23
The evidence
Some of the strongest evidence in development: multiple randomised trials, published in top journals, including effects on people who didn't receive cash.2223
The lesson
Use cash as the yardstick. If a programme can't show it beats simply giving people the money, ask why not.
It worked
The Gates Foundation
Bill Gates and Melinda French Gates · 2000–2045 · Foundation
Tasting noteThe largest bet on measurable health in history, with power that deserves the scrutiny it gets.
What they did
Founded in 2000, the foundation helped launch Gavi and the Institute for Health Metrics and Evaluation.2425 In 2025 it announced it would spend more than $200 billion over twenty years and close at the end of 2045.26
Why it counts
In 2009 The Lancet credited it with transforming the funding, science and politics of global health, while criticising how its grants tracked disease burden and calling for better governance and transparency.25
The evidence
Its greatest successes run through institutions it helped build, such as Gavi, whose evidence we summarise above. Its influence is large enough that independent scrutiny of it is itself a public good.25
The lesson
Build institutions others can join, measure everything, and set an end date. Then invite the criticism, because at that scale you'll need it.
It worked
Chuck Feeney and The Atlantic Philanthropies
Chuck Feeney · 1982–2020 · Giving while living
Tasting noteHere for the principle as much as any single result: give it away while you can still see it work.
What they did
Feeney put almost all his fortune into a foundation he set up in secret, gave anonymously for years, and made more than 6,500 grants worth over $8 billion in more than 25 countries. Atlantic made its last grant commitments in 2016 and closed in 2020, as planned.27282930
Why it counts
Atlantic describes itself as the largest foundation ever to commit its whole endowment within a fixed period and then close, showing that "giving while living" works at billion-dollar scale.3129
The evidence
We can't vouch for every one of 6,500 grants, and we don't claim to. We include Atlantic for its model: anonymity, urgency and a deadline. Feeney's own advice to the very rich was "try it, you'll like it".29
The lesson
Money in an endowment helps no one yet. Give while you can still watch, learn and adjust.
Honourable mentions
Worth knowing, with a caveat.
One is a government programme, not private giving. The other is a classic we admire, judged honestly against what the evidence can and can’t show.
It worked
PEPFAR
The United States government · 2003– · Government programme (honourable comparison)
Tasting noteNot philanthropy, but a humbling reminder of what public money can do at scale.
What they did
The President's Emergency Plan for AIDS Relief has invested over $100 billion in the global HIV response since 2003.32
Why it counts
The State Department says it has saved 25 million lives. That is the government's own estimate.32
The evidence
We include it as a yardstick, not a donor: no private gift is on this scale. Its future funding is a matter for governments, not this page.
The lesson
Private donors are usually the venture capital of doing good. The best private gifts prove something that public budgets can then scale.
It worked
Carnegie's public libraries
Andrew Carnegie · 1883–1919 · Individual donor (judged honestly)
Tasting noteGrander in legend than in the evidence, but the evidence that exists is good.
What they did
Carnegie, who wrote in 1889 that "the man who dies thus rich dies disgraced", helped build more than 2,500 libraries around the world.3334
Why it counts
A free library was, in his words, the "best gift which can be given to a community".34
The evidence
One careful study compares US towns that got a Carnegie library with towns that applied but never built one, and finds patenting rose by about 10 to 12 percent over the next twenty years. That's real, and narrower than the legend: it measures invention, not reading, learning or wellbeing.35
The lesson
Love the story, then ask what it changed. Even our favourite legends deserve a blind tasting.
Sources
Every fact and figure above links here. We checked each source on 24 October 2026. Figures are as published by each source; estimates are theirs, not ours.