Five problems.
Ten organisations.

Global plastic production runs at over 400 million tonnes a year and is still climbing. An estimated 8 to 11 million tonnes of it reaches the ocean annually — a figure that sounds abstract until you stand on a remote beach that nobody visits and find it lined with bottle caps.

Most of it doesn't arrive by ship. It arrives by river. A relatively small number of rivers, running through dense population centres without waste infrastructure, carry the large majority of plastic that reaches the sea. That matters enormously, because it means the problem has chokepoints — specific, findable places where a barrier or a collection crew intercepts waste before it disperses into two thirds of the planet.

Once plastic is out at sea it fragments. Sunlight and salt break bottles into flakes, flakes into microplastics, and microplastics into something you cannot filter out of anything. It's in fish, in salt, in rain, in us. The Great Pacific Garbage Patch is not an island of bottles you could photograph — it's a soup roughly three times the size of France, mostly invisible from a boat deck.

There is no technology that un-fragments a microplastic. Which is why almost everything worth funding happens upstream of the damage: stop it at the river, or collect it while the pieces are still big enough to catch.

The tropics lose in the order of 3 to 4 million hectares of primary rainforest every year. Reported as a rate, that's a football pitch of irreplaceable forest gone every few seconds, continuously, while you read this page.

Rainforests hold roughly half the world's known species on about six percent of its land surface. They also run the water cycle for entire continents — the Amazon generates much of its own rainfall by recycling moisture through its own trees. Cut enough of it and the system stops raining on itself. Scientists studying that feedback loop put the tipping point somewhere around 20 to 25 percent deforestation, and the Amazon is already close to the bottom of that range.

The bleak arithmetic behind conservation funding is this: protecting standing forest costs a fraction of restoring cleared land, and unlike restoration, it works immediately. A protected hectare keeps its carbon, its species and its hydrology on day one. A replanted hectare is a promissory note against a century of successful management.

Land that is legally titled and locally stewarded — particularly by indigenous communities already living in it — is the most durable protection anyone has measured. Not fences. Deeds.

Start on the street. The WHO estimates more than 200 million stray dogs worldwide, and welfare organisations put stray cats higher still — animals that are born, get sick, get injured and die without ever having anyone responsible for them. Almost nobody funds their medical care, because they belong to no one.

Then the farms: tens of billions of land animals are raised each year, the overwhelming majority in intensive confinement systems designed around throughput rather than the animal. Add an estimated 115 million animals used in laboratory experiments annually — a figure that is itself uncertain because much of the world does not require anyone to count — plus wildlife trafficked at industrial scale, working animals driven past exhaustion, and pets abandoned when they become inconvenient. The common thread is that the suffering is legal, routine and invisible.

What makes this cause unusual is how little it costs to move. Very little philanthropic money goes to animal welfare relative to the number of animals affected, which means the marginal donation still buys extraordinary leverage — a street hospital in one country, a corporate policy commitment in another, testing methods that don't require an animal at all.

A single successful corporate campaign can improve conditions for more animals than a lifetime of individual rescue — and a street hospital saves the specific animal in front of it. We fund both, because both matter, and we do not pretend they are the same size.

Around 2.2 billion people lack safely managed drinking water, and roughly one in ten has no basic access at all. The consequences are not exotic: waterborne disease remains one of the leading causes of death in children under five, and it is caused by something a filter and a borehole fix permanently.

The second cost is time. Where there is no nearby water source, collecting it is unpaid labour, and it falls overwhelmingly on women and girls — collectively hundreds of millions of hours a day walking to a source and back. Those are hours not spent in a classroom or earning an income. Install a well within walking distance and you have not only prevented disease, you have handed a village back its own working day.

This is the cause we fund with the least agonising, because it is the rare global problem where the only missing ingredient is money. No breakthrough required, no political consensus, no behaviour change. Roughly forty dollars puts clean water within reach of one person for the rest of their life. A t-shirt company can generate that.

Around 250 million children and young people are out of school. But the deeper crisis is the children who are in school and still can't read: in low-income countries, the large majority of ten-year-olds cannot read and understand a simple story. A classroom without a teacher, a book or a language the child speaks is attendance, not education.

The returns are unusually well documented. Each additional year of schooling measurably raises lifetime earnings, and the effect carries into the next generation — educated girls in particular go on to have healthier, better-educated children, which is why girls' education keeps appearing at the top of development-economics rankings.

Conflict and displacement are now the largest single driver of children losing school entirely. Which means the education budget and the humanitarian budget are not separate problems: keeping a child in school during a crisis is often the intervention that prevents a decade of lost ground.