Social impact bonds (or “pay-for-success”) can be an incredibly useful tool for funding change.

They can be good for social enterprises and nonprofits in need of capital to scale, for the investors looking to fund those organizations and for governments looking for results.

As popular as these types of investments are getting in the United States, there’s still some serious confusion surrounding exactly how they work. So, we made a GIF.

A social impact bond is a way for government to, yes, “pay-for-success.” Here’s how the whole thing works, in five steps:

• A government entity picks a public sector issue like homelessness or recidivism, then chooses an intermediary partner to broker a deal.
• The intermediary, often a nonprofit advisory firm, is charged with raising funds from investors and finding a nonprofit/social enterprise that directly combats the public sector issue with a proven, scalable program or service.
• The investors — often large financial institutions such as Goldman Sachs or Merrill Lynch — fund the program or service.
• The program or service must deliver the desired results set in the initial contract.
• Once those results are evaluated and deemed successful, the government repays the investors.

Here’s the rub — they don’t always work, because of the subjective, sometimes murky nature of social impact metrics. What looks like sturdy, successful metrics to Goldman Sachs might seem fragile or miscalculated to parties evaluating those metrics. As measurement tools become more precise and efficient, it could very well become much easier to navigate public-private-partnerships like social impact bonds.

This type of partnership is still really green — the concept was introduced for the first time just over five years ago in London, and didn’t make its way across the pond until it was tested in New York City in early 2012 with Goldman Sachs and then later that year by New York state.