Soon after starting GiveWell in 2007, we guessed that many charities didn't have the type of evidence we were interested in: High-quality, rigorous evidence demonstrating that the intervention they are implementing works. But we assumed that the most effective charities would.
We learned, however, that due to the expense and difficulty of collecting this kind of evidence, most charities—even our current top charities—don't have this kind of information.
Academics, however, often collect this type of evidence through their study of health and development interventions, which is why the first step in our process is now to review the independent evidence of a program's effectiveness before seeking charities implementing the program well.
More of our views on evidence: