It's clear that by giving to GiveWell's top charities you can do a lot of good. I wouldn't know about those charities, though, without GiveWell working to find them. Perhaps I could do more good by giving to GiveWell than to their top charities, by allowing them to put more research into identifying even better ones?

GiveWell is an example of a "meta-charity", one that evaluates charities. There are also meta-charities that promote giving more and giving better, such as 80,000 Hours, Bolder Giving, and Giving What We Can. They could all use more money to hire more professional giving advocates. [1] If they can turn a dollar of donations into substantially more than a dollar of increased donations to effective charities, isn't that the best use of my money?

GiveWell isn't currently limited by donations [2] and Bolder Giving doesn't emphasize giving effectively, so I'm really just looking at 80,000 Hours and Giving What We Can. How do I know whether a dollar spent on advocacy through one of them goes farther than a dollar given directly to the charities they advocate? What is the additional amount of donations they can move per dollar I give them?

The simplest thing is to survey the people who you've influenced to find out how much they're giving to effective charities that they wouldn't be otherwise. Add to this a discounted estimate of what they will continue to give accounting for attrition. Divide by the money you've spent, converting volunteer time to money at whatever your full cost per employee would have been. [3] Is this good enough? What should I watch out for?

One thing that does worry me about giving to them is that I might displace other donations. Let's say it turns out they can make a strong argument that they're moving 10x in additional donations for each dollar they get, but that you can only grow so fast and so they have much less room for more funding than the amount they could bring in. There may be people who will be moved by this 10x boost to give more than they would otherwise. I'm not one of them: I set aside money to donate and being confronted with a particularly effective giving oppertunity isn't going to get me to give more. So it's better if these flexible people do the metacharity supporting.

[1] I do some as a part-time giving advocate, blogging, maintaining the Smart Giving subreddit, and hosting effective giving meetups, the next one of which is the evening of Friday July 6th.

[2] When Julia wrote to them about a month ago they said they we're fine for the time being.

Update 2014-12-22: As of December 2014 GiveWell now does think they could use additional funding effectively.

[3] Giving What We Can tells me they've done this, and I'm trying to get data from them to see if I can trust their methodology.