Sean emailed me the following (posting it with his permission):

When I thought about it I imagined it working like this:
-webmaster visits site
-browses by cause category (eg third world / disease research / education)
-gets banner image plus link to download
-puts banner on site, linking directly to the charity.

I saw the biggest barrier being to get the charities to commit once the architecture was sorted. That's because it would require a lot of approaches to hard-pressed charities to start with, convincing them the idea was worth their time.

The sponsorship might make it easier to allocate time to the project, but I don't think it's necessary really for the resources given that domain names are so cheap and webspace is free. In fact, I could sort the hosting probably.

The main benefit of sponsorship would be bringing in traffic, potentially.

I had a couple of friends interested in helping me when I was looking at it. I think the first stage would be to get a dummy site together that explains the concept to charities and websites so they can sign up to participate when it launches and then try to get some banners from charities.

Once the dummy site is up, it's possible for someone to start promoting it through search engine registrations, someone else to start approaching charities, someone to start designing banner adverts for charities who don't have the resources to do it themselves.

I did consult a friend who worked for a major UK charity when I was researching this idea and she said that some charities are vary wary of people who try to help. I had already concluded that the only way for it to have credibility is if it's run as a non-profit venture, which makes it difficult to accept cash sponsorship I think although it might be good to get a company to give free webspace or something.