Project Wonderful Part 2

Assessing the effects of advertising on Project Wonderful. If you've been following the blog you might remember my odd results from last time. I've run the test for more times. Each time my page views stayed static, but the number of downloads? See for yourself:


You can click for a larger image. Yellow indicates the period an ad was live. The blue line shows the page the ad links directly to. The others show linked products.

Exactly the same number of people hit the page, but in each case they were statistically more likely to download. Also, following shortly after the main bump, there was a smaller rise in the number of people who downloaded associated products. The two middle increases are smaller. These two tests were run with button ads (117px x 30px), while the others used banners. I'd need to run a few more tests before I can see if there is a correlation between ad size and likelihood of downloads, but right now it looks like it.

This was with a free ebook on Smashwords, so the next test is to try this again with a commercial product.

The other thing I have done is to try it as a publisher. The Stephen and Matilda blog now has ads served by Project Wonderful. My assessment? It's easy to set up and editing your ad is simple. They also take Paypal. It takes a few days to get approved, unlike the signup for adsense, but then they are more selective.

A small side benefit I didn't expect is the metrics. Tracking how many people view an adbox, and where in the world they come from, also means tracking site visitors. One of its best features is that you can display your own ads in the adbox if no one else bids, making it easy to co-ordinate a campaign across multiple sites.

We've submitted a second blog to them and, if the results are promising, you might find them serving the ads on here.