Its become fashionable to run AB and multivariate tests for just about anything on the internet. Of course, direct marketers have been practicing tests vs. control groups for a long time before the rise of the world wide web. It would be a shame to spend so much time testing and yet not know when you’ve found something meaningful. Knowing when and how to check for statistical significant belongs in every marketer’s toolbox (or at least, you better get know a friendly web analyst who will run the numbers for you). Here are two of my all-time favorite posts on this subject:

RKG on finding statistical significance in two Adwords tests Interesting commentary on the value (or lack-of-value) in copy testing in PPC
Avinash Kaushik on separating signal from noise with statistical significance

What I am reading

URL’s are totally out In Japan, someone has taken the time to study how people really use search. Instead of listing URL addresses, innovative advertisers have started showing images of a search box with recommended search terms. A commenter mentions that this idea is reminiscent of AOL Keywords back in the 90s. Regardless of the origins, the idea is brilliant, assuming you’ve done the SEO footwork to ensure you rank first for your brand keywords.