Direct marketers and the customers who buy from them (or contribute to them) will find fascinating this transparent look at the A/B testing of banner advertising by the Wikimedia Foundation, the non-profit organization that brings you Wikipedia. The A/B testing results page is part of a detailed wiki-style documentation of the entire fundraising effort, including a PDF of the results of the research project that guided the development of the banner options tested.
I especially like seeing this inside look at the process that went into developing these banner ads because, like I’m guessing a lot of other people do, when I see those photos of Jimmy Wales, I understand why Nat Torkington “eloquently” described them as, “Jimmy Wales’s hollow haunted gaze staring at you with the eerie intensity of a creepy hobo talking about how tasty human liver is.” (Sorry, Jimbo.) On the right, you can see a visualization (infographic) created by David McCandless, of the A/B test results that (with a bit of wry British wit) shows how the “gore” approach trounced the other options tested.
Too often, we (and by “we,” I mean “me” and every client I’ve ever had) believe our personal tastes and opinions are shared universally. We rationalize it this way: If we like an ad or page layout or advertising message, then certainly the customer (or donor) will also. This ego-centricity leads us to seek-out others (typically, those we know will agree with us) to join in a dance called the confirmation bias tango.
However, as the ever-quotable Finnish communications theorist Osmo Wiio (my go-to guy on this topic) once said, “Being content with the formulation of your message is a sure sign of having formulated it for yourself.”
It’s not very often we get such an inside look at the process of finding out what works — and why it’s very often not what we think will work.
Oh, and they worked on me. I’m upping my annual contribution. You should also.