Dr. Gardners presentation focused on the use of A/B testing for websites. The key point of A/B testing is to decide if we put our efforts and $ on A or B. This approach is organic and Darwinistic there should always be an A and a B, even if A is Excellent, it should be challenged constantly by a B. Markets change and the only way to be ready is to challenge A with a B. One benefit of A/B Testing is it is forward looking research where Analytics is backward looking and historical.
Your testing must be relevant. If you are doing customer satisfaction surveys that don’t drive change in the company why bother doing them? It must be part of driving a strategic initiative within your company. Ask who your website customer is and what is it you want them to do: a conversion to a sale, a phone call in, download a paper, watch a video, what is it you want the visitor to do? How are your customer segments the same vs. different? What is the dependant variable? Then using two choices be it 2 different web pages, questions, selections it can be anything, track their choices. Using the results keep tweaking until you are satisfied with the results.
We applied the Kano model to website performance, a comparison of the expected, un-expected that delights, the wow factor, just meeting expectation, and not meeting expectations. You don’t get points for doing the expected and meeting basic expectations. Dr. Gardner emphasized “What is it you need to see for you to feel that you can trust this: product, idea, website, presentation etc?” Our goal as New Product Developers is to find unknown Wows. People will pay more for the same product if it involves a better emotional experience.
Dr. Gardner gave several examples of the companies with the top converting websites using this type of testing:
It would be tempting to simply copy these sites, but what works for their customers many not sell so well for other products or demographics.
I want to thank Dr. Joely Gardner for her great presentation on A/B testing.