Every time you design a new landing page, add a new product to your website, add keywords to your campaign, rewrite a pay-per-click (PPC) ad, craft an email campaign, or redesign your website, you are making assumptions about how your visitors will react.
Jeff was telling me all about his efforts to relaunch his company’s websites this past weekend. Of course, my conversation went to customer experience, analytics, optimization and testing. For numerous reasons, Jeff didn’t really have any analytics for the old site, so he was figuring it would make sense to launch the new site — with its vastly improved customer experience — and wait six months to collect data before even considering any testing. His approach is not all too uncommon.
Let me make this clear: Jeff and his team are top notch. These are just the realities of online retail; it’s difficult, and sometime more so in a company that has such a long retail tradition as HBC.
Jeff and his team made assumptions about new technologies, features and layout for every page of their website. Some of these assumptions were explicit (launching customer reviews with Bazaarvoice would improve the experience and conversion rates) and some implicit (we’ll talk about that below). Some I am sure would be validated right after launch. Others would require testing.
Right there on the table I had Jeff draw the product page layout they had developed. (Well, it was on a piece of paper on top of the table cloth, but you get the point.) Jeff and his team had made the decision that product images would be on the right side of the page. One of the other retailers at the table said theirs was on the left side. So I asked Jeff, “How do you know it won’t cost you a million dollars during the next six months by having your product image on the right-hand side?”
While it might be easier for readability to have it on the right-hand side of the page, the majority of retailers have the product image to the left of the copy. I could see the agita stirring right there (and it wasn’t the Italian food we just ate). “You made an assumption that your product page would work better this way during the design process,” I said, “and you know what happens when you assume. Why not take it to the court of last resorts, the folks who vote with their dollars — your website visitors — and test it?”
I know how smart Jeff is, but do his bosses appreciate everything he brought to the table? He doesn’t have analytics showing lots of benchmark data, but testing would let him prove the assumptions and, most importantly, continuously improve the website. He could then show his organization the value provided by testing his assumptions, learn from both the successes and failures, and focus on improving the experience by adopting a culture of “Always Be Testing.”
I only gave Jeff one or two simple test ideas because I didn’t have the sites in front of me, but they are now live. You can check them out and feel free to offer suggestions in the comments below for things you think Jeff should test or things you’d would like to test if you could:
When would you like to prove how smart you are and make more money with your website?