Many of you know about A/B testing on websites: showing one of two versions (A and B) of a given page to visitors, and seeing which one generates a better result.
The ElasticPath e-commerce blog, GetElastic (recommended), has a nice post about A/B testing for wine, complete with a real-life example of a wine product page. The results are, to say the least, interesting (and counter-intuitive).
We'd all like to think we know best when it comes to what appeals to website visitors (and, in my experience, wineries are particularly guilty in this regard), but testing reveals the (sometimes surprising) truth about what drives visitors to take action.
If you have an A/B testing story to share, please do so in the comments.
Update: here's a link to the full case-study documenting a 41% lift in revenue per visitor.