A/B Testing Cover, Title and Book Description

A/B experiments are a controlled form of testing where extraneous factors are eliminated as much as possible. At Jellybooks, we can A/B test a range of parameters.

Firstly, I a multi-title test Jellybooks measures how a change in cover, title or description affects the popularity of a title, i.e. is it more or less likely to be picked compared to other titles in the test?

Secondly we measure how changes in cover, title, description, length or form of content affect reader engagement. Does the completion rate increase or does the recommendation factor drop if the title is changes? These and other factor can be scientifically tested with reader analytics in combination with A/B testing.

AB testing Covers

Covers are one of the most common reasons for why great content does not sell. Rarely is down to poor design. More commonly it is design that is too adventurous or poorly communicates what the book is about. A cover does not only have to say “pick me” but also “pick because you will get this” and be honest about it. Reader analytics in combination with A/B testing gives an opportunity to test how different cover designs perform even before a book is published.

AB testing Book Descriptions

After the cover, the book description is probably the second most powerful influencer affecting purchase decisions. A good description should not be an academic summary of the book, but also a sales pitch for the book. This is notoriously tricky to get right, so it make sense to actually test it, both in terms of how it influences selection, but also in terms of how it affects completion rates and recommendation factor. Among all the meta-data elements that a publisher should optimise, book description is probably the most important, but also the most tricky one. Unlike cover design, it can be done by the author or editor, but the motto is test, test, and test again. The work of many months could depend on getting both the cover and the description right.

AB Testing Book Titles

This is a more unusual element to test, but in some cases, especially for non-fiction books it matters greatly and can represents the difference between stellar success and middling performance.