Two recent articles –  a Wall Street Journal article Your E-Book is Reading You and a New York Times Article Moneyball for Publishers – discuss the ways in which publishers and book retailers are using digital data to understand how readers react and engage with e-books. New data analysis techniques can look at how quickly readers finish a book, how far readers get in a particular book before giving up on it, and can even assess differences between readers (based on gender, age and other factors) in terms of their reading behavior. Sounds intriguing, doesn’t it? Especially given many publishers would admit they don’t really know a lot about their readers or their reading behavior when it comes to specific books.

Jellybooks, a so-called ‘reader analytics company’, offers publishers the ability to track reading behavior by giving a group of readers free e-books and digitally recording their reading behavior. In this way they are hoping to demonstrate to publishers how often the books are opened, how quickly they are read and (if a reader fails to complete the book) at what point a reader’s interest began to wane. According to Jellybook’s data fewer than half the books tested are finished by the majority of readers (ugh!) and that most readers give up on a book in the early chapters (which is hardly unexpected). Again, intriguing…

So what could publishers potentially do with these data? Well, given the plan is to track reading behavior prior to a book’s publication, these data could be used by publishers to formulate their marketing plans (spending less, I assume on the books that ‘failed’ in the test group, and spending more on those the test group completed quickly). Publishers could also use the data to identify the type of readers that respond well to a book and produce a more targeted marketing plan. I assume another option, in the future, could also be the possible ‘casting adrift’ of authors and books that failed to catch fire with test readers.

Both the WSJ and NYT article point out the potential pitfalls for these kind of ‘deep’ digital reader analytics programs. The test group might not represent, for example, the kind of readers a particular book would appeal to, or the group might not be a large enough (or diverse enough) to adequately represent the general book buying audience. There are also privacy concerns if this kind of analytics became widespread – although almost e-book publishers like Amazon, Apple and Google can already can track though their apps how many times readers open the app and spend their time reading. No doubt they already analyze their own data to glean a great deal of information about reader behavior.

Authors may also be cautious – while it would be pretty cool to see how readers respond to your books – what if they didn’t react as favorably as the publisher would like?? Is an author more likely to get dropped if the test audience doesn’t respond the way an author or publisher was hoping? Does a lukewarm reception in the test group mean that an author is likely to receive minimal marketing success (could failure become a self-fulfilling prophecy depending on the reaction if the test group?) I wonder too if reliance on digital data analytics could have a freezing effect on acquisitions of more quirky, eccentric or less mainstream books.

So TKZers, what do you think of the move towards deeper ‘reader analytics’. As a reader and as a writer, what benefits or risks do you see?

By the way, I am traveling to Nicaragua so, depending on access to the internet, I may or may not be able to join in, what I hope is a stimulating discussion on this topic!