By Edward Nawotka, Editor-in-Chief
Ebook distributors and platforms have an unprecedented amount of data about what and how you read. A panel at Digital Book World last Thursday, offered several insights on how companies are using their data, what they have found, and the need to develop standards.
The following is an edited transcript of their remarks, sometimes paraphrased for clarity:
Chris Kenneally, Director of Business Development & Author Relations at Copyright Clearance Center moderated the panel and began by surveying the panel as to how many people are reading on one device or multiple devices.
Jared Friedman, Co-founder & CTO of ebook subscription reading service Scribd said “We were really excited to see half of readers using two or more devices, and 10% of reading three or more devices. Readers are allowing digital devices to permeate their lifestyles.”
Michael Tamblyn, President and Chief Content Officer of Kobo noted that you really need to consider whether you are talking about e-readers or through apps. Generally, he said that 15% to 20% of Kobo readers are multi-device users. “But when you look at other countries it is different. In Japan, there are no e-readers. It is a smartphone and tablet market, so the multi-device users are much higher.” This was confirmed by Micah Bowers, Founder and CEO of white-label e-reading app developer Bluefire: “We see that 40% of our users are reading on multiple devices.”
Of course, this is the just one small aspect of the user data digital platforms are tracking today, which covers everything from the amount of time you read, when you read, to the types of books you read at certain times of the day or even certain times of the year.
“For example,” said Kobo’s Tamblyn, “At Kobo we see romance reading really pick up in the summer, while New Year’s usually starts a trend in people reading non-fiction…but that said, most readers are pretty stable throughout the year. A romance reader is a romance reader all the time, and science fiction reader is an science fiction reader all the time.”
What is always interesting to hear more about is what types of books readers are most likely to finish. At Scribd, only 5% of readers opening a health and fitness book are likely to finish it, and yoga is the least finished category of books. But, noted Tamblyn, “You need to look at the differences between fiction or nonfiction, as nonfiction tends to be dipped in and out of it a bit more.
David Burleigh, Director of Marketing & Communication at OverDrive, which is a major supplier of ebooks to the library market, said, “When we looked at the top three titles from last year — Unbroken, Gone Girl, The Fault in Our Stars — some 6-11% of books are finished, 30% made it halfway, but this may be due to the end of the lending period or their having gone to buy the book.”
Bluefire’s Bowers confirmed that “the chance of someone finishing the book goes up the further someone has read in the book,” and added, “What was surprising to us was the sheer amount of time people spent reading: 35% read for an hour a day, 17% read for over two hours a day.”
Meanwhile, at Scribd, just 10% of people who left a fiction title for a month or more are likely to return to it.
When it comes to book purchases, Kobo found that if you can get 10% of the way into a book you have some momentum and you are likely to become a purchaser (though that doesn’t necessarily mean you will finish it…see above). At Bluefire, which does not have an ebookstore of its own, “33% are side-loading files, 2.5% of people are reading free books from the app, the remainder are downloading them from libraries, retailers, and other platforms.”
Data analytics is also moving well beyond the merely quantifiable. It is best applied in the book business to looking at human behavior. To this end, Scribd conducted an experiment which looked specifically at the habits of users who read Kurt Vonnegut novels. Friedman said that the company was “motivated to go beyond just looking at a book and looking inside a book as to why or why not something would work.” And so Scribd “took a bunch of popular authors who had authored a considerable number of books and look at the reading behavior. Did it change depending on the difficulty of the book? Where there fall off rates? Is there a particular point where the author seems to lose the reader? For Kurt Vonnegut in particular…there was an enormous difference between the reading speed and completion of the book. Cat’s Cradle is more accessible and was read relatively quickly, while his more experimental work is read about half the speed.”
As time wound down, Bluefire’s Bowers reflected that when it comes to data analytics, “you have a several-hundred-year-old industry that is only beginning to access this data in the past year. And as we move forward we are going to need to look at the nomenclature, to develop standards.”
Echoing this point, Tamblyn underscored just how difficult it is to make even the simplest assumptions about reading habits: “[At Kobo], we are still struggling to figure out where books end and where they actually begin…”
You can listen to the full panel here.