No Longer Playing Games: The Significance of Analytics in Book Publishing

Until recently, book publishers and authors played a tiresome game of pin the tail on the donkey when it came to surmising how a book would perform on the market. Thanks to the growth of books in digital platforms, publishers and authors do not have to play that guessing game anymore. Instead, they now have access to measurable analytics obtained from e-reading devices that provide many insights into readers’ consumption of e-books. The knowledge gleaned from this data in turn enables publishers and authors to make better-informed decisions on what to publish. Of course, the matter is not all as simple as using this data solely for market research. Traditionally, publishers have ultimately controlled the shaping of content by selecting and publishing what they supposed readers would want to read—playing the role of the gatekeeper[1]. Analytics, however, are rapidly changing the process in which content is shaped, because readers can now give both direct and indirect feedback on the success of a book. The effect that analytics has had overall, then, is that it has eliminated the static process of publishing books, and developed a new publishing model that is agile[2]. From a social perspective, however, there are concerns that “just as Web sites try to adjust their content to move as high as possible on the Google search results, so will authors and publishers try to adjust their books to move up the list” (Johnson) and that “a data-driven approach could hinder the kinds of creative risks that produce great literature” (Alter). In other words, the social currency[3] of books as cultural capital could become compromised due to analytics. However, should traditional publishers follow in the footsteps of data-driven book publishers, such as Kobo, Coliloquy and Sourcebooks, and embrace the use of the agile publishing model, not only will they thrive in the future but, in theory, they could use the analytics garnered from digital publishing platforms to successfully preserve the social currency of books.

To delve further into this idea, one must first look at the kinds of analytics that are available and how data-driven publishers are utilizing them. There are two sets of analytics available to publishers: indirect and direct. Indirect analytics are objective pieces of information collected through e-reading devices recording the habits of readers. These analytics are acquired the moment someone purchases an e-book and opens it up on their e-reader. Once this occurs, publishers can see in real time “how many times a book is opened, how many pages are read at one time, how long a book takes to finish, how far readers get into books before quitting (if they dislike it), what consumers read before and what they read next” (Richardson). By coupling analytics on reading habits with additional information about a reader’s age, gender, and the genres they are reading, publishers are now aware that books with a female protagonist are 40% more likely to become a bestseller; that the top genre men read is historical fiction, while the top genre women read is romance; that women are 50% more likely to finish a book than men; and that the average session length of readers under the age of 40 is twelve minutes, versus the twenty minutes of those over the age of 40.[4] Direct analytics are subjective pieces of information that readers provide actively. This information is obtained when readers highlight certain passages, make comments about the content, or rate the book. With these two types of analytics available, publishers are able to fully ascertain how readers are engaging with their books and can no longer say they do not know their customer base.

When publishers employ analytics like these in their publishing programs, this is what is known as agile publishing, and there are three digital publishers that currently stand out with this initiative and that traditional publishers should take note of: Kobo, Coliloquy, and Sourcebooks.[5] Kobo launched its self-publishing and self-service platform in 2012 called Kobo Writing Life (KWL) and the service’s strength lies more on the side of indirect analytics. KWL provides “a dashboard offering a variety of data that allow an author to track the sales performance of a book across multiple markets and track the impact of promotions and sales” (Reid). The main beneficial feature of KWL is its “live metrics that give you a sense of where your books are selling and how many, all in real time” (Kozlowski). The other notable feature of KWL is that it allows authors to set the price of their e-books[6]; so, given the live data they receive, this frees authors to play with prices and learn the optimal price-points for their books. The advantage to receiving live analytics is that it gives flexibility in how to go about tweaking the sales strategy of a book once it has been published.

Coliloquy is a digital publisher that has its strength on the side of direct analytics. The company strives to actively integrate reader feedback in their publishing methods, and as they say “push the boundaries of how we think about narrative and storytelling.” What separates Coliloquy apart from other digital publishers is that they make all of their books into “active applications, rather than static files,” allowing their authors to create episodic, serial stories and use “engagement mechanics like choice and voting, branching story lines, re-reading loops, and personalized content.” Enabling readers to directly comment on a book and influence its development increases their engagement with both the author and the book itself. The nice thing about Coliloquy is that its use is not restricted to new titles alone. Traditional publishers could easily take advantage of Coliloquy’s analytics to revitalize backlist materials and receive feedback, since all that would need to be done is to create an active application. Thus, Coliloquy’s use of analytics in their books goes to show that analytics do not necessarily hinder the creative risks that produce great literature. If anything, they are opening the door for publishers, authors and readers to collaborate, and making the book an intersection for social engagement, unlike anything we have seen before.

As seen with Kobo and Coliloquy, the use of analytics can help enhance publishers and authors enhance digital book content in different ways. But how can analytics serve to help enhance print books? Sourcebooks is a good example of a publisher balancing between “both classically physical and dynamically digital” book formats. Of the three digital platforms outlined here, Sourcebooks is the one that traditional publishers and people worried about the decline of print books can look to to alleviate their fears. The company’s focus is on authorship in a culture of collaboration, and so Sourcebook also focuses on agile publishing. Rather than publish a print edition first and then release a digital edition, Sourcebook reversed the order and “began experimenting with a new model of serial, online publishing” (Alter). With the traditional publishing model, publishers do not have a grounded way of knowing how a book will do once it is released. What Sourcebooks has done is to release “early online editions for half a dozen titles, ranging from romance to young adult to nonfiction books, and has solicited questions and suggestions from readers (Alter). This is a clever way to incorporate customer feedback, and a much more practical way to publish print books. In this manner, publishers can improve a book and feel more confident about how the book will fare in the market. Publishers may also feel a little more comfortable with this method as well, because an author will have originated the content and the creative risk is still at the heart of the work. Receiving feedback from readers earlier on then is really not much different from how an editor queries an author during the manuscript phase. And much like how the expert knowledge of editors helps to strengthen the quality of a manuscript, so too would the use of direct analytics from readers help add to the social currency of the book.

Technology can be an intimidating thing for publishers who have grown accustomed to traditional publishing methods. However, there are some brave souls who have already started paving the way to making the publishing process dynamic. As these three data-driven publishers show, there are many reasons why publishers ought to implement analytics into their publishing programs with the main reason being that it opens up opportunities for creative collaborations. Analytics are simply tools that help publishers and authors “work to create better, more satisfying content” (Richardson).



[1] See the concept of the pre-artifact system in Craig Mod’s article “Designing Books in the Digital Age.”

[2] Sourcebooks has an excellent explanation of what agile publishing is: “The agile publishing model (APM) relies not only on the author for providing the content and overall direction of the book, but also on the community of readers to provide proactive reviews and feedback on the materials provided. Working together, the author and community will shape and change the content as the book moves from its initial stages as an interactive, digital platform to the final published product.”

[3] As defined by James Levy in his article “DNA of a Successful Book.”

[4] Statistics from Hiptype’s infographic on the DNA of a Successful Book: http://www.hiptype.com/infographic

[5] There are several more leading companies in this venture, such as Lean Publishing, Barnes and Noble’s Pubit, and Amazon’s KDP, but this paper intends to give a sampling of the different ways to benefit from data-driven book publishing platforms.

[6] Authors can even set the price to free.

 

Works Cited

Alter, Alexandra. “Your E-Book is Reading You.” 19 April 2012. [http://online.wsj.com/article/SB10001424052702304870304577490950051438304.html].

Coliloquy. “About.” [http://www.coliloquy.com/about/].

Hiptype. “The DNA of a Successful Book: Infographic.” [http://www.hiptype.com/infographic].

Johnson, Steven. “How the E-Book Will Change the Way We Read and Write.” 20 April 2009. [http://online.wsj.com/article/SB123980920727621353.html].

Kozlowski, Michael. “Kobo Writing Life to Be Launched Soon.” 8 July 2012. [http://goodereader.com/blog/e-book-news/kobo-writing-life-to-be-launched-soon/].

Levy, James. “The DNA of a Successful Book.” 25 July 2012. [http://blog.hiptype.com/the-dna-of-a-successful-book-1].

Mod, Craig. “Designing Books in the Digital Age.” Book: A Futurist’s Manifesto. Eds. McGuire and O’Leary. [http://book.pressbooks.com/chapter/book-design-in-the-digital-age-craig-mod].

Reid, Calvin. “Kobo to Launch Kobo Writing Life Self-Publishing Portal.” 5 June 2012. [http://www.publishersweekly.com/pw/by-topic/digital/retailing/article/52275-kobo-to-launch-kobo-writing-life-self-publishing-portal.html].

Richardson, Andy. “Ebook Analytics: Knowing Your Audience.” 24 August 2012. [http://www.econtentmag.com/Articles/Editorial/Commentary/Ebook-Analytics-Knowing-Your-Audience-84556.htm].

Sourcebooks. “Why Sourcebooks.” [http://www.sourcebooks.com/company/why-sourcebooks.html].

3 comments:

  1. Duany, great paper. It very clearly laid out some fascinating methods and potential benefits of using analytics.

    I’m not convinced reader analytics belong anywhere near great literature, but I can certainly see the use in other types of publishing, like maybe textbooks, instructional books, or genre fiction. With any book that has a concrete, desired outcome for the reader (even if that goal is just to while away the hours in a heightened state of arousal, like the latest pub phenom), or a book whose goal is to appeal to as broad an audience as possible, it makes sense to understand exactly how the book is failing or succeeding in getting its readers to that outcome, and then tweaking successive iterations accordingly.

    I think the various approaches you’ve mentioned could make a lot of terrible books better, but I’d be surprised if they could make good books great, and analytics at its most basic form is already being blamed for thwarting the emergence of new writers — over the break there was an article in the Globe (another Rowly cameo, natch) blaming BookNet Canada’s sales tracking for publishers dropping mid-list authors.

    For more literary works I think publishers should be incredibly weary of enlisting crowd wisdom. I guess in its most innocent form it’s no different than an author soliciting opinions from friends, and then ignoring the ones he doesn’t like — a timeless method of social engagement and crowd collaboration — but anything much beyond that pretty quickly runs the risk of destroying the author’s voice and vision, and bumps into maddening questions of authorial integrity, art vs content, whether a writer should ever concern herself with her potential audience (don’t tell Mary), or just focus on bumbling along towards wherever her instinct takes her, and hope that she happens to be good enough for people to want to read her.

    I guess it’s back to art vs commerce: this sort of thing could work well for books that are more tailored to a definable audience (or an audience of all), but I think it’s of less use in producing great art, which requires equal parts hard work and magic. That should go over well in job interviews…

    Thanks for the interesting read!

    Mike

  2. Although I do believe analytics are going to play a greater role in publishing, and should, I agree with Mike that on a title-by-title level it’s not going to make good books great. My belief here is based really on experience related to website analytics. Just because the data is available, it doesn’t actually give a full picture of what’s going on. And the data is really only as good as the data interpreter.

    Publishers haven’t historically invested in data analysis so although they can gather data like where in a book people start and stop or how many times the book is opened, I’m not sure it’s clear what one would do with that information.

    Now, if we look at your examples, it’s really Kobo and Coliloquy vs. traditional publishers (even innovative ones like Sourcebooks) who stand to gain from analyzing “big data”, data sets that are so large they require management of that data to an extreme that is well beyond what is collected through one-off website or ebook stat gathering tools, which publishers already struggle with assessing.

    Kobo, for example, is more likely to be interested in what books people read before and after certain titles that are selling well. They are more likely to be interested in vast data sets that provide aggregated information like you quote above related to how women engage vs. men, total dollar value spent, and categories that perform well but have low publisher competition. i.e., if Kobo sees a high degree of consumer interest in a category but a low degree of titles in that category, it’s a good opportunity for Kobo as publisher to step in with titles of their own.

    The difficulty for traditional publishers is that Kobo isn’t necessarily sharing that data but rather capitalizing on the business opportunities themselves.

    So I see your point at the end that it’s important for publishers to implement analytics with the main reason being “to open up opportunities for creative collaboration,” but I think the main reason will be to unveil publishing opportunities.

    This Publishers Weekly article offers some further insights here, along with some of the tools like Bookseer and CoverCake that publishers can use to actually implement analytics.

    http://www.publishersweekly.com/pw/by-topic/digital/retailing/article/52119-can-b-g-data-fix-book-marketing.html

  3. Another useful tool for publishers to acquire analytics about eBook engagement is through Hiptype.

    Publisher’s Weekly wrote a good article about it. Here is a snippet from that article explaining what kind of information can be obtained: “Among the information Hiptype provides is performance data (how far readers are getting through titles and the areas with the highest reader attrition) and anonymous, aggregated audience insights. The latter, to help publishers understand who is reading their books, is not shared with any third parties and a reader, notified at the beginning of the e-book about the anonymous usage statistics function being enabled, can choose to opt out.”

    They offer a 30 day free trial, a basic plan that costs $19 a month, and a pro plan that costs $99 a month. There are no setup fees to begin.

    What I thought was really beneficial about Hiptype is that it works in eReaders your audience already uses. There is no need for them to download a new app.

    It’s great that publishers can now have access to such valuable information.

    http://www.hiptype.com
    http://www.publishersweekly.com/pw/by-topic/industry-news/publisher-news/article/53364-hiptype-analytics-for-e-books.html
    http://www.hiptype.com/#pricing

Comments are Closed.

css.php