Wednesday, October 05, 2022

Where's all the data?

Fascinating discussion about how book data can inform cultural awareness and potentially improve publisher and the reporting on book publishing.

Where Is All the Book Data?

Perhaps most importantly, however, it is likely that books end up much more racially homogenous—that is, white—as a result of BookScan data, too. For example, in McGrath’s pioneering research on “comp” titles (the books that agents and editors claim are “comparable” to a pitched book), she found that 96 percent of the most frequently used comps were written by white authors. Because one of the most important features of a good comp title is a promising sales history, it is likely that comp titles and BookScan data work together to reinforce conservative white hegemony in the industry.


The many ways that SPL checkout data might be used to understand readers or literary trends are still relatively unexplored. In 2019, The Pudding constructed a silly “Hipster Summer Reading List” based on SPL data, highlighting books that hadn’t been checked from the SPL in over a decade (a perversely funny list but definitely a terrible poolside reading list).

This checkout data is also used internally for a variety of purposes, including to make acquisition decisions, as SPL selection services librarian Frank Brasile explained. But the factor that apparently influences library acquisitions the most is simply what the Big Five choose to publish. “We don’t create content,” Brasile reminded me in a somewhat resigned tone. “We buy content.” To a large extent, then, public libraries inherit the pervasive, problematic whiteness that is endemic in the publishing industry.

As head of Bowker many years ago, we collected book data for higher ed publishers and charged a lot of money for it. At the time we were very interested in expanding into trade books but were not able to pull off the deal to buy the UK firm which launched Nielsen Book Data in the US. Even then, it was clear there were many holes in the data - even with more corporate booksellers - and it only reflected a segment of the market. While this article focuses on trade data, I would speculate that college textbook authorship is singularly (if not more so) (mis)representative.




No comments: