Book People Cling to Fuzzy Math; Hard Numbers Are … Kinda Scary!

Whoever said “Change is good” clearly never worked in the publishing industry. The book business generally likes to do things the way it has always done them, and any proposed alteration to the plan is usually met with ambivalence at best. So only if you’d just dropped down from Mars would you expect book people to be overjoyed by the quick rise and ubiquitousness of Nielsen BookScan, a sibling of the TV-rating company and the first credible data-gathering that the industry has ever seen. Now publishers, agents, bookstores, journalists and any other publishing watchers can, for $100K plus, contract with BookScan to gain access to a Web site that will tell them, in hard figures, just how well or how badly their books have sold. (Publishers can also ask BookScan to furnish regular reports, breaking down the numbers by region of sale, price of book and other criteria, but that’s an additional charge.) The service has been around since 2000, when it had just a handful of clients, but in the past year it has achieved critical mass: All the major trade publishers have signed on.

You’d think this would be good news for everybody: Publishers and agents and authors will have a more realistic basis on which to negotiate future projects, because they can know for sure how similar ideas or the author’s previous books have fared. Also, because BookScan can “slice and dice” the numbers a lot of ways, including by price and by region, publishers using the service can get an edge on pricing and marketing. The advent of BookScan promises that news outlets and their readers will no longer have to try to decode best-seller lists, the most important of which-the New York Times best-seller list-compiles its data by a complex system of weighting and extrapolating that only a Kremlinologist could love. As for authors-well, the BookScan reports that your agent or publisher can show you are a whole lot easier to read than your average royalty statement. And besides, such reports can be made available weekly.

But this being the book business, reaction to BookScan has been far more complex. Why? Because for all the bottom-lining and conglomerating of recent years, publishing is still a business built on passion and perception. “We talk about the BookScan numbers in editorial meetings,” said one prominent publisher who didn’t want his name used-and who then admitted that the numbers are not always factored in when projects go to auction. “On the one hand, we want to buy books that sell, and BookScan can give us an indication of how well a project might sell,” this person continued. “On the other hand, when you want a project, you usually have to pay more than somebody else. The competition can get very heated-and you end up paying more than you probably ‘should,’ based on the numbers, because you don’t want somebody else to publish it.” Besides, an editor hell-bent on acquiring a book can be perversely happy about being in the dark, numbers-wise. “O.K., sometimes I knew the agent had puffed up the author’s track record,” said one former editor. “But I was grateful to be able to talk back to a skeptical marketing department that had doubts about selling the book I really wanted. Nobody had inarguable numbers.”

Agents, too, are wary of BookScan, probably precisely because hard data undercuts what they do best: the enthusiastic spinning of a project’s viability. (“Spinning” in this context is a “generous word,” one disgruntled publisher told me.) “These numbers are both a good thing and a bad thing,” said the agent Robert Gottlieb, chairman of Trident Media Group and agent to such megasellers as Janet Evanovich and Dean Koontz. “They are an indicator, but an overreliance on them can be misleading.” For one thing, he and others point out, they tell only part of the story; according to Jim King, vice president and general manager of Nielsen BookScan, the service receives reports from most, but not all, retailers-Wal-Mart is conspicuously absent on the roster, for example-and the company suggests that it reports around 70 to 75 percent of all sales. So a book that sells very well in non-reporting retailers-a book like The Purpose-Driven Life , for example, which has been sitting at or near the top of virtually all nonfiction best-seller lists for many weeks-as well as the kind of mass-market, Wal-Mart- and book-club-popular authors that Mr. Gottlieb represents, may not appear as huge sellers on BookScan.

So BookScan isn’t foolproof-but it’s the closest the book business has ever come to knowing itself, and you’d think that fact-seeking journalistic outlets would embrace it. For one thing, it efficiently collects data that, until now, the periodicals have had to collect from retailers themselves by phone, fax or survey. But ambivalence reigns in journalism, too, apparently. While The Washington Post recently contracted with BookScan to receive its services-under their agreement, the paper can publish its rankings but not the hard numbers, according to Mr. King-the venerable New York Times has made no such arrangement, and a spokesman for the paper declined to say whether its executives were even considering doing so. “We sample a portion of the known book-selling market, and we essentially extrapolate from that,” Richard Meislin, The Times ‘ editor of news surveys, told me.

But it was his next comment that made me think, for one minute, that I was talking to a dyed-in-the-wool book publisher:

“We like doing things the way we do them,” he said.