People often ask me about the books I have written, “how many copies have been sold?” And it surprises them to find out that I have no idea. For instance, my book Parentonomics was published initially by UNSW Press and then MIT Press before being translated into, I think, 6 other languages each with their own publisher. For each and every one of them, total sales are a mystery. Sometimes I plead and can find out this quarter’s sales but the overall picture cannot be grasped. And certainly there isn’t enough information to work out if I should write another book or not. And trying to glean this from Amazon sales data is, at best, a rollercoaster of wild changes in rankings. (You might think that some payment might generate that incentive but in reality, for books that aren’t best sellers my interest is in reach because a pure monetary motive won’t cut it when there isn’t much money).
This article in the NYT struck a chord with me. I’m not alone. Just when I thought this sort of frustration was a small publisher thing, here we find out that the biggest publishers have had the same issues.
Dave Cullen, the author of “Columbine,” a nonfiction book published in 2009 by Twelve, part of Hachette, said he had become accustomed to haranguing his publisher for sales data. While his publisher was patient and accommodating, Mr. Cullen said, he frequently wondered why he could not check the same information himself.
“Some of this is the publishers trying to be competitive,” Mr. Cullen said. “And some of it is that they’re opening their eyes. Publishers didn’t realize the frustration that authors have.”
Three big publishers — Random House, Simon & Schuster and Hachette — will now put that information online. This is supposedly in response to Amazon who was vertically integrating into publishing with the hook of providing information. But why was that so hard to do? What barriers could there be to providing authors with information before? Surely, sales data has been electronic for decades. Even if it isn’t online, how hard could it have been to request? Why would this have been something “years in the making”?
Every speculation I have about this is sinister. But the economist in me wonders whether conspiratorial sinister motives could have really sustained themselves for this long. Perhaps publishers kept information for themselves to make it harder for authors to sell themselves to smaller publishers. That said, my guess is that the data existed somewhere and even smaller publishers could assess an author’s popularity.
Anyhow, if anyone has theories or information on the “the iron curtain of book sales data” please feel free to comment on it here.
Digitopoly 
So how does anyone know that they are getting contracted royalties, even now? You just have to trust them?
A few large players in the field with a virtual monopolistic grip on the market? The answer is quite simply that the big publishers run a confusopoly (term coined by Dilbert’s Scott Adams). Note that authors are to a large extent the customers of publishers.
Isn’t part of it the returns problem? That is, publishers know how many copies they have shipped to retailers, but not how many of those are going to be returned – and the proportion can be large.
In my own case, I got quarterly sales figures from the (small) publisher and then annual, but the answer was a polite “not many” no matter what the counting system.
Up until about 2000, with the rise of Bookscan, publishers had no way of tracking their sales with any kind of accuracy. They had to wait until they got either sales data or returns back from retailers (independents, who weren’t great about reporting, and the chains, who produced lots of churn, sometimes returning a title and ordering multiple copies in the sam period). Bookscan, the Nielsen service, gave them some sense of what was selling, but only tracks sales through certain retailers. For some publishers, Bookscan only captures 50% of sales.
Providing authors with realtime sales data divorced form context can produce more problems than it solves. What does 300 copies a week mean? Is that good? What does it include? How about direct sales, international sales? Why did the total number sold go *down* this week? How does my rate of sale compare to James Patterson? Am I happy? Should I be? Please call me and tell me I’m a good author . . . and so on. Seriously. It can get angsty.
That’s not a good reason not to provide data, but it does point to why publishers might be hesitant. Figuring out the presentation of the raw data (something publishers are not historically good at) can be difficult and time consuming, especially when you’re understaffed and chasing sales, marketing, and publicity opportunities.
By the way, Tom: I love Whimsley.
The problem is partly returns, which lag many months from the “sell-in” to give the publisher the net “sell-through” at retail. The Neilsen data does not capture all the market, as not all retailers report. Plus the data is sometimes manipulated by publishers which “cashier” phantom sales through reporting retailers at a short margin to the retailer. For example, this may be done for books that are given away, or sold at a conference where the author is speaking, in the hope of driving the title up the best seller lists.
One more question” Amazon … was vertically integrating into publishing with the hook of providing information. But why was that so hard to do? What barriers could there be to providing authors with information before
What information does Amazon provide to authors? The ranking is addictive, as many authors know, but deducing actual sales figures from it is almost impossible. Is there more information that I am not aware of? Because if it’s just ranking then Amazon is doing no better than publishers.
(PS: Thanks Tim!)