Last week Amazon made another announcement with disturbing implications. Amazon revealed that their Kindle ebook sales “continue to overtake” the sale of print books on Amazon.com. But in the next few sentences, Amazon added some big disclaimers. So the truth is apparently that they’re not selling more ebooks than printed books.
Amazon issued a press release announcing that over the previous 30 days, they’d sold more Kindle ebooks than printed books “for the top 10, 25, 100, and 1,000 bestselling books on Amazon.com.” But it seems like an odd distinction, almost like they’re playing a game with the numbers. In the universe of all books sold, just how small is the piece that’s occupied by this month’s best-sellers? When you walk into a bookstore, how many of the books around you aren’t in the top 1,000? I’d guess it’s an awful lot — at least more than half.
And that points to the biggest quirk in Amazon’s calculation. Wikipedia notes that Amazon’s own list of best-sellers “tends to favor hardcover, more expensive books, where the shipping charge is a smaller percentage of the overall purchase price or is sometimes free, and which tend to be more deeply discounted than paperbacks.” In fact, best-selling books are often new books — which are first available only in hardcover editions. So Amazon isn’t talking about a typical sample of all books that are sold; instead they’re sampling an unusual subset where hardcover books are still very much over-represented.
In the real world, hardcover books represented just 23% of all books sold last year, according to Nielsen’s Bookscan service. But Amazon used this anomaly to announce that ebooks in their Kindle store were outselling all those expensive hardcover books. In July one analyst did some quick calculations based on Amazon’s other public statements, and concluded that over 70% of the books Amazon sells were still printed books. And since Amazon sells more ebooks than just about anybody, he reports that sales still remain very strong for the printed book, with Amazon’s ebook sales only representing “the equivalent of 6% of the total print book market.”
I’d publicized the analyst’s conclusions, and it ended up getting some attention from an MIT technology blog and the popular web site TechDirt. I half-wondered if Amazon’s latest press release was an attempt to address their skepticism by creating a new announcement where ebooks now seem to be finally outselling books — when they actually aren’t. After all, if Amazon really were selling more ebooks than printed books — across the board — obviously they would’ve announced that instead. (And how else could ebook sales “continue to overtake” print book sales? They’ve either passed them, or they haven’t!)
Amazon’s press release quotes Steve Kessel, the Senior Vice President for the Kindle, saying that Amazon’s ebooks “are also outselling print books for the top 25, 100, and 1,000 bestsellers — it’s across the board,” though apparently “across the board” actually means “the small portion of the board which contains expensive and pre-dominantly hardcover best-sellers.” But I also noticed their calculation specified sales to “Amazon customers” rather than “Kindle owners.” This seems to confirm reports earlier this month that 1 in 5 people buying ebooks from Amazon’s Kindle store don’t actually own a Kindle (according to a new technology survey). So Amazon may be selling more ebook versions of (expensive and pre-dominantly hardcover) best-selling books — but a lot of those are only being read on iPads and iPhones.
And Amazon also specified that their statistic was for “the last 30 days” — which could represent a one-time spike in the month after Amazon released the cheapest Kindle ever. I know I’m being cynical, but at least I’m not the only one. A reporter at Barron’s financial blog complained that Amazon’s announcement was “completely lacking in informative quantitative detail.” And a columnist at PC World notes it’s not the first time their statistics have made a strange comparison.
“Amazon has a tradition of playing these stupid mind games with the press… Amazon really took the cake for its silly numbers game last December when the company announced it had sold enough 8 gigabyte iPods during the holiday season to play 422 years of continuous music. The company also claimed it had sold enough Blu-ray disc players during the 2009 holiday sales blitz that if you lined up all the players side-by-side they would stretch for more than 27 miles. Huh?
He suggests that Amazon is guilty of foisting on the public “some random statistic that would be more at home in the Guinness Book of World Records than a quarterly sales announcement. ‘Hey look, we sold more Kindles in Q2 FY2010 than the weight of three pregnant Kenyan elephants.’ Good for you. Oh, did I mention that Amazon said it sold more Kindle books than print books for the top 10, 25, 100, and 1,000 bestselling books on Amazon.com during the past 30 days? Wonder what that actually means?