Among the many overzealous, under-synthesized tech business stories today, perhaps the most surprising is the news that Amazon is now selling more ebooks than print books. 105 ebooks for every 100 print books, as it happens. While 105 > 100, a more accurate but less scintillating headline might be, “Amazon ebook sales on parity with print book sales.” But I suppose I could write something every day about journalistic spectacularism.

I want to make but one point about this news, and that’s to remind all the mouth-foamers that Amazon sells far, far more copies of their top 100 books than the rest of their titles, which number in the seven figures. Here’s a rough estimate from Morris Rosenthal, who has been attempting to track Amazon sales rank to units sold for some time. The left is an estimate of the top 1,000 books, and the right is an estimate of books ranked 1,000 – 1,000,000. Click for a larger version.

The point I want to make here is that the vast majority of sales of books in general on Amazon comes from the bestseller list… that is, the top 10 books are responsible for something approaching 20,000 sales per day according to Rosenthal’s estimates.

All of which means that when a report like the one linked above averages all of Amazon’s sales and attempts to draw a conclusion about ebook readership and market patterns, it’s naturally going to be heavily skewed toward the top 10-100 books. Which is fine, that’s just how the market works. But people read lots of books for lots of reasons, and it’s not yet reasonable to conclude that ebooks are have overtaken print books—and that’s clearly the provocative, sensationalist claim that Amazon, the press, and perhaps the tech industry more generally want to make.

Just to be completely clear: I’m not pooh-poohing the rising popularity of ebooks. It’s just annoying when reports like this are read without greater thought to what the data really tells us. And that’s something quite specific: that ebooks are as popular as print books among books that sell well to a general audience.

published May 19, 2011

Comments

  1. SabotageGigante

    The other place this reporting fails is that it counts number of books rather than the sales values of the book. It fails to account for free, or nearly free ebooks.

    Do a search for “Alice’s adventures in wonderland”. The cheapest ebook, free, several other editions are $0.99. The cheapest paperback 4.99 plus shipping.

    How many free ebooks do I have to “buy” from Amazon to skew these numbers?

    This is one of the virtues of ebooks. But it makes the statement that much weaker.

  2. Ian Bogost

    Good point.

  3. Gilbert

    Bug report: all of the text past the figure turned into one giant hyper-link for me. I’m running Google Chrome 11.0.696.68 on Windows 7.

    Thought that might be helpful if you’re using some kind of custom blog software

  4. Ian Bogost

    Oops, just a missing tag. Fixed, thx.