Kindle Sales Rank

How Many AmazonE Kindle eBooks Are Selling Based On Sales Ranking

Note: This analysis is in no way sponsored or approved by Amazon.com. For overall book industry sales regularly updated on this site, see: Amazon sales compared to Barnes&Noble and Borders. The following graphs were created in June 2009 with actual sales estimates solved by ratio from my page on estimating book sales from the Amazon sales ranking.

The first graph addresses Kindle ebooks with an average Kindle sales rank between 10 and 100,000. I don't try to estimate sales for Kindle ebooks with sales ranks below 10 because they are in the head of the curve where I have very limited data and where the actual sales fluctuate greatly with season, number of blockbusters, and I wouldn't put too much faith in the eBook sales estimates for the ranks lower than 1,000. My data for books in those ranks (which was translated into Kindle ranks) was based on a frantic data gathering session when Amazon (accidentally) revealed stock number in real time for all books a while back. BTW, I'm not sure if AmazonE is a good way to express the difference between the paper books and the eBooks stores on Amazon, I just wish they would do something to unmix the bestseller lists:-)

It turns out that Kindle eBooks are now selling a lot more copies than I would have guessed just yesterday. I suppose it's reasonable that Kindle owners are highly motivated buyers of bestsellers, where Kindle ranks are essentially equivalent to book ranks, and even if there are only a few hundred thousand Kindle readers in existence (Amazon has yet to publish a number), it's not like everybody in America is an active book buyer. Just for the sake of example, let's pretend that only 2 in 100 Americans is a regular book buyer, which gives you around 6,000,000 book buyers. And lets assume that there are 600,000 Kindle readers in existence. That would say that 1 in 10 regular book buyers owns a Kindle. Now, there are plenty of places non-Kindle owners can go to buy bestsellers, from Barnes&Nobles and Borders to independent bookstores, to supermarkets and mass merchandisers. So it wouldn't be so surprising that when it comes to top bestsellers sold on Amazon, about as many Kindle copies as paper copies are sold.

The second graph shows how Kindle ranks stack up against regular sales ranks with a high degree of certainty, assuming that Amazon isn't purposely skewing the data:-) I've already made one change to the head of the curve (the top 100 sellers) since posting it last week to reflect a shift in balance, it may take some time get the average relationship of top selling Kindle eBooks to top selling books since there are so few of either and and their relative positions change frequently. I was able to gather a series of very accurate data points for Kindle sales ranks bracketed by book sales ranks in Amazon bestseller lists for the bulk of the curve. As you can see, in the very top ranks, Kindle rank and sales ranks for books sometimes sell at the same rate, the bestselling "book" on Amazon at any give time may even be a Kindle eBook. By the time we get to a Kindle sales rank of around 500, the relation has already diverged to around 2:1, ie, an eBook with a Kindle rank of 500 is selling about as many copies as a paper book with a sales rank of 1,000. By the time we reach a Kindle rank of 10,000, the relationship is down to a little worse than 4:1, so a paper book with a rank of 40,000 is selling more copies. By the time we reach a Kindle sales rank of 100,000, the relation is just under 10:1, so a paper book with a sales rank in the low 900,000's is outselling a Kindle eBook with a rank of 100,000. Jeff Bezos has tossed out the stat that for books available in both paper and Kindle form, Kindle sales are running 35% of paper sales, or 3:1 paper, but that would be averaged over all Kindle titles. The 3:1 (paper:Kindle) comes at a Kindle rank of around 2,000, so something like half of Kindle titles sold are from the 2,000 most popular.

Another interesting thing is that the sales curve suggests (in the reliable section, no less) that AmazonE store is selling north of 600,000 Kindle eBooks a week. If we stick with the guess of around 600,000 Kindle readers in existence, that would mean that the average Kindle owner buys a book a week. That may be possible, or it could be that a lot of iPhone users are buying Kindle ebooks. If we double our estimate of the number of Kindles sold to 1.2 million, that would lower the Kindle eBook purchase rate to 2 per month per Kindle owner. Insider estimates for the number of Kindles sold are all over the lot, from around 400,000 to 1.5 million, my own guess is that when Amazon has actually sold a million Kindle readers, they will announce it publicly.

Given that there are only around 300,000 eBooks available for Kindle (last time I checked) it's impressive that they are making such and impact on the head of the long tail. And if California follows through on moving their textbooks to electronic form, I'm betting the Kindle 2 will be a major candidate, and California would be wise to go with Kindle readers and Amazon as the distributor rather than some clunky system of providing laptop equivalents and broadband access to all students to download texts. I should also note that many Kindle eBooks are free, classic out-of-print titles provided by Amazon (over 7,000, with the Adventures of Sherlock Holmes currently the most popular at #38). The number mean that Amazon, already a force to be reckoned with in the publishing world, may become unstoppable. What remains to be seen is if Kindle sales are exhausted with elites, techno-junkies and social signalers, or if it enters the mainstream, perhaps through the public schools as I've long predicted the dominant eBook reader of the future will.

I'm currently looking at the large number of Kindle publishers who are trying to make some money by selling various public domain works. Some publishers are using cheap copies of out of print books to attempt to promote their genre related titles - it doesn't appear to be working, but a good bid. The smart ones break story collections into individual short stories, which minimizes overlap with Amazon. You don't have to go very far up the cheapie list (Kindle eBooks arranged by price) before the Kindle sales rank is above 100,000, roughly equivalent to a million for a book rank. So while Amazon's free books may make up 5% of the popular Kindle titles, it doesn't look like the cheap promo copies and attempts and inexpensive public domain books (competing with Amazon) add up to much.A little further up the chain, 80 cents, etc, there are a ton of publishers trying to make a living off Kindle public domain books, but most aren't selling at all. Hard to compete with free, and the more obscure titles aren't in demand.

I'm also planning to redo the relationship curve a few times offline over the coming weeks just to see how it average out, whether it turns out that there are "hot" periods for Kindle, like weekends or nights, which have very such different selling patterns that multiple curves will be required.

I plan to update this page as frequently as I get new data, if you have something to contribute, please contact me.

Visit My Publishing Blog