An Earthquake Hits Amazon’s Sales Ranking

Anyone who’s ever written a book — and these days, who hasn’t? — can tell you that watching your sales rank on can be a pretty fun sport.

But something happened recently that made it a lot more fun for some people, and a lot less fun for others.

I noticed the change the other day when I checked the Amazon page for Freakonomics. For the first 12 or 18 months of the book’s release, we were generally ranked somewhere in the top 20, hanging out at No. 2 for a long time (we couldn’t get past the pre-release Harry Potter) and finally hitting No. 1 briefly after a 20/20 special on the book.

Then, for the past 6 or 8 months, we’ve floated around the 30 to 40 range. When something significant happened — a TV appearance, e.g. — there might be a bump, but it wasn’t usually drastic and it didn’t last long. We were like the guy who just hangs out at the end of the bar nursing his beer, not really hurting anyone but not leaving either.

And then, the other day, we dropped like a stone. As I write this, Freakonomics is ranked No. 194 on Amazon.

What happened? Did the steady stream of people who’d been buying one book for months suddenly stop buying it?

No: what happened is that the Amazon best-seller sales rankings suddenly got a radical overhaul. Here are the top 25 books on Amazon. As I write, Alan Greenspan‘s book (despite Levitt’s prediction) is still No. 1, and the runaway paperback hit Eat, Pray, Love by Elizabeth Gilbert is at No. 3. Those are the kind of books that have always been ranked high on Amazon. But the other eight books in the top ten are all either Curious George or Clifford the Big Red Dog books — inexpensive children’s paperbacks. This kind of book never used to rank high on Amazon, and I am not sure if this was a premeditated change, a quirk, or a mistake; but it certainly has changed the feel of the Amazon rankings.

Although Amazon is secretive about how its sales rankings are determined, I have always suspected that the list is weighted toward a certain kind of book — new hardcovers, to be precise. There could be a lot of reasons for this: a best-seller list looks more appetizing if it is full of new books instead of old ones; perhaps Amazon wants to sell more of the expensive hardcovers than it wants to sell cheaper paperbacks; etc.

I have no idea what precipitated this change, but I wonder if it has something to do with shipping costs. In the old days, customers often paid the shipping on Amazon orders. Nowadays, assuming that Amazon has been successful in converting a lot of its customers to its Prime membership (I sure signed up), which offers free shipping on most orders, the company is paying the shipping. In that case, selling a cheap and light 20-year-old Curious George paperback, even for $3.95, may be as profitable as selling the Alan Greenspan book. Does anyone out there have any insider knowledge of the Amazon change? And/or offer alternate theories?


The Principia Discordia is 100,047


I think this is just a big mistake.
First of all I think, that selling those cheap books cant be as profitable as selling the 20 dollar+ hardcovers.
And then I think, if it was as or more profitable, this strategy would be too obvious.
If you see the bestselling statistics for the first time, you just think "that cant be true".


Is it possible that the aforementioned sales are coinciding with back to school stuff, and schools are buying 'class sets' of these books, sending them through the roof?

Amhed Herrera

Maybe it has something to do with Amazon changing the way that books are ranked. A couple of months ago you would just get "#456,129 in Books", but now you get that number and you get something like the following:

#7 in Books > Children's Books > Educational > Explore the World > Fiction > Africa
#9 in Books > Literature & Fiction > Authors, A-Z > ( A ) > Allende, Isabel
#12 in Books > Children's Books > Animals > Elephants > Fiction


Yesterday afternoon, I happened to catch a rerun of your appearance on Beauty & The Geek. After reading this entry, I can't help but wonder - Did that appearance have any effect on your ranking?


My own experience with Amazon is mixed. I have been buying books and other stuff from Amazon for some time now. When I pre-ordered the last Harry Potter book, it did not arrive as promised. After complaining they sent a $5 coupon which you can use only on Amazon-shipped books. I buy books from other sellers especially those who carry technical and/or out-of-print books. I'm just dismayed that they would restrict the use of the coupon to their own supply.

As far as the ranking is concerned, I have seen major shifts in matter of days or even hours and their recommendations sometimes baffle me when they have nothing in common with what I searched for. I think that they promote some books with a tie-in to publisher's incentives and the author's publicity tours.


Just in time for your new children's book. Coincidence?


Never underestimate the power of sales and deals websites. Yesterday Amazon had a sale on children's books, news of which proliferated to sites like Slickdeals and Fatwallet. Slickdeals listed 9 Clifford books and 7 Curious George books all for a dollar or less.


Who cares? If I'm interested in a book, say 'Freakonomics', I'll buy it whether it is ranked #1 or #4356.


Well, if it is just a blip because of a sale, so be it. But if the algorithm has changed, then it certainly eliminates the need to spend time browsing the Top 25 list for interesting things that other people are reading.


I just checked out that SlickDeals forum, people were buying books in double digits due to the deal Amazon was making. Must be a big reason for the shock. All these kids books for less than $1 each will have teachers and parents flocking for them.

I suspect things will be normal after this deal.


I imagine the deals websites can have transient, significant impact on sales numbers on a quite regular basis. I have Amazon Prime, and regularly monitor them for cheap items I can have delivered for free (though I didn't check yesterday, and missed this deal for my son).

A secondary impact at a place like Amazon is that the very cheapest items can become filler items. So if you want to buy something that's $23 and don't have Prime, you can either pay $3.95 for shipping or pick up two one-dollar items to reach the $25 mark for free shipping. Items that are very cheap may therefore see elevated sales to people as a mechanism for saving money on shipping.


Re: Doug

Your comment is asinine. The question is about the nature of the Amazon rankings, not about what prompts you, Doug, to purchase books.


Here's a theory:

Someone at noticed that they keep track of which books sell the best, by actually counting ... you know ... how many books they sell each day.

From that, they figured out that their bestseller list was a bunch of BS. Those books that used to be on the bestseller list never were the bestsellers ... the list was a fake, a feint, a lie designed not to determine after-the-fact which books people actually buy, but rather to convince people which books they should buy.

You authors might not like it that you're outsold by the guy who wrote the latest Spongebob opus ... and I'm sure your publisher hates it ... but the word "bestseller" ought to mean "most books sold."


I watch the Amazon rankings of a number of books, many of which have not yet hit their publication date. I've recently noticed that, whereas before, many of these books would suddenly rank much better on the day of their release (because all of the pre-orders were shipped on that day, giving the book an incredible one-day sales number), now the before-pub and post-pub numbers are not much different. Perhaps Amazon (in addition to any other ranking changes) has begun including books' sales figures as soon as they are clicked as "sold," rather than when they actually ship? (Sounds like a cash vs. accrual accounting issue, almost!)


I would venture that ever since you decided not to offer full feed RSS you've lost a lot of potential readers who might have been inclined to purchase your book.

Definitely the RSS.


Perhaps you finally suceeded in selling a copy of Freakanomics to every man, woman, and child in America, thus, sending it's demand plummetting.

Pedro H. Carvalho

Did you check if this ranking drop happened to other books too? Couse I think this fact would pretty much confirm that the algoritm has changed. Otherwise, it can be just speculation...


I think Charles has got it right. If you check the list, most of the new kids books you mentioned are shipping in 2-5 weeks. What probably happened is word of a sale got out, people flocked to buy them and bought all the stock.

Perhaps the rate of sales figures prominently into the metric, so the rapid influx of Clifford and Curious George sales pushed them up the rankings. Once the deal is gone and the rankings have a chance to settle, I wouldn't be surprised to see the book land up near the top again.


I know exactly why Clifford and Curious George moved up. Yesterday Amazon started a children's book sale, with many prices starting as low as 70 cents. Many of the books were eligible for a "buy three, get one free" promotion. Supply and demand :)

Or it could be the RSS...