The simple answer is that it means different things to different people, but I’ll elaborate.

If you’re an Amazon.com customer looking at “Product Details” for a specific book, you might interpret the Amazon.com sales rank as a measure of the popularity of that book. The lower the Amazon.com sales rank number, the more popular the book is as determined by the number of recent sales of that title through the Amazon.com marketplace. This is true in general, but if you really want to know the popularity of a specific title, you should also consider the publication date and how long it has been trading on Amazon.com. For example, a book with an Amazon.com sales ranking of 100,000 that has only been released and traded on Amazon.com for a few days may be very popular in the near future and break into the top ten. That it’s only ranked 100,000th right now is a bit misleading because the time period over which this rank was determined isn’t long enough to measure true popularity.

For a bookseller, Amazon.com sales rank often correlates with how quickly a particular title can sell. A book with a very low number Amazon.com rank can be expected to sell out within minutes or hours of being listed, and a book with a 4,000,000 sales rank can take several years to sell. Booksellers frequently use the analogy when looking for books to add to their inventory. If a bookseller finds a book that has an online value of $10, an Amazon.com sales rank of 2,105,878, and is faced with buying it for $4, he is likely to pass, but if the sales rank were 45,017, is suddenly could be a book worth adding to your inventory.

When searching for books, I look at the Amazon.com sales range for higher-priced books because I don’t want to stack too much firewood on my shelves, but it’s certainly not the measurement I use for most book purchases. The scientist in me wanted to know more, so I compiled Amazon.com rating data for thousands of books I’ve sold, using the Amazon.com rating both at the time I bought them and at the time I bought them. I sold them, respectively. Being an engineer, I had to collect and torture data to see if there was any correlation between Amazon.com sales rank and the time it takes to sell a book. In fact, there is, at least in my experience, and curve fitting the data yielded a correlation coefficient of 0.93. The good fit of the data is particularly evident in a semi-logarithmic plot.

So, you see, there is a meaning to the Amazon.com sales rank.

For the book buyer it is a measure of the popularity of a book.

For the bookseller, it is a measure of how quickly a book can be sold and whether or not the book can be just firewood.

For the engineer, it’s just a bunch of data that can be tortured into a kind of confession.