At his Foner Books web site, Morris Rosenthal has lots of good advice on print on demand (POD) publishing. In fact, Foner Books has used POD publishing for its own titles. One of its current titles—Print-on-Demand Book Publishing—is all about how to do POD publishing, including selling on Amazon. Morris Rosenthal also talks about estimating the sales figures from the Amazon sales rank; that's part I want to talk about today. He provides an analysis—Estimating How Many Books Sold by Amazon Rank—that includes a plot of sales per week against the Amazon sales rank, based on data he gathered during November 2004. The plot is on a logarithmic scale and it's shows linearly decreasing sales with increasing rank. For example, a title with a sales rank of 1000 sells 90 copies a week whereas a rank of 100,000 translates to 3 copies a week. Although it's probably not a good idea to extrapolate the data into the lower sales ranks, I was tempted to get to an equation for the sales per week (S) versus rank (R) and it's roughly the following power law equation with a power of 0.74 (this is the "shape parameter"):
Then I plotted the function and as expected it's the usual power law curve except that in this case, the plot shows an empirical plot of copies sold on Amazon per week versus the Amazon sales rank:
By the way, there is an economic analysis of online book sales by Judith Chevalier and Austan Goolsbee—Measuring Prices and Price Competition Online: Amazon.com and BarnesandNoble.com—that also includes an estimation of rank from online sales. Their formula expresses rank as a function of sales, but if I were to use their formula and rewrite it as sales expressed in terms of rank, the shape parameter would be 0.83 instead of 0.74. This causes the sales to drop off faster as the rank increases. Both of these examples gives you an idea of the power law distribution at work for online book sales. By the way, the term "long tail" comes from the power law distribution's tail that shows small, but non-zero sales figures for books way down in popularity. The impact of the long tail is that the sales of a large number of high-rank (low demand) books could add up to a sizable amount, so they're worth selling for online stores such as Amazon that can keep all those titles in stock.
Tags: ebooks epublishing trends long tail longtail POD print-on-demand
It is a really good go through for me, Must admit that you are one of the best bloggers I ever saw.Thanks for posting this informative write-up.
Posted by: Jordan AJF 8 | August 02, 2010 at 02:34 AM