Amazon and B&N suggest wildly different books

When I look at my book pages on Amazon and B&N, I find an enormous gap. B&N tells customers that people who buy The Numerati also purchase extremely heady books about math (which are way beyond me). They point them to Conceptual Spaces: The Geometry of Thought, and Foundations of Measurement, Vol II, Geometrical, Threshhold and Probabilistic Representations!

Now, maybe there are people who bought my book and also bought those. I should alert them, and others, that anyone capable of reading those books will learn next to nothing about math in mine. It’s written for a general interest audience. There’s not one formula in the book, not one Greek letter. I issue a stark warning in my book blog. (My fear is that they’ll buy my book expecting sophisticated math, be disappointed not to find any, and then trash it to their friends and on blogs.) The Amazon recommendations are entirely different. They point to popular books, like Dan Ariely’s Predictably Irrational Dan Roam’s The Back of the Napkin.

What’s the difference between the two approaches? I can only guess. Let’s assume that they’re being honest, and that buyers of my book actually purchase the ones they list. But which ones best represent the readership? It would seem to me that B&N has a set of key words associated with mine (mathematics, calculus, etc) and it links mine to books with similar tags. Amazon, by contrast, appears to pick the books that sell the best. (Irrational and Napkin are both best sellers.) It’s not a sophisticated formula. But it makes more sense for me—and for Amazon.

Before it's here, it's on the Bloomberg Terminal.