What kind of academic books sell well and which ones do not? This depends on how you define academic, book and well. Even so, I have one authors data points to share. My advisor Harry Lewis has written several books of very different types and was kind enough to share his numbers and some comments with me.
- Unsolvable Clases of Quantificational Formulas 1979. A monograph on an extremely specialized topic. He says I don't know how many it sold, but if it sold 1000 it means my mother bought 500.
- Elements of the Theory of Computation (with Papadimitriou). Textbook in Automata Theory. 1981. First edition sold 38,000, second sold 19,000. I'm surprised how well it sold- its a fine book, but there are lots of books on automata theory out there. Of course, there were fewer in 1981. Also, the used book market was not as efficient then as it is now.
- An Introduction to Computer Programming and Data Structures using MACRO-11 1981. 4000 copies. Specialized, not surprising. And clearly irrelevant now.
- Data Structures and their algorithms (with Larry Denenberg). 1991. 5000 copies. He says Disapointing, we worked very hard on that one. By 1991 there were already quite a few books on the market. And CLR, which is both broader and superb, began to dominate the market.
- Excellence Without a Soul: How a great university forgot Education. 2007. Sold 13,000. Very happy with that- books like this usually sell about 3000-5000. AND there is a paperback version coming out now.
- Blown to Bits: Your Life, Libery, and Happiness After the Digital Explostion (with Hal Abelson and Ken Ledeen) He says: Also a trade book. Just appeared June 15, but seems to have sold 3000 in the first two weeks. People like it, it is even sold in some airports, but there haven't been any newspapers reviews and there have been few online reviews too, so its been hard to attract a lot of notice. I'll review it in my column and on my blog soon. Need to finish it first!