One factoid immediately struck me in Chris Anderson's widely read October 2004 Wired magazine article, "The Long Tail." He put the percentage of "long tail"--that is, the relatively obscure books that aren't among the 100,000 or so stocked by a typical Barnes & Noble superstore--at 57%. In all my years writing about Amazon, I had never heard that number. Apparently for good reason: Turns out it was way too high. Amazon CEO Jeff Bezos later said it was closer to 20%--still reasonably impressive but not nearly the amazing number that everyone who read the article kept talking about.
Anderson has taken another, more rigorous look at Amazon's long tail, and concluded that it's more like maybe a quarter to a third of Amazon's overall sales. Again, pretty interesting, but not far off the traditional 80-20 rule. As a classic long-tail customer, I don't doubt the importance of the idea and the Internet's role in making the long tail profitable. But especially with just about every startup in the land boasting a PowerPoint slide about how they're enabling the long tail of [insert product or industry here], it's worth casting a more critical eye on the notion.