The Long Tail, Revisited

The Stalwart, one of the blogosphere's finest sources on markets and investing, raises a valid objection to the love-in over the long tail. Some evidence from NetFlix indicates that the most popular movies (i.e. the "head" of the sales graph) don't get rented as much because the physical supply of the hottest DVDs can't keep up with the demand, which flies against the theory behind the long tail. However, to be fair to Chris Anderson, he is pretty clear on the point that sales can't be all "tail" - a strong head is necessary as well, as he notes when discussing the failure of MP3.com:

By contrast, the success of Netflix, Amazon, and the commercial music services shows that you need both ends of the curve. Their huge libraries of less-mainstream fare set them apart, but hits still matter in attracting consumers in the first place. Great Long Tail businesses can then guide consumers further afield by following the contours of their likes and dislikes, easing their exploration of the unknown.

So while the long tail may not provide the revolution some have bestowed upon it, to say that its proponents are missing the importance of the hits is somewhat disingenuous. What would be (and will be) interesting to see is how the demand for hits and misses changes when it's all available through streaming, thereby cutting out the physical component and theoretically allowing for a more accurate representation of consumer preferences. Will the most popular movies prove to be even more popular, or will demand for titles further down the tail remain strong? Another aspect of the NetFlix situation that might be interesting is the role of price sensitivity on demand for movies, both mainstream and not. Would people be willing to pay more for the movies with higher demand? Would a lower per-unit price change the demand for less-popular movies? (Obviously, this kind of tinkering would be a profound alteration of NetFlix's business model, so it's primarily of academic interest).