Long Tail Falls Flat? HBR Refutes Anderson

By now, most of us have heard Wired editor Chris Anderson's theory of the long tail. It boils down to the fact that search and other technologies are making it easier for consumers to find more obscure items that they really want to buy, rather than heavily marketed items that companies push on us. Since technology can also enable businesses to offer more (usually digitized) items at a lower cost, the business wave of the future is making as many options as possible available to as many people as possible--selling a few copies of hundreds of items rather than hundreds of copies of a few items, and shattering the traditional "blockbuster" mode of production in which one or two big-selling products fund the rest of a company's output.

Harvard Business School professor Anita Elberse has a new article for Harvard Business Review challenging Anderson's theory ("I got your tail!"). She says it's "highly disputable that much money can be made in the tail," and cites statistics showing that even consumers who are interested in the long tail--and there aren't enough of them to be significantly profitable--consume popular, blockbuster-type products in addition to obscure ones. Consequently, she concludes, companies should keep pushing a few big-name products over all others in order to make money, and bundle the big names with more obscure stuff to try and broaden consumers' tastes.

Citing William McPhee's theory of exposure, Elberse distinguishes "light" or infrequent consumers from "heavy" or frequent consumers, and describes how the infrequent consumers focus on the popular products. More avid consumers also buy popular products, and--because of their high demand for product--expand into more obscure items as well. In other words, no one is buying only obscure stuff. Additionally, more popular products are consistently rated higher than more obscure ones, suggesting that even those who consume "underground" items don't necessarily like them that much.

Elberse uses some data from Rhapsody's online music service, but much of her argument hinges on information from the online movie rental service Quickflix. Quickflix seems to be the Australian version of Netflix, except with about a third of the selection (32,000 vs. over 100,000 titles), making it perhaps a poor choice for refuting a long tail theory based on the availability of many options to the consumer. In addition to using limited data, Elberse also conflates "superior" products with "popular" products--sensible from a business perspective, perhaps, but overly dismissive of the amount of enjoyment that people can derive from more obscure products.

Based on the customer survey data from Quickflix, Elberse concludes: "Hit products remain dominant, even among consumers who venture deep into the tail. Hit products are also liked better than obscure products." Because "the tail is long and flat... content providers will find it hard to profit much from it."

Elberse's assertions come at an interesting point in our business evolution. As we see it, Anderson's argument is not necessarily so much that the long tail is already hugely profitable, but that the companies who figure out how best to exploit the long tail effect will be poised to make bunches of money when the long tail really takes off. Elberse is focused on data about what's happening now, just a few years into the online sales revolution (can you believe the iTunes Store is only five years old?), while Anderson is focused on what's going to happen if technology is used to its potential. The nature of business is bound to continue changing drastically as technology develops, and any company remaining bound to one sales strategy is likely to fail.

Chris Anderson himself addresses Elberse's arguments on his blog. He says the differences between his and Elberse's ideas result from a "subtle difference in the way we define the Long Tail, especially in the definitions of 'head' and 'tail.'" Anderson asserts that "'Head' is the selection available in the largest bricks-and-mortar retailer in the market," while "'Tail' is everything else, most of which is only available online." Elberse is coming at the issue from a perspective that "head" is popular product and "tail" is obscure, while Anderson is defining the issue more in terms of availability. The two may sometimes be meaningfully related, but not always.

Anderson also questions Elberse's reasons for extending the data on Quickflix customer ratings to other products. This is the only data suggesting that consumers appreciate more obscure titles less, yet Elberse uses it frequently to suggest that the obscure is not truly desired by consumers. Having seen plenty of crappy independent films, we can relate to the idea that the obscure is not always amazing--but we've probably seen even more crappy blockbusters than crappy independent films. And Rotten Tomatoes is an easy place to go for evidence that critics, at least, think many blockbusters suck.

Commenters on Anderson's blog criticize Elberse for focusing on percentages rather than numbers, noting that the top 1% of any particular market is now much larger than ever before, thanks to tail-enabling technology. This was central to Anderson's original thesis, and seems largely ignored by Elberse.

Regardless of your perspective on the long tail, these thoughts make interesting reading for anyone concerned with the future of business.

Images from from Flickr users Dan Taylor and Chris Gin