Once Again, the Long Tail Refuses to Be Buried

200px-long_tailsvg Wired editor Chris Anderson’s theories about the Long Tail have been the source of considerable controversy almost since the day his first Wired magazine piece on the topic was published in 2004. The initial criticisms of his thesis centered on whether there was such a thing as a “long tail” at all — in other words, whether digital distribution of music and other forms of content have allowed little-known songs, movies, and so on to prosper where they might otherwise have been ignored. Later attacks, however, have focused on how the Long Tail theory functions in certain markets, and whether or not the existence of such an effect actually helps anyone in those markets create a workable business model.
The most recent criticisms came a few days ago, at a mobile telecom conference in London, where an economist named Will Page — who works for the MCPS-PRS Alliance, a British copyright licensing-fee collection agency — spoke about research he conducted into music-buying behavior. The results of this research, Page said, didn’t conform to the “power law” distribution described by Anderson’s theory, but instead followed a more common “log normal” distribution (if you really need to find out more about a topic only a statistician could love, you can check here and here).
The bottom line is that Page reportedly argued the data didn’t support the existence of a Long Tail for music buying (a claim that The Register pumped up into a post about how the entire concept is flawed, and how this is “bad news for Californian technology utopians”). Anderson, for his part, says the data appears to have come from research into mobile music-buying patterns — since mobile music provider Mblox was a partner in the study — and that he has already admitted mobile behavior is subject to different effects (music-industry theorist Gerd Leonhard makes some excellent points about other reasons why we shouldn’t necessarily believe the numbers, as does Yankee Group analyst Benoit Felten).
Why does this debate matter? Because Anderson’s theory suggests that content providers should expand their catalogs of music and movies to include more obscure titles, as a way of appealing to consumers with broader, Long-Tail type interests. By extension, the theory also suggests that musicians, writers and directors who are outside the mainstream might be able to pursue their creative dreams and still make a living. If there’s no Long Tail, then all bets could be off, and the Top 40 mentality could once again rule over content-related industries.
There have been previous attempts to bury the Long Tail, including one launched by a former research partner of Anderson’s, Anita Elberse, who wrote a piece for the Harvard Business Review about flaws in the theory based on data she collected. In that case, the former Wired editor made a fairly convincing argument that, far from torpedoing his conclusions, much of the data actually helped enhance the theory. Page’s study may not do that, but it is a long way from a smoking gun.
Contrary to what some might think, Chris Anderson didn’t invent the idea of the “long tail” — similar theories about the effect of diminishing production and distribution costs on digital media were
being discussed at least a decade before he wrote his Wired article, and many of the central concepts have been around since the mid-1940s. Whatever its flaws, it is still a powerful way of expressing the
changes the web has wrought in content-related markets of all kinds. Whether content producers, distributors and creators want to adapt or not is a different question.