May 7, 2008 1
The Long Tail
There’s a lot I hope to get to eventually in talking about the future of the media landscape, but a lot of it is predicated on Long Tail theory, so I suppose I should start with that. Named by Chris Anderson, an editor from Wired, and elaborated upon in his book (which I highly recommend), Long Tail theory became one of those models that, encapsulated by a cute phrase, were fashionably thrown around for a while, in the same category of ideas as black swan events and tipping points. Long Tail is directly applicable to the current models of distribution in some industries, and so describes a very real and important trend in 21st century economics.
Earlier this year, iTunes passed Wal*Mart as the US’s largest distributer of music (at least, some people think so – Apple is quite secretive about iTunes’ actual sales figures). This event validates Anderson’s theory – iTunes follows the Long Tail model; Wal*Mart, with its finite shelf space, does not. iTunes’ ability to outsell the nation’s largest retailer is sturdy evidence that Long Tail is the wave of the future.
But I still haven’t described exactly what Long Tail is. Basically, it comes down to the pattern of popularity of given items for sale, handily illustrated in this graph (which I lifted from Slate.com):
In a traditional model, a handful of wildly popular items (“hits”) accounts for the bulk of sales (that would be the head, at the left of the graph, obviously), and the rest of the items sell a few each, but even together don’t generate as much profit as the hits. The shorthand is 80-20: 80% of profits are generated off 20% of the titles in stock.
This makes sense when a retailer, like Wal*Mart, has limited space, and keeps only the top selling items in stock. But with an model like iTunes, one which can keep a virtually infinite number of titles for practically nothing, the figures are reversed. The hits generate a respectable chunk of profit, but it’s the list of titles that sell only a thousand or a dozen or even fewer copies a year, but titles which number in the millions, that cumulatively generate the greater profit.
If that doesn’t make sense, try visiting the We Feel Fine website I linked to a few posts down, and clicking the “Mounds” link at the bottom left. This visualisation shows all the terminations to the phrase “I feel ______” gleaned from the Internet by lumping them into wobbly mounds. 128,155 people finished the phrase by saying “better,” the most popular answer. But by the time you scroll to “satisfied,” only 1418 results appear. And that’s less than a tenth of the way down the tail. Keep scrolling right … only 20 people feel “vexed,” but the number of other responses with only 20 or so results is staggering. In fact, the tail cuts off at mounds of 20, so perhaps it could continue indefinitely. Clearly, the vast majority of the total volume of all the mounds exists in the tail.
As Anderson explains, when carrying inventory is virtually free, it just makes economic sense to keep as many as you can. It costs a fraction of a cent to keep an obscure .mp3 on a server for a year; even if it only sells once a year, at $0.99 a pop, you’ve just made roughly $0.99 profit, and pretty much effortlessly. The benefit for the retailer is obvious. The benefit for the consumer is just as obvious: unlimited selection means the opportunity to indulge even the most arcane tastes.
This doesn’t work in all industries, obviously, but digital media is particularly well suited to carrying unlimited stock. Netflix, while still carrying physical inventory, represents a successful transition between two models, and is a great example of Long Tail at work; cinemas, with their dozen screens, are not. Amazon is another business based on the Long Tail model: their stock is much, much larger than any bookstore can carry (but, it’s important to remember, is limited to only those books in print), and is distributed by mail, as opposed to through retail outlets. The Kindle, Amazon’s solution to both the problem of keeping books in print and of distributing them by mail, has yet to catch on, but don’t be surprised if some iteration of it someday replaces books made from trees. For that matter, Netflix is already offering subscribers the ability to download movies for viewing on home computers – which not only eliminates wait times, it also dramatically reduces delivery costs. Digital books and downloadable movies both represent the fullest potential of digital media to extend to the very tip of the tail.
The ramifications of Long Tail are far reaching and shouldn’t be underestimated; not only does it change the music we listen to, the movies and TV shows we watch, and the books we read, but eventually, it will reshape the way we navigate the media landscape. With endless choice, the process by which we filter the signal from the noise, the wheat from the chaff, will become more sophisticated, eventually evolving into an entirely new manifestation of culture unto itself.