Chopping the Long Tail down to size
Wired's advice could seriously damage your business
Exclusive The most comprehensive empirical study of digital music sales ever conducted has some bad news for Californian technology utopians. Since 2004, WiReD magazine editor Chris Anderson has been hawking his "Long Tail" proposition around the world: blockbusters will matter less, and businesses will "sell less of more". The graph has become iconic - a kind of 'Hockey Stick' for Web 2.0 - with the author applying his message to many different business sectors. Alas, following the WiReD Way of Business as a matter of faith could be catastrophic for your business and investment decisions.
Examining tens of millions of transactions from a large digital music provider, economist Will Page with Mblox founder Andrew Bud and Page's colleague Gary Eggleton, looked to see how large and valuable the "Tail" of digital music may be. They produced a spreadsheet with 1.5 million rows - so large, in fact, that it required a special upgrade to their Excel software (and more RAM) - and the three revealed their work at the Telco 2.0 conference this week.
They discovered that instead of following a Pareto or "power law" curve, as Anderson suggested, digital song sales follow a classic Log Normal distribution. 80 per cent of the digital inventory sold no copies at all - and the 'head' was far more concentrated than the economists expected.
"Is the 'future of business' really selling more of less?" asks Page. "Absolutely not. If you had Top of the Pops now, you'd feature the Top 14, not Top 40."
Anderson bet that the orange portion - the "Tail" - has more value than the red portion - the "Head". But it doesn't.
As Andrew Bud explains:
"The Long Tail's argument is that the pattern of consumption for media is bent out of shape by the limits of the shops selling them. Digital media lets the nature of people's demand flow free. Well, we now know what the shape of that demand curve looks like."
Bud told the conference that the basic shape of consumer demand for digital music clearly fits the Log Normal distribution, "with eye-watering accuracy". That's no surprise, he says, because so many sales curves he's seen over the past ten years follow this distribution.
"Now we've seen what happens when tens of millions of choices are thrown in the air and people can go pick them up. What was astounding was the degree of inequality between the head and the tail - by a factor of three. It's specifically the Log Normal shape that leads to a rather poverty stricken Tail.
"There are Tails where the Tail lives as a kind of welfare state. Not this one. You starve in this Tail."
Brown's 1956 lognormal curve fits digital sales data much better than "The Long Tail"
This really isn't the upbeat fairy tale message Anderson has spent four years selling on the conference circuit. However, as he took his "message" to Davos and beyond, the Long Tail has gradually developed into a 'Policy Based Evidence Making'. Having convinced himself of the truth of his hypothesis by looking at one US music service, Anderson widened his search for facts that might fit his theory. But he didn't examine the numbers closely or critically enough, say the economists.
“Now we've seen what happens when tens of millions of choices are thrown in the air and land on the floor.”
"You need to consider much more than just some flimsy volume-based Rhapsody data if you're going to say the world's changed," says Page. "For instance, understanding value both in terms of retail spend and then marginal profitability to the artist and songwriter would have been a logical extension"
In another surprise, 80 per cent of the revenue came from 52,000 songs. What's eye-catching about the number? Well, the typical inventory of a conventional high street record store was around 4,000 CDs. Or ... around 52,000 songs.
Sponsored: Hyper-scale data management