"The blockbuster effect has been even more striking on the digital platforms that were supposed to demonstrate the benefits of the long tail. On iTunes or Amazon, the marginal cost of “stocking” another item is essentially zero, so supply has grown. But the rewards of this model have become increasingly skewed towards the hits. Anita Elberse, of the Harvard Business School, working with data from Nielsen, notes that in 2007, 91% of the 3.9m different music tracks sold in America notched up fewer than 100 sales, and 24% only one each. Just 36 best-selling tracks accounted for 7% of all sales. By last year the tail had become yet longer but even thinner: of 8.7m different tracks that sold at least one copy, 96% sold fewer than 100 copies and 40%—3.5m songs—were purchased just once. And that does not include the many songs on offer that have never sold a single copy. Spotify said in 2013 that of its 20m-strong song catalogue at the time, 80% had been played—in other words, the remaining 4m songs had generated no interest at all."
"The blockbuster effect has been even more striking on the digital platforms that were supposed to demonstrate the benefits of the long tail. On iTunes or Amazon, the marginal cost of “stocking” another item is essentially zero, so supply has grown. But the rewards of this model have become increasingly skewed towards the hits. Anita Elberse, of the Harvard Business School, working with data from Nielsen, notes that in 2007, 91% of the 3.9m different music tracks sold in America notched up fewer than 100 sales, and 24% only one each. Just 36 best-selling tracks accounted for 7% of all sales. By last year the tail had become yet longer but even thinner: of 8.7m different tracks that sold at least one copy, 96% sold fewer than 100 copies and 40%—3.5m songs—were purchased just once. And that does not include the many songs on offer that have never sold a single copy. Spotify said in 2013 that of its 20m-strong song catalogue at the time, 80% had been played—in other words, the remaining 4m songs had generated no interest at all."
http://www.economist.com/news/special-report/21716467-techno...