In a post entitled New Study: Internet Radio Plays 800,000+ Stiffs, we suggested that advertising supported Internet radio could not succeed if it chose to play songs that the majority of listeners didn’t want to hear. We noted:
Across the entire (music) spectrum, there are only so many songs worth playing (if your goal is to have a measurable audience). That means there are only so many "marketable" artists that listeners want to hear.
We were taken to task for our comments. Internet radio supporters argued that there is plenty of room for long-tail Internet radio, stations that play songs that the majority of listeners don’t want to hear.
A Billboard analysis of Nielsen Soundscan data raises further doubt about the economic viability of long-tail radio. This is from Nielsen's blog:
For most people, Chris Anderson's 2006 book The Long Tail marked a new way of thinking about selling goods on the Internet. Being free of the physical limits of shelf space, he predicted, would alter what people bought. For music, this would mean the most popular music titles would become less popular as consumers were able to tap into vast online catalogs. In most corners of the business world, and especially in the music industry, The Long Tail was controversial. Would consumers actually start to ignore the hits?
A Billboard analysis of Nielsen SoundScan data going back to 2004 shows Anderson wasn't correct on all points. Hit digital albums have lost market share to far less popular titles. But hit digital tracks have gained market share over the years. The top 200 tracks accounted for 14.5% of sales in 2004 and rose to 15.8% in 2005, 17.1% in 2006 and 2007 and 17.2% in 2008. Through October 25, 2009, the top 200 tracks' share stood at 18.7%.
This is not the first analysis to find that online music’s long-tail appears to be a myth. The U.K.’s Guardian noted the myth of music’s long tail in 2008. Recently they wrote:
The study discovered that, of the tens of millions of tracks available for sale on the web, 80% sold no copies at all - and that 80% of the money spent on the 20% that did sell went on just 52,000 songs. As Andrew Orlowski pointed out in his excellent Register article on the subject, the typical inventory of a conventional high-street record store is around 4,000 CDs, or 52,000 songs.
Rather than create a longer tail, digital downloads appear to be shrinking the tail. This in turn suggests that the Internet may be expanding access to diverse music, but the majority of listeners are rejecting this diversity and instead focusing on an even smaller portion of music.
It would be ironic if crowd-sourcing actually tightened radio stations rather than opening them up.