Josh Bernoff of Forrester triggered quite a tempest online earlier this month with a report that said iPod buyers weren't buying many downloadable tunes from Apple, and their purchases were declining sharply. You can fault Josh for looking at sales sequentially instead of year-over-year; music sales are notoriously seasonal, after all. But I don't think you can argue with the central point about iPod owners not buying a whale of a lot of tracks from iTunes. Apple's own numbers showed that 1.5 billion tracks had been sold as of September to a world equipped with about 70 million iPods, which works out to a little more than 20 tracks per player. Even if you acknowledge that many iPod buyers have multiple players, the average at the time still amounted to only a couple of CDs' worth.
Focusing on iTunes sales, however, misses the larger point about how these devices are changing music consumption patterns. As anyone who got an iPod for Christmas (or even one of the less ballyhooed alternatives) will attest, having the ability to take a large quantity of music anywhere and everywhere alters how people feel about music. Unlike radio, an MP3 player is utterly personalized, not bounded by genre and minimally repetitive (or at least repetitive only to the degree that the user demands). So it seems inevitable that this new platform -- and as the legion of add-ons for the iPod demonstrate, these players are platforms, not just devices -- will stoke people's appetites for music.
That leaves only the question of how those appetites will be fed. The iTunes 99-cent-per-track model is an impossibly expensive way to load up even a 1 gigabyte iPod Shuffle, let alone a 30 or 80 gigabyte model. Most people find another way to stuff songs onto their players, whether it be legal (by ripping tracks from their CDs) or not (by downloading songs en masse from a file-sharing network). My guess is that iTunes will goose sales but won't dramatically change the amount of music people acquire, at least not in its current iteration. Remember, the average American doesn't buy much music -- 700 million CDs shipped in 2005 translates to about 4.5 per adult under 54. That's, what, 54 tracks per year? The current rate of 20 downloads per iPod may be significantly higher than the number bought by the average adult (about 4 this year and 2 last year, according to Nielsen SoundScan), but it doesn't come close to filling an MP3 player.
That's why I'm bullish on music subscription services like Napster and Rhapsody, as well as bulk MP3-buying options such as eMusic. The former have to do a better job persuading music fans to buy access to tracks instead of the tracks themselves, and they have to find a way to interoperate with iPods -- no mean feat, given Steve Jobs' refusal to do any licensing deals with iTunes competitors. The latter, meanwhile, won't be a mass-market option until they win licenses from the major labels, who have refused to sell songs as MP3s. But these approaches strike me as the ones best adapted to the new platform created by the tens of millions of iPods and iRivers in use today. Even if iTunes sales fell to zero overnight, the appetite for vastly more music will still be there.