Here at the D Conference, Walt Mossberg just asked Jerry Yang and David Filo about Steve Jobs' assertion on the same stage two days ago that Yahoo!'s new music service was priced below cost, and that they had a betting pool at Apple about how long it would be till Yahoo! raised their price. Steve's bet: 5 months. Jerry's reply: "Well, we have a pool at Yahoo! about how long it is before Apple introduces a subscription service. And I think David's down for 5 weeks!" That being said, Jerry did admit that perhaps their offering was an "introductory price." [tim at O'Reilly]
I've written at length about the fact that people should not judge a subscription service like Napster's by its current pricing. Napster etc. don't serve the needs of people who don't buy much music; it's too expensive compared to what they would spend otherwise. This has been the source of a lot of criticism of the subscription model vs. the purchase-a-track-at-a-time model.
I have argued that that criticism is barking up the wrong tree because one can't judge the future of subscription services by the initial, exploratory, yet-to-be-fully-formed offerings we are now seeing. I argued that subscriptions for such consumers would come down in price.
Yahoo's cheap pricing isn't the arrival of a solution for such users. It can't be, because it's still one-size-fits-all service. Since it is offering the same price to all users, even those whose use will be very heavy, Steve Jobs is probably right that it is unsustainable, and Jerry Yang's "perhaps" should probably considered to be more than a "perhaps."
The solution will come when a vendor offers different pricing' structures depending on the subscriber's needs. Those who listen to more music should pay more for their monthly subscription. That would ideally require portable music players to report play counts back to the subscription service; something that I don't believe is supported now.
But I'll be surprised if it is not in our future.
I should also note that one can imagine a purchase-a-track-at-a-time model that would also fit the bill. It would provide most of the advantages of subscription models but allow the user to keep the music if he stopped paying, obviating Steve Job's criticism of the subscription model: "when [you] stop paying all the music goes away".
The solution would be an infrastructure wherein a) consumers paid for songs over time, so that the amount they pay for a song is proportional to the amount of benefit they derive from it, roughly analogous to paying a mortgage on a house and having the option of selling it early. But with a more linear schedule for paying off the principal. And b) a really transparent and automatic mechanism for users to trade, buy, and sell songs from each other; this should happen relatively invisibly to the user, with the result that the user experience is almost the same as it would be for a subscription service, even though it's really a purchase-a-track-at-a-time model.
I could write a lot more about that, and possibly will, but I just don't have time now. Moreover, I currently think it's likely that the simpler solution -- a straight subscription service with a tiered pricing model -- will win out.
Comments