The recording industry is in trouble. While the transition from analog records to digital CDs was a profitable one for Capitol, Universal, Sony/BMG, Warner Brothers, and the other big labels in the 1980s and early 1990s, the ongoing shift from physical media sales to digital delivery over the last decade has been a disaster. The signs of wrenching change are obvious: Shuttered CD stores, counterproductive DRM schemes, and a heavy-handed RIAA legal campaign against alleged infringers.
In recent years, some in the industry have recognized earlier missteps. We are now seeing a variety of business models to support legitimate digital music downloads. But they look unrealistic to me -- does anyone really think consumers will jump on yet another variation of online music subscriptions, crafted by Columbia's Rick Rubin or anyone else?
There are also per-song and per-album purchasing schemes. Are they the solution? Maybe not. Two recent, high-profile experiments involving Radiohead and Trent Reznor look iffy, and the labels can't seem to make up their minds about the iTunes/iPod/iPhone troika.
And here's one of the big problems with the per-song/per-album schemes: Unless you are already familiar with the music, how can you be sure it's worth the amount of money you pay (e.g., 99 cents per iTunes song) or even the prices that you suppose the music is worth at the time of purchase? Nothing is more frustrating than slapping down cash for something you think is going to be great, and then finding out that it's actually a dud.
This was my experience with the new Radiohead and Trent Reznor album downloads. I am a fan of both, and paid for each album -- $5 for the Trent Reznor/Saul Williams set, and over $9 for Radiohead's In Rainbows. Both su ... I mean, both were quite disappointing. And there weren't any good feelings associated with supporting two independent acts that I have liked for a long time. I just felt ripped off.
But I would have felt less ripped off had the price been cheaper. Say, $2 for each album. It also occured to me that I would have paid a lot more for earlier albums that they'd released. For instance, I bought Radiohead's OK Computer on CD when it was released ten years ago, and I have played it hundreds of times since then. Its value to me is significantly more than the $13 I shelled out for the disc in 1997.
And this made me think: Why doesn't pricing reflect the true value of the music to the individual consumer, rather than the unit value assigned by the band, the label, or the distributor? In the days of LPs and CDs, such a proposition would have been impossible. How would you know how much a consumer liked a disc or the songs on it, unless you asked them individually? But in the digital age, determining whether or not a consumer likes the music he or she has bought is easy. A measure already exists -- namely, the number of times someone plays a song -- and there is already a system to capture that metric, thanks to the software that people use to play mp3s and other music files. For instance, iTunes already counts the numbers of times I play each song. It's safe to assume that a new song I have played 26 times in the past two months is more valuable to me than a new song I have played just once.
So, instead of paying 99 cents for a song through the iTunes store, what if I agreed to pay a range of 30 cents to $1.50, and then charge my credit card one month later, based on the number of times I have played the song on my computer and in my iPod, relative to the total number of song plays I have initiated in the past four weeks? If I listen to it every day for an entire month, I would gladly pay $1.50. If I listen to it two times in the first week, and then never touch it again, then take my 30 cents. I don't feel so bad about shelling out 30 cents for a substandard song, and I don't mind paying $1.50 for a song that sings to my soul. This model would make buying online music much easier for me -- there wouldn't be any lingering doubts about whether or not to purchase something, or buyer's remorse.
Note that this is not the same as the scaled pricing model described by Chris Anderson on his Long Tail blog. The model I have described actually determines price based on a measure of like/dislike for individual listeners. Would this work for you, or are the existing pricing schemes for music downloads the way to go?