Two British researchers have published a study rebuking the “Long Tail” idea that in the digital economy, you can survive by selling less copies of each item, yet more variety of total items. Instead, they found that “more than 10 million of the 13 million tracks available on the internet failed to find a single buyer last year.” The British researchers also found that “for the online singles market, 80 per cent of all revenue came from around 52,000 tracks” and for albums, 85% did not sell one lousy copy all year. “I think people believed in a fat, fertile long tail because they wanted it to be true,” one of the researchers said. The idea’s original author, Chris Anderson, said that the new study consisted of only one data set and so proved nothing, although a data set consisting of all the music available for sale on the Internet would seem to be a reasonably significant one. But how about those people bullish on online music specifically? Surely they must be troubled by this news, right?
Nope! Hypebot’s reaction:
Does this meanthat we should all be back to chasing hits? No, but if correct, the study should remind us that making music available for sale is the beginning and not the end.
Ah, that must be the problem: no one bought those 10 million tracks because they just weren’t marketed right. If only they had some sort of company that could market and promote them without the artists having to pay them money up front! Or, I guess, if only the artists weren’t so lazy making music and touring and working day jobs when they should be friending people on MySpace.
This should be a huge story for everyone that talks about digital music, because this is exactly the kind of evidence that’s been missing from the debate. We know what major labels do wrong, and we know what people think should be done differently. But we don’t know whether that works. This makes sense, in a way; it’s easy for a blog to reprint a press release for a new company whose model conforms to your ideological expectations, but it’s hard to do the legwork to see if those companies are actually successful months or years down the line. It’s easier just to assume that what you think will work will work, or if it doesn’t, to assume this is due to the big evil major labels manipulating things. But this is not how the world works. If the new model is better, it should be beating the supposedly failed establishment, and if we care about music, we should be asking ourselves honestly if the new model is actually working by checking back on things and assessing their success. If they’re not selling music, then they’re not working. Period.
Much of the new digital economy seems to be new ways to distribute music, not new ways to make music. Widgets that stream, unlimited download services, pop freeganism, yes; but the content is supposed to make itself magically. In creating our new model, we’ve taken the myth for the truth, assumed that the heroic lone artist is how music actually get made, when of course, it’s not. A single artist can always make music on their own, for free. But that’s not how most music gets made, and without an actual revenue stream, which can only come from selling music (even if the “long tail” exists, at that end of it no one would be able to actually turn a profit from touring), then all of that is lost.
Long Tail theory contradicted as study reveals 10m digital music tracks unsold [The Times via The Daily Swarm]
[Photo of a long-tailed tit via Sergei Yeliseev]
One day people are going to look around and ask themselves what happened to all the content. At that point, things will start getting better. Until then, things are going to get a whole lot worse.
This is not much different from the old model, either, in terms of sales. A couple of years ago Billboard did a wrap up for 2006 that showed that only a third of albums released sold more than one hundred copies (and I’ll bet a large number of those sold none at all), and most of the remaining third sold less than 10,000. The delivery system may be changing, but the realities of the market don’t seem to have changed much at all.
This is a killer post, Mike. The last two grafs, especially.
Here’s my response/continuation, if you’re interested:
[www.thecontrarianmedia.com]
I don’t really find this troubling or encouraging from either standpoint. First of all, it seemed like Anderson was less dismissive of the results in the article from the Times. Instead he seemed to acknowledge that the data provided showed that the long tail model did not apply, but was hesitant to say that it meant that the long tail model COULDN’T apply in general (that is, for business as a whole, not just the music business). Additionally, it also seemed like he would need to read the entire published report, a not unreasonable response.
More importantly though, I never considered the idea of the Long Tail as immediately game changing as everyone seemed to think it meant. It’s an attractive idea but it’s only possible given the relatively new model provided by Amazon. The datasets gathered by Anderson seemed to prove his initial theory but it seemed naive to me to think that it meant the model was already happening, despite the hype that surrounded that idea.
I don’t know. In the end, this is how data analysis and research works: contradiction after contradiction. If Page and Bud’s analysis prove nothing, it’s because Anderson’s research proves nothing and vice-versa. They’re just ideas, meant to be discussed and complained about.
Studies, schmudies. Long tail, shmong tail. I still say ‘yay for content, boo for major labels as filters’. I don’t care how much shit there is out there that no one buys – I can still sift through and find better music than the turds at record companies.
Nice post Mike. It prompted me to actually make a post of my own.
[30frames.blogspot.com]
A counterpoint from John Brissenden:
[johnbrissenden.tumblr.com]
yeah spot on really. its almost as if people would rather this research hadn’t been done, as it challenges the prevailing ‘alterna-wisdom’.
@Mike Barthel: It seems like point number one contradicts point number two.
“The report hasn’t even been published so we can’t draw any conclusions about its veracity, but while we wait why not call the authors’ credibility into question?”
Why is this surprising at all? What surprised me is that ONLY 10 million of the available singles didn’t sell at all. Almost every work of any kind is bad, or at least exceedingly mediocre–painting, sculpture, dance, television, fiction, and, yes music, even if it’s distributed online and without DRM and without a major label deal. Fighting overly-restrictive copyright policy and the out of touch label exec dinosaurs is valiant, and I mean that, but we are all joking ourselves if we think The New Age of Media Distribution will simultaneously usher in some second renaissance.