A new study of digital music sales by Will Page, chief economist of the MCPS-PRS Alliance, the not-for-profit royalty collection society, challenges the idea that niche markets are one of the most important economic models of the internet -- a theory known as the "long tail".
Page carried out the economic modelling for Radiohead's 'In Rainbow' album, which was released free on the internet.
The idea of the long tail first came to prominence in a book called 'The Long Tail' by Chris Anderson, published in 2006.
It theorised that the internet economy would shift from a relatively small number of mainstream products at the head of the demand curve to a long tail or "huge number of niches in the tail".
The online retailer Amazon is often cited as an example of a business that generates huge revenues based on the long tail economy.
The study found that online sales success still relies on big hits, revealing that of the online singles market, 80% of all revenue came from around 52,000 tracks.
In the albums market, the figures were even starker, with only 173,000 albums purchased online from the 1.23m available to buy.
The data and its sources in Page's study have not yet been published, but 'The Long Tail' author Anderson has already refuted its claims in an interview with The Times.
Anderson told the newspaper: "I respect what Will's done and have no doubt that he has indeed found a dataset where it doesn't work, but I'm not sure you can conclude much, if anything, beyond that."
Comments
Two immediate flaws spring to mind:
1. The Long Tail didn't say that there wouldn't be a 'short head' of mass hits that would continue.
2. A study of legal downloads is going to miss the huge amount of illegal downloading, free streaming services etc. With the current disuption to the music industry, perhaps the legal portion is made up of the hits, and it's in the free/illegal sections the long tail really exists?
Dan, nice points, totally agree!
Never mind the flaws in the interpretation- take a second look at those numbers.
The "old" idea is that the "head" is where the business is- so by having all the "hits" in stock, record shops were doing good business.
The "new" idea is that with more choice, there is more demand- that there is actually more business is selling everything outside the "head" \(ie. the Long Tail) than selling everything in it.
Although the numbers are meaningless without a bit more of a context, lets say for arguments sake that it's been counting since the iTunes Music Store launched in 2003 and made legal downloads a realistic option 5 years ago.
And, for the sake of simplicity, lets say those 52,000 tracks were released evenly over the course of those 5 years.
That would mean that 80% of singles sales came from 200 singles that were released every week. Now the idea of a record shop with 200 singles for sale in a single week is pretty much unthinkable- let alone 200 different singles every single week. \(And remember, record shops usually don't sell old singles- if they're out of the charts, they're off the shelves. Not so with digital sales...)
The same calculation for the albums would mean that 665 albums every week were released and sold. Again, have you ever seen a record shop that had over 600 different new album releases on their shelves every single week?
Of course, the assumption that those were evenly distributed is nonsense- in actual fact there would have been more releases \(and more sales) as time went on and the digital market grew, so the number of singles/albums released right now that are getting sales would be significantly higher than those numbers.
So, the Long Tail theory says that the future of business is "selling less of more"... and this is supposed to DISprove it? Seems to me that it validates the theory perfectly well... \(that is, if you understand the theory to mean something other than "increased choice magically creates a market for everything being sold"...)
It looks like the old 80:20 rule still applies. Which means that there is a long-tail worth 20% of revenue, predicted to represent 80% of all available music. It's still a long-tail though.
Brilliant Scott, it's a very good argument. To push it even further: it would be great to compare the number of titles \(albums and singles) responsible for 80% of sales BEFORE e-commerce really took off. Whatever it is, I'm darned sure that it is much more narrowly spread than it is now. As you said, this actually proves the theory, not the other way around.
As someone who has managed big music and ents websites over the years, albeit not pure ecom sites like Amazon, I've always struggled to see the long-tail thesis working in practice. If I had to sum up the reasons in one word, that word would be 'marketing'. Even with no physical barriers between music buyer and artist, the buyer still needs to discover the artist and be able to judge whether or not to buy their music. From the publisher's perspective, which I guess is what I represent in this debate, it's still all about hits. What has changed is that there are more genres and therefore more, smaller hits - and yes they do add up to more volume. There's also more chance of an outsider becoming a hit due to the viral power of the Internet community. What has not changed tho is that most tracks/artists need a big marketing/editorial focus to achieve significant sales, whether that means one very clever, very busy social media ace or \(far more usually) a campaign team. This isn't to say I disagree with Scott's comments, I just think there is a trap in the long-tail thesis which implies a simplistic equation between unlimited choice and unlimited sales.
Comments
Dan Thornton - 22/12/2008
Two immediate flaws spring to mind: 1. The Long Tail didn't say that there wouldn't be a 'short head' of mass hits that would continue. 2. A study of legal downloads is going to miss the huge amount of illegal downloading, free streaming services etc. With the current disuption to the music industry, perhaps the legal portion is made up of the hits, and it's in the free/illegal sections the long tail really exists?
David Llewelyn-Jones - 22/12/2008
Dan, nice points, totally agree!
Scott Thompson - 22/12/2008
Never mind the flaws in the interpretation- take a second look at those numbers. The "old" idea is that the "head" is where the business is- so by having all the "hits" in stock, record shops were doing good business. The "new" idea is that with more choice, there is more demand- that there is actually more business is selling everything outside the "head" \(ie. the Long Tail) than selling everything in it. Although the numbers are meaningless without a bit more of a context, lets say for arguments sake that it's been counting since the iTunes Music Store launched in 2003 and made legal downloads a realistic option 5 years ago. And, for the sake of simplicity, lets say those 52,000 tracks were released evenly over the course of those 5 years. That would mean that 80% of singles sales came from 200 singles that were released every week. Now the idea of a record shop with 200 singles for sale in a single week is pretty much unthinkable- let alone 200 different singles every single week. \(And remember, record shops usually don't sell old singles- if they're out of the charts, they're off the shelves. Not so with digital sales...) The same calculation for the albums would mean that 665 albums every week were released and sold. Again, have you ever seen a record shop that had over 600 different new album releases on their shelves every single week? Of course, the assumption that those were evenly distributed is nonsense- in actual fact there would have been more releases \(and more sales) as time went on and the digital market grew, so the number of singles/albums released right now that are getting sales would be significantly higher than those numbers. So, the Long Tail theory says that the future of business is "selling less of more"... and this is supposed to DISprove it? Seems to me that it validates the theory perfectly well... \(that is, if you understand the theory to mean something other than "increased choice magically creates a market for everything being sold"...)
Tom Barnes - 23/12/2008
It looks like the old 80:20 rule still applies. Which means that there is a long-tail worth 20% of revenue, predicted to represent 80% of all available music. It's still a long-tail though.
Lazar Dzamic - 23/12/2008
Brilliant Scott, it's a very good argument. To push it even further: it would be great to compare the number of titles \(albums and singles) responsible for 80% of sales BEFORE e-commerce really took off. Whatever it is, I'm darned sure that it is much more narrowly spread than it is now. As you said, this actually proves the theory, not the other way around.
Dan Patton - 23/12/2008
As someone who has managed big music and ents websites over the years, albeit not pure ecom sites like Amazon, I've always struggled to see the long-tail thesis working in practice. If I had to sum up the reasons in one word, that word would be 'marketing'. Even with no physical barriers between music buyer and artist, the buyer still needs to discover the artist and be able to judge whether or not to buy their music. From the publisher's perspective, which I guess is what I represent in this debate, it's still all about hits. What has changed is that there are more genres and therefore more, smaller hits - and yes they do add up to more volume. There's also more chance of an outsider becoming a hit due to the viral power of the Internet community. What has not changed tho is that most tracks/artists need a big marketing/editorial focus to achieve significant sales, whether that means one very clever, very busy social media ace or \(far more usually) a campaign team. This isn't to say I disagree with Scott's comments, I just think there is a trap in the long-tail thesis which implies a simplistic equation between unlimited choice and unlimited sales.