Thursday, December 04, 2008

long tail is dead

so i'm on this list serve (FA Worldmusic) where I occasionally weigh in. today i weighed in heftily - check it out.


Nothing like being called out to make one sit down and start writing. ;) Thanks Dmitri!

Mark, I think the simple answer is that no one is making money from digital sales yet - even the top overall sellers like Katy Perry are not seeing large percentages of their income come from digital sales. 

Now onto the long tail of my response - The first thing I think it is important to mention in a discussion of the Long Tail is that it is a theory, or a concept. For those who weren't all geeked out when the term came to common use, the idea of the Long Tail as it relates to music and digital content was put forward by Chris Anderson, WIRED editor. For the Wiki history of this check out
His blog is now called The Long Tail (
And the original article is here:
You can also get the book from the retailers he has linked to in his blog.

The quick overview (cribbed from wikipedia) is that TLT (the long tail) describes the niche strategy of businesses that sell a large number of unique items, each in relatively small quantities. Given a large enough availability of choice, a large population of customers, and negligible stocking and distribution costs, the selection and buying pattern of the population results in a power law distribution curve. This suggests that a market with a high freedom of choice will create a certain degree of inequality by favoring the upper 20% of the items ("hits" or "head") against the other 80% ("non-hits" or "long tail"). The idea among those of us who live in the this long tail was that digital would free us from the oppression of major label domination and allow smaller labels and artists to make a living from sales within the long tail. 

From my own experience within a company entirely dependent on long tail sales (, it was painfully apparent that the "negligible stocking & distribution costs" (along with other costs like licensing, marketing, software development, etc) were in fact major blocking points. To overcome these costs would require investment of outside money, which has (to my knowledge) not yet happened with this particular retailer. 

Others apparently have been seeing this and other flaws in the theory, and as Anderson noted on his blog, there has been a flurry of reappraisals of the Long Tail ( These reappraisals I think are what caused Ian to originally post his note here.

So I guess the question for us in the theoretical Long Tail (if it exists), the question remains, where is the Long Tail as a theory today and how can we (if applicable) put it to use for our artists, our market, our companies? From what I'm reading and seeing, the general consensus is that TLT (the long tail) is dead, or that the theory has been disproven. 

Currently, I'm working inside a recommendations company (a key part of Anderson's original thesis), and what I see troubles me in terms of looking at "world music" and other related independent music. As has been well documented, overall music sales are declining (with the growth in digital not nearly enough to offset losses in physical sales, in which non-top100 sales are even more depressed), and has been amply stated, discussed, and anecdotally shown on this list, independent artists now do NOT make a living off selling recorded music (not sure if they ever did, but that's a different story). The existing modes of distribution don't work horribly well for long tail artists, and future modes (mobile phones) certainly don't. For most mobile services I've seen from the inside, only 1-3% of the catalog ever sells, and maybe only 5% gets looked at. These numbers just don't work for "long tail artists". We did just start working with a retailer focused on independent music, so it'll be very interesting to see how/if recommendations can drive customers down the long tail to more unknown content.

I am in agreement with a line in the original article "With cheap production tools and the internet as a new distribution channel, some costs of production are indeed lowered, and some artists can indeed cut out (or "disintermediate") the middle man. But those old rules still make a significant difference to your business strategy. "In particular, the division of labour and economies of scale still have tremendous relevance to understanding today’s market”, Page notes. The division of labour means it can benefit you to employ a specialist intermediary, while economies of scale mean the bigger you are, the better terms you can negotiate."

In laymans terms, I think the biggest blocking point to accessing new audiences, or to accessing an audience that cares are marketing (dollars & strategies) and filters. Megan discussed filters, and I concur with her - we should be cultivating sites that feature only certain types of music, developing good and interesting writers, cultivating fan writers/promoters, etc - creating & loving filters (Mondomix, Calabash, AfroPop, etc etc etc). 

That's the easy part.

The hard part is from that piece I quoted above and has to do with marketing & access. Small artists can't get the same terms as major labels. And even small labels can't get those same terms. And the kicker - major labels are not sharing their well negotiated terms with individual artists on their rosters. Also, majors have much more money to experiment with strategies of marketing & distribution to find their fanbase, or develop a fan base. 

Folks who are way smarter than I are saying that things are in general shifting the other way of positive (Google's Eric Schmidt says we're in a 90/10 economy - worse than the traditional 80/20 model for us in the smaller piece -

And I think that Dmitri's comment "What I take from the Long Tail model is that artists who in the old model did not have access or ability to sell music or make a living, now can" is a misreading of the long tail and emergent technologies. Sure new technologies make artists available and accessible to the world in new ways. However, those technologies existence does not guarantee (or even offer in many cases) any financial return. 

Dmitri can place his clients all he wants on NPR, in the NYTimes, and in every other conceivable news outlet. But if those artists don't actually make money through sales, or aren't available (or easily available) in a sales outlet, then they don't have money to hire him again. More important than album sales, if those artists can't develop a fan network through touring or social network outreach, they won't have money to record again, to tour again, to hire marketers again. In short, most artists are in the same boat as before. Lowered costs of hosting, recording, etc don't change the demand equation, only the supply side equation. We have to be in the business of developing demand, not supply. Supply will always be there. But we have to be develop demand. 

Artists have to follow Megan's advice and build a career, build a fanbase, not just put out a digital release and expect to get rich due to the long tail. They DO have to know about digital technology, and how to use it. They do have to know that things are interconnected. For example, do you (or your artist) know what is on their wiki page? Or know that this page is then shot off to for use on their pages? Or used in Songbird's music player implant mashTape? If that page is blank, your artist is a nothing to users of those services. 

Again, back to developing demand. This is our job - create audiences, find audiences, get them interested in the music we love. And we're not alone, which is great. This means we can work and share models and actual festivals/venues/etc with folk musicians, indy rockers, jazz artists, etc. These I think are the topics to discuss at APAP. How to collaborate, share knowledge, share resources, and build audiences. 

Anyway, that is more than enough from me at this point. If you want some more contrarian points of view, check out Will Page's critique of the 1000 fans model -



soungalo said...

couple interesting posts counter to the long tail critique -

soungalo said...