Sunday, July 13, 2008

Long tail not so debunked after-all


There’s been much ado about Anita Elberse’s article in the Harvard Business Review lately – where she seriously questions the validity of the long-tail theory. The press (and that includes bloggers, and me) love anything that runs contrary to a popular opinion in business and culture, and Chris Anderson’s The Long Tail, has certainly been influential. Within days of the issue hitting the stands, I started receiving emails from people making sure I had read it – mainly because they know my organization has launched a project working somewhat off the long-tail theory. I nuance this with the “somewhat;” however, as I’ve never agreed with most of what the adherents of the long-tail seem to believe. Pretty much everyone in the film world who has read it, and I meet someone everyday who hasn’t, seems to think that it says that obscure, niche content can now make more sales. These people believe that by building a better web system, doing more niche marketing or whatever, that their small movies can now become much bigger. This is false, of course, but became a popular belief because, hell, every filmmaker is in need of some golden business rule that can help them sway investors that their little movie can someday be big. A few people have even accused me of thinking this in relation to Reframe. (Which has never been true.)

But what they all miss, and what Elberse’s article doesn’t really address, is that the real truth of the long-tail is simply that in a business environment that allows for more long-tail content and transactions, more value will accrue to those companies that exploit an aggregate of long-tail content. Not more than hit-makers, but more than they made before the web. No single niche title (yes, there are a few exceptions) will become much more popular, but if you aggregate many of them, you can add up those small sales to something approaching profitability. This isn’t refuted by Elberse, and I’m still pretty bullish on the aggregate model; in fact, even more so in relation to this new study.

Elberse points out some very fascinating things about the nature of long-tail business, and while seemingly intuitive in retrospect, it’s great to have some data to back this up. First, she proves that yes, a hit is still a hit and that hits sell a lot more than niche titles. She also shows that this trend is growing. Importantly, however, she identifies two other trends – from 2000 to 2005 sales of the most obscure titles doubled their sales, but the number of titles that didn’t sell a single copy quadrupled. In short, more content is entering the marketplace (thanks, digital, thanks a lot) but not all of it is going to get purchased just because it is available. So, your small title won’t necessarily gain a huge audience just because it’s on every platform available, but you can expect a better marketplace than not to long ago. Those things with some value, however, are seeing an increase in sales due to digital availability. This is especially true on the thinner part of the tail. As Elberse says, “When I differentiate between artists on smaller, independent labels and those on major labels, I find that the former gain some market share at the tail end of the curve.” Clearly, more research is needed on the long end of the tail, but there is a marked increase in activity that you don’t see at the middle end – this is potentially good news for the smaller indies.

When Elberse investigates what does drive sales to the long tail, she finds some fascinating stuff. She analyzes the buying habits of people on Quickflix and finds that people who tend to buy long-tail, niche items buy a lot more of them than anyone else, and that they also buy a lot of popular product. She found that people who rent obscure titles tended to rent more than twice as many titles as those who didn’t (50 per 6 months vs. 20). I’ve always had this hunch, and we built it into Reframe’s design, but she has the data to prove it.

She goes further and shows data that debunks the idea that there is an audience that simply prefers obscure, niche titles – it turns out that even those who buy the most obscure films/music tend to give it much lower ratings than they give to more popular product. This is pretty profound – her data seems to debunk the idea of an “iconoclast” consumer who purchases/rents a lot of obscure media because they like it more. On the contrary, she found strong indications that they like such media less than popular media – they just like to consume a lot of media and therefore go further down the tail. She found this particularly true of consumers who tend to focus on a broad category or niche – perhaps a genre like horror or comedy. Consumers who really like a genre – and importantly, any genre – tend to seek out more obscure titles in this genre. Those who consume more media look for more titles than those who passively consume, but they feed off both popular and niche content. Yep, they read Lapham’s Quarterly and People, watch both Batman and Reminiscenses of a Journey to Lithuania and if they like Wanted a lot, they are more likely to hunt out more obscure titles from the same director, like Day Watch and Night Watch, and the most obsessive will seek out other films from the same region. But don’t think you can capitalize on this too easily, as Elberse cautions that “given that obscure products tend to be appreciated less than hits, it will be very difficult to earn any kind of price premium for them.” That’s an interesting thought in relation to the current model for educational sales!

She gives many scenarios of advice for both producers and distributors of content to consider, but the implications are pretty clear – the long-tail does exist, but the business models to best exploit it may not be what many in indie film have thought. It’s very clear that aggregators, online stores, etc. need to have a mix of both popular and niche content – there isn’t some mythical consumer that only values niche content, and your little film is much more likely to be found if someone can get there while investigating something much better known. This is nothing new of course. Film festival programmers have always used the strategy of mixing an experimental short, say, in front of a more popular feature to build audience for the more obscure title. It works this way online as well. Anyone thinking about how the long-tail impacts the indie film business -festivals, distribution, producing - should study these findings closely because it’s very possible that the idea of separating indie/niche content from popular content (i.e. current practice) is not a good idea in an online, interconnected world.

BTW - That her article is brilliant is not surprising – she is colleagues with and did part of this study with Felix Oberholzer-Gee, and his research has helped debunk the myth that piracy has been hurting sales of music. A couple of debunkers of popular internet myths like these are surely business-minded folks the film world would do well to listen to more clearly.

2 comments:

HL said...

I think there might be some value in comparing the trends (before/after Web) of items not heavily marketed. The focus on content/products in general doesn't give enough consideration to the power of featuring a certain product or piece of content on a homepage or on a "most popular" list. Because the internet is full of networking tools that enable users to directly link a friend to a product they like (long tail or not) and there is no shortage of availability if a user seeks out a product on their own, it would be interesting to see if there is a fattening of the tail due to social media, if we consider only those products that have not been advertised or heavily marketed.

Anonymous said...

It is certainly interesting for me to read that post. Thanks for it. I like such themes and everything connected to this matter. BTW, why don't you change design :).