Whither the Long Tail?

Tom Slee is writing an extensive critique of Chris Anderson’s The Long Tail.

Why am I bringing this up here? Well, there’s a lot of discussion these days about what implications online distribution channels (and things like Apple TV) have for indie media, and one side of the debate (the one I’m usually on) relies heavily on arguments that are in essence affirmations of the validity of Anderson’s long tail concept. (Which, for anyone who is unfamiliar, is basically the idea that, with the unlimited “shelf space” available on the Internet, an unlimited variety of products can be offered and discovered, and demand will shift significantly from “hits” — the small set of extremely popular items — down to niche-market items.)

As a critique of Anderson’s book, Slee’s posts are very effective. He deftly takes Anderson to task for making sweeping statements and failing to adequately back them up. The consequences for the long tail arguments often advanced by indie media proponents are, fortunately, not as dire as this might seem to imply.

One of the big problems Slee points out with many of Anderson’s examples is that when Anderson looks at the “hit driven” world of the past, or of traditional retailers, he looks in the wrong places. For instance, Anderson notes that Amazon makes 25% of its sales from books you can’t find in retail stores — but his standard for what you can find in retail stores is Borders. This completely ignores all the specialty retail bookstores that specifically cater to various segments of the long tail market. Slee calls Anderson on making making a similar mistake with music — assuming that the (lack of) diversity on radio was an indication of the overall diversity available in the market, when in fact there was vastly wider variety available in record stores, and there were commercialy successful records that never got radio play.

Why is this not a complete disaster, for people who hope the Internet will change the media landscape? First, because there are some media where, for fundamental technical reasons, there really hasn’t been a long tail market, and now one is developing, and if you want to produce one of those types of media, the fact that the long tail concept might not work so well in some other markets is irrelevant. Secondly, because the long tail isn’t the only reason the Internet and the other changes brought about by digital technology matter to indie media.

To address the first issue, consider podcasts. Most of the podcasts I listen to are targeted at very narrow audiences, like digital media professionals, or even programers using specific programing languages. Prior to the Internet, there was no plausible delivery mechanism for what amounts to niche-market talk radio. (And it would have to be radio; unlike with the music example above, almost nobody is going to buy discussion shows at the record store.) Are there enough people in the US to support a talk radio show about digital media production? Sure. But they’re spread out all over the place. You’d have to have a national radio show to reach a decent number of them, but the audience that show would draw clearly wouldn’t justify tying up scarce spectrum that could be used to broadcast content with vastly wider appeal.

This is what Anderson calls the “tyranny of geography”, and this is a case where it definitely used to be present, and the Internet has eradicated it. I can now wander through Central Park listening to talk radio (AKA podcasts) about design patterns in object-oriented programming or whatever, that 99.9% of the population doesn’t even have the technical background to understand. The people who produce that content can find an audience, and while it’s still not huge, it’s big enough. It’s also highly targeted, which advertisers are willing to pay a lot of money for, so content producers can still make a profit.

Consider the implications for the market that (I assume) most people reading this blog are working in if, over the next few years, devices like Apple TV, along with ever more pervasive broadband Internet access, enable the same kind of thing for high-quality video content.

Slee specifically addresses Anderson’s view, about the future of television, which is somewhat similar to what I’m suggesting. Anderson focuses on what he sees as the the success of Rocketboom, while Slee raises questions about the validity of Anderson’s numbers. From the perspective of the larger issue, this is mostly beside the point. Rocketboom isn’t representative of the sort content we’ll see emerging in the next few years; it’s a few minutes of web video a day; it’s not even really trying to take on television programing. In fact, the infrastructure that Internet video requires to do so is only now slowly emerging.

Slee quotes an MIT Advertising Lab post that advertisers are hesitant to do “small deals”, but from what I’ve seen in the podcasting world, it depends heavily on what audience content reaches. If content accurately targets a valuable narrow demographic, advertisers who have a product that appeals largely to that demographic will play ball. If the advertiser would have to market to a general audience 10x the size in order to reach the same number of people who were actually interested in the specific product, a “small deal” might not be that small after all. Admittedly such valuable narrowly defined audiences are more likely to be found with e.g. podcasts targeted at specific types of professionals than with, I would expect, most narrative content, which is something to keep in mind.

Moving onto the second issue I raise above… what if we assume Anderson’s basic thesis is completely wrong? What if we assume there really won’t be any significant new market opportunities in the long tail? Does this destroy the hopes of anyone who believes the Internet and digital distribution make a difference for indies?

No. Even if you still have a hit-driven market (“hit” here in the “it’s a big hit” sense, not the Web sense), it doesn’t mean nothing has changed. The Internet changes the way hits get created. (For the record, Anderson recognizes this effect. He tries to claim it as part of his long tail thesis, but as Slee points out, it’s really not; a new way of finding hits is not the same thing as a shift toward niche markets.)

When videos go up on YouTube, they’re all basically equal (ignoring channel features, etc. which in my experience the vast majority of users don’t bother with). Yet, some of them eventually get more viewers than a primetime network TV show.

I don’t know what fraction of YouTube video views are of, say, the top 1000 videos; it’s possible most of them are, and YouTube’s traffic is therefore very hit-driven. But those hits weren’t created because the people who posted the videos spent a lot of money on advertising or knew the right people to get word out or had the access to insert them into the right (exclusive) distribution channels. They’re “organic” hits, driven by word of mouth and sometimes by clever promotion that almost anyone has the resources to undertake. While this has always happened, to some extent, the Internet makes it vastly more powerful; word of mouth travels farther, faster. Perhaps even more importantly, digital distribution mechanisms can scale more quickly to meet increased demand.

Of course, most people reading this blog would probably be more interested in a successful theatrical release than a successful YouTube video. What I’m saying here is directly applicable to that. Proponents of theatrical digital projection point out that it will make it easier and cheaper for indie content to get into theaters. But this frankly isn’t that important; if what you’re interested in is getting your movie playing in a few arthouse theaters, that can and does happen today. Digital exhibition will make it cheaper and more common, but that probably won’t have a huge impact on anyone’s bottom line.

What’s really important is that digital exhibition allows a movie to scale to more screens, fast. A distributor is going to be a lot more likely to pick up your movie if they know they can put it on a handful of screens and ramp things up in a short period of time, if it does well. The lead time on striking and delivering film prints makes this difficult with film. With digital projection and network distribution (and with Internet-speed word of mouth), one could see theatrical releases that start small, and then explode with the same force as this month’s Internet meme.

So, to summarize:

  1. Even if a long tail cultural shift (and the market opportunities it implies) isn’t nearly as substantial as Anderson makes it out to be, there appear to be some real market opportunities for people interested in producing video and audio content.
  2. Though many markets are still hit driven, lower barriers to entry, the enhanced power of Internet-enabled word of mouth, and the ability of digital distribution to quickly scale, will still result in some outsiders having success that wouldn’t have otherwise been possible.

Note that this post is intended only to examine how Anderson’s thesis and Slee’s critique of it apply in a few narrow sets of circumstances; both authors make much broader and more complex arguments beyond the scope of this blog. If you’re interested in the wider subjects, I’d encourage you to read the original works.

(By the way, for anyone in the audience getting tired of my industry commentary… there will probably be a lot more technical stuff again following all the new information everyone expects from Red at NAB.)

This entry was posted in Industry Commentary and tagged . Bookmark the permalink.

Leave a Reply