As the Information on Demand conference wound down in Las Vegas last week, I had the opportunity to sit down with Wired magazine's editor-in-chief, Chris Anderson.
During our 30 minute conversation, Anderson talked about his groundbreaking book "The Long Tail," about the economics (and disruption) being created by the power and reach of the Internet, about rock band Radiohead's recent attempt to master its own destiny without the assistance of a major record label by going direct to consumers, and much, much more.
(Warning: Like the long tail, this blog post is long and covers topics interesting and niche, but it is more a marathon than a sprint, so feel free to print out for a rainy day if you find that it's overloading your RSS reader).
Turbo: Chris, thanks for taking the time to speak with me. I really found the ideas in the book fascinating, and I’m really interested in understanding how you found your way to the idea of the Long Tail.
Was there a particular “Aha” moment where you realized there was something going on here that needed to be called out and explained?
Anderson: My background is science and economics. I’m really a data guy, and what we have today is the largest proliferation of data the world has ever seen. Google’s basically built on that. And what I realize was that the economics of the 21st century were lying in the servers and databases of the Yahoos and Googles of the world, and we just had to get the data and look at it, and we’d be able to understand who we are, and what we want in these new marketplaces.
So I started with a music company called Rhapsody, and I got the data set, and got all their tracks ranked by popularity. And it looked like this (L shaped). And I was like, what the hell, I’ve done something wrong. And I suddenly realized that the difference in popularity between the #1 track and everything else had exaggerated the vertical axis. And the sheer number of tracks…there were 3 million of them…had exaggerated the horizontal axis.
And in fact what had happened was that the whole thing had been stretched in both directions by the dynamic range of popularity. And the variety, the vast variety…well, we’d never really seen markets with 3 million songs before.
And the only way I could analyze this was to first get rid of the vertical axis…I’m gonna cut off the top 100 tracks…and just started from 101 to 3,000,000…and recalibrated the scale.
So the scale went down by a factor of 100. And then you saw the curve of the Long Tail begin to take shape. But in a scale set by a number 1 hit, the tail was obscured. So looking to the right, rather than to the left – to the market beyond the blockbusters – revealed there was a lot of energy there.
There were thousands and millions of tracks that were still selling…you just weren’t noticing because we were so focused on the top 100. And then what I did was that I added up all those millions of tracks that would be on the top 100, and those would add up to being 34% of the total market.
So in other words, the top 100 was 66%, the bottom was 34%, and that’s now become closer to 50/50 today. And that moment was just recognizing that we’d been focused on the blockbusters and the best sellers, because we’d never really before had markets that had room for everything else. For the first time we had room for a market that had 3 million tracks.
We could then not just assume what people wanted, we could measure what people wanted. And when you do that, it turns out that half the market is so-called “misses.”
Turbo: I think I read that when went about writing the book you actually leveraged your blog, the Web, and input from others to use as a sounding board. I’m curious how that went and whether or not that’s something you plan to continue with future books?
Anderson: Yeah, absolutely. So the Long Tail started with that sort of epiphany moment we talked about, which then turned into a series of speeches, which sort of crisped up the term. The Long Tail, by the way, comes to us from statistics.
That curve is called the “powerlaw” and is known as a “Long Tail” distribution. So I crisped up the name, and then published the article in Wired in October 2004, which became the most cited article Wired had ever done, so it was on the basis of that that I decided to write a book.
One of the things you realize when you write a book is that they take a long time. Not just to write, but also the whole publishing and distribution process. So I realized that if I started the book – we signed the deal in December 2004 – and the book was scheduled to come out in June 2006. So I realized that there was going to be eighteen months of radio silence.
This big idea that I’d started, that I’d launched out into the world, was going to go quiet! And I just couldn’t bear letting the momentum die. I couldn’t bear the idea that someone else was going to take it and run with it. I couldn’t bear losing ownership of the meme.
So I started the blog entirely selfishly to own the idea during the dead period when I was working on the book. And what happened was that worked, and it did sort of keep the focus around me and I could retain ownership of the idea. But what also happened were two surprising things that turned out to be even more powerful.
The first surprising thing was that this practice of giving away my ideas, my data in progress, and sharing my research as it went, stimulated this response through comments and emails and suddenly I had created an open-source collaborative research project.
Simply by the act of giving something away brought back so much more in return that I had accidentally built a research team who were thinking about the theory in fields that I had never considered, like travel and NGOs and medicine and food…people who got the idea, and who knew their own world and brought in data and examples from that world.
So in a sense it became a collaborative research project that not only made my job easier, and hugely expanded the range to which the theory was being applied, but it also caught my mistakes.
You know, the same way when we open source software, or with peer review with scientific articles, we use our peers to catch our mistakes, then we finally publish the article. You wouldn’t publish software without beta testing it. Why would you publish a book without beta testing it?
So in a sense the blog became the beta test of the book, and when the book came out, it was better because it had already been stress tested by the many thousands of people who were reading the blog. The third big advantage of the blog only emerged at the end. By the time the book was done and ready to be published, I had probably on the order of 10,000 daily readers of the blog and they became the greatest distributed marketing channel ever.
They became…this is how word of mouth got started. And I offered to give away ARCs, or author review copies…galleys. My publisher said, “Here, you can have some ARCs to give away to influentials.”
And I said, “Okay, how many can I have?” He said, “How many do you need? 10, 20? 30?” I said, “How about 2,000?”
No one had ever asked for 2,000 before. But I said, here’s what I want to do: On my blog I’m just going to put one post that says anybody who has a blog who wants a copy of the book for free, to blog about it – positive or negative, whatever – just send me your address. Free books for bloggers.
And so we ended up sending out 1,000 books, which costs maybe $2.50 a piece, so maybe it cost us $2,500 to send it out, and we ended up with 800 reviews. And the book sold better on Amazon than on Barnes and Noble. Amazon represents about 15% of the industry, Barnes and Noble is about 40%, and the book sold better on Amazon because I think all those blog reviews pointed to Amazon. That’s how we got word of mouth going.
So the blog community that had built up became a grassroots marketing machine and that’s what helped the become a best seller.
Turbo: You also talk in the book about the end of the blockbuster era for music. I’m curious what you think a world without hits looks like? Will the mass, culturally-shared watercooler experience we had in the past go the way of the Edsel?
Anderson: One of the biggest misunderstandings in the book, one I’m slightly responsible for because of overstatement and hyperbole, but I think I tried to go to pains to correct in the book itself, is that it’s not the end of the blockbuster, but rather the end of the monopoly of the blockbuster.
That power law invariably has a head and a tail, it has a hits and non hits, it has hits and niches. What you have in the twentieth century was just the blockbusters, just the products which were popular enough to make good use of scarce distribution channels, be they screens, channels, stations, or shelf space in stores.
Now we have marketplaces that have room for everything, the hits and the niches. And so what you’re seeing is this kind of shift of attention, shift of populations, away from total concentration around the blockbusters to more of a shared marketplace. We spend some of our life watching Hollywood movies, we spend some of our life listening to indy music or watching documentaries.
So what you’re seeing is not the end of the blockbuster, but rather the end of the dominance, the end of the exclusive domain of the blockbuster. Now the blockbuster is going to have to share the marketplace with smaller niche products.
Turbo: Building on that, as you know we’ve seen some big moves by major artists over the last several weeks – Nine Inch Nails, Radiohead, to some extent, Madonna – who announced they are going to go direct to their fans. Do you think this is a trend that will continue to play out in the music industry, and do you expect that that will bleed over into Hollywood? And if so, what do you envision to be the longer term impact of that?
Anderson: I think yes to the first, and no to the second. I think that what you’re seeing in the movie industry is something that should have been obvious to everybody. Which is that our definition of the music industry as the sale of plastic disks is a very narrow definition of this industry.
When you look at the actual numbers, Radiohead is going to make more than 80% of their money from touring, the Rolling Stones make more than 95% of their money from touring.
Turbo: I didn’t realize that.
Anderson: Well, think about it. What is a ticket to Radiohead gonna cost, maybe $600 for good seats, or something like that. It’s a huge amount of money. I’m actually working on a post right now that I hope you don’t beat me to. I believe that every single part of the music industry is growing, except for plastic disks.
Digital downloads, ringtones, concerts, license fees, merchandising, even twelve inch singles are growing thanks to the DJs. By the way, if you include the iPod, as I believe a fair assessment would, then the music industry has never been bigger than it is today. And in fact, those plastic disks represent something like a low double digits percentage.
And yet, in all of our sense of what the policies and politics and economics of the music industry should be are still focused on the sale of plastic disks. I think music is going to become free because music as a product can be distributed to zero marginal costs. And economics will tell you that price tends to follow cost.
So, what you should do is give away the music to sell the show. You want to put the music, the product, which is a commodity that can be replicated, into the hands of as many people as possible, so that the performance, which is a non-commodity that cannot be replicated, can attract the broadest possible audience.
Turbo: Well I think the precedent is already there. That’s essentially what the Grateful Dead did…they set up a little section where everybody could place their tape recorders?
Anderson: That’s right. And Prince gave away his album in the Daily Mail…you know, but the problem is the major labels are in the plastic disk business, whereas the artists are in the artist business.
And I think what you’re seeing right now is a misalignment of incentives, that the artists see their financial health being a much broader perspective of licensing, touring, merchandising, and all that. And the labels are just focused on one portion of that. Their interests have just now diverged, and as a result, they’re legally diverging as well, they’re leaving the labels.
Turbo: I actually did a little of asking for input from others as well, and one of the questions was about the original Wired article on the Long Tail, in which you laid out this idea that access to information about Touching the Void from reviews about another book Into Thin Air could drive sales of the Touching the Void and cause it to come back to print.
Victory for the Long Tail: But would Touching the Void have had any chance of getting attention in the Long Tail had Krakauer’s book not been around for Amazon readers to review or Amazon to suggest via the people who had also read it?
Put another way, does it still take some external entity to drive the “hit” which could lead to the increased traffic and awareness and aren’t such “hits,” per se, still selected by some filter. It’s just now they’re crowd-sourced or socially selected instead of identified by a reviewer or some other person of “authority?”
Anderson: Sure, the old model of hit was what I call a “synthetic” hit, which was hits only happened when the hitmaking industry made them happen. Now some of those hits were very good A&R functions, the talent-spotting functions of this industry, spotting good stuff and then improving it and elevating it and marketing it.
Some of it is just crap that’s generated and it becomes successful because they’re so good at marketing. You know, look at Hollywood movies: some of them are great, some of them are terrible.
But even the terrible ones end up with millions of dollars of box office sales because the marketing engine is so effective. So those are the synthetic hits. These are the hits that basically are top down, that are sort of pushed out with a powerful marketing machine.
What we’re seeing now is a different kind of hit, this is the bottoms up organic hit. These are the MySpace bands that build their fan base show by show, kind of word of mouth, very organic, sometimes not signed to a major label at all.
These are the ones that kind of come from below and come to popularity through the authentic meritocracy of what people actually think. Now they may get signed to a major label or they may get broad distribution and then be taken to the next stage, but they basically started at the bottom.
And I think the top down synthetic hits, those filters tend to be professionals whose job it is to spot talent. The bottoms up hits, those filters tend to be the crowd, as the question suggests. I think you’ll continue to see both of them.
It’s just now those bottom up hits were never really possible before, and now they are because we now have the Web, which is the greatest word of mouth amplifier ever created.
The question also said, would that book have become a hit without the mainstream hit to coattail on, and maybe not. That is absolutely true. You know, one of the examples I give in the book is the importance of having both head and tail in the marketplace.
You need to start with the world you know, and that is often blockbuster stuff. You know, you start with Madonna, and if you like Madonna, you might like Pink, and if you like Pink, you might like No Doubt, and if you like No Doubt, you might like The Selector, a 1980s ska band from Coventry, England.
So you basically started with the top of the head with the mainstream, and then through this process of collaborative filtering and getting to know more about you, they take you down to the tail.
So, it’s not just one, it’s head and tail.
Turbo: Okay. So this kind of takes the other end of that as a consumer. You know there’s a lot of stuff out there, and I’m sure you deal with this in your day-to-day existence. I know I do. There’s not enough time to scan everything, to read everything…there’s a lot of stuff out there. And I don’t think it’s just a matter of attention span.
There’s a real temporal cost to reading all this stuff: Web Pages, the YouTube videos, etc. And I think in the past we had a clearer focus for generating hype promoting these special interests, and we also had these clear lenses for filtering and focusing attention on the targets of the hype, whether it was the movie or book reviewers or whatever. Now, anybody can create a firehose of attention and focus anywhere they want.
Anderson: Is that true? What you mean is that successful amateurs can participate?
Anderson: Yes, that’s true, although not everybody can, but it occasionally happens.
Turbo: The capability is there.
Anderson: Yes. The capability is there.
Turbo: Yes, which leads us back to the whole declining costs of production thought. So I wonder overall, do you think this is a social or economic good, A, and B, do you think we might face now instead a tyranny of time? Might the limited capacity of our own attention be overwhelmed by the volume of abundant choice?
Anderson: Yeah, well you know we used to have very limited filters: The prime time broadcast schedule, The New York Times, Hollywood, and Top 40 radio. You know, it was easier because there were fewer choices. We didn’t really know how satisfied we were because we didn’t know what our options were.
We now have way too many choices and we’re sort of groping our way to a new class of filter. These filters are now amateurs, they’re people whose tastes we respect. And it’s messy. The reality is I’ve got 226 RSS feeds and I can spend my whole life reading my feeds. I haven’t got it right yet.
The good news is that these feeds collectively are relevant to my interests better than any one of those previous filters. Better than The New York Times, better than prime time, better than etc. So they’re better, but there’s still too much. I do still feel a sense of information overload.
These are early days yet. Over time, I will get better at finding individuals who are filters, I will get better at finding software filters, and the technology will get better at filtering. And I hope that the proverbial “Daily Me” will start to emerge.
But I have to admit it gets messier before it gets cleaner, and we’re all just trying to figure out what peer filtering really means and how best to do it.
Turbo: Why don’t we close with you opening the kimono a bit on your new book, Free.
Anderson: So we were talking about the music industry and how music, the product, wants to become free to market the performance, which is anything but free.
My new book is about broadly that. It’s about the emerging industry of making money by giving things away. The Internet is built on free. The underlying technologies, of bandwidth and storage and processing, are close to free and as a result phone calls are free and emails are free.
The music industry is a glimpse into all industries, or to many industries, and I suspect it’s a glimpse into the book industry as well. One of the experiments we’re going to do in the course of this is we’re going to make the book free. The audio book will be free, the ebook will be free, the web book will probably be advertising supported, maybe even the physical book will be free and advertising supported.
All in the effort to sort of maximize the reach of the idea, in the same way that Radiohead wants to get their music into more peoples’ hands as marketing for their concerts, I want to put my idea into as many hands as possible at the lowest possible cost. And I’ll figure out how to monetize it, whether it’s speaking, consulting…that’s a good problem. Maximum celebrity, maximum recognition gives you maximum opportunity for new business problems.
Turbo: Thanks for taking the time, Chris.
Anderson: Thank you.