The Long Tail

by Chris Anderson

  Think of these falling distribution costs as a dropping waterline or a receding tide. As they fall, they reveal a new land that has been there all along, just underwater. These niches are a great uncharted expanse of products that were previously uneconomic to offer. Many of these kinds of products have always been there, just not visible or easy to find. They are the movies that didn’t make it to your local theater, the music not played on the local rock radio station, the sports equipment not sold at Wal-Mart. Now they’re available, via Netflix, iTunes, Amazon, or just some random place Google turned up. The invisible market has turned visible.

  Other niche products are new, created by an emerging industry at the intersection between the commercial and noncommercial worlds, where it’s hard to tell when the professionals leave off and the amateurs take over. This is the world of bloggers, video-makers, and garage bands, all suddenly able to find an audience thanks to those same enviable economics of digital distribution.

  THE 98 PERCENT RULE

  This book began with a quiz I got wrong. One of the things I do as the editor of Wired is give speeches about technology trends. Because I started my career in the science world and then learned economics at The Economist, I look for those trends first in hard data. And, fortunately enough, there has never been more data available. The secrets of twenty-first-century economics lie in the servers of the companies that are all around us, from eBay to Wal-Mart. Although it’s not always easy to get the raw numbers, the executives at those companies swim in that data every day and have a great intuitive feel for what’s meaningful and what isn’t. So the trick to trend-spotting is to ask them.

  Which is what I was doing in January 2004, in the offices of Robbie Vann-Adibé, the CEO of Ecast, a “digital jukebox” company. Digital jukeboxes are just like regular jukeboxes—a big enclosure with speakers and blinking lights, often found in bars—with the difference that rather than a hundred CDs, they have a broadband connection to the Internet and patrons can choose from thousands of tracks that are downloaded and stored on a local hard drive.

  During the course of our conversation, Vann-Adibé asked me to guess what percentage of the 10,000 albums available on the jukeboxes sold at least one track per quarter.

  I knew, of course, that Vann-Adibé was asking me a trick question. The normal answer would be 20 percent because of the 80/20 Rule, which experience tells us applies practically everywhere. That is: 20 percent of products account for 80 percent of sales (and usually 100 percent of the profits).

  But Vann-Adibé was in the digital content business, which is different. So I thought I’d go way out on a limb and venture that a whopping 50 percent of those 10,000 albums sold at least one track a quarter.

  Now, on the face of it, that’s absurdly high. Half of the top 10,000 books in a typical book superstore don’t sell once a quarter. Half of the top 10,000 CDs at Wal-Mart don’t sell once a quarter; indeed, Wal-Mart doesn’t even carry half that many CDs. It’s hard to think of any market where such a high fraction of such a large inventory sells. But my sense was that digital was different, so I took a chance on a big number.

  I was, needless to say, way, way off. The answer was 98 percent.

  “It’s amazing, isn’t it?” Vann-Adibé said. “Everyone gets that wrong.” Even he had been stunned: As the company added more titles to its collections, far beyond the inventory of most record stores and into the world of niches and subcultures, they continued to sell. And the more the company added, the more they sold. The demand for music beyond the hits seemed to be limitless. True, the songs didn’t sell in big numbers, but nearly all of them sold something. And because these were just bits in a database that cost nearly nothing to store and deliver, all those onesies and twosies started to add up.

  What Vann-Adibé had discovered was that the aggregate market for niche music was huge, and effectively unbounded. He called this the “98 Percent Rule.” As he later put it to me, “In a world of almost zero packaging cost and instant access to almost all content in this format, consumers exhibit consistent behavior: They look at almost everything. I believe that this requires major changes by the content producers—I’m just not sure what changes!”

  I set out to answer that question. I realized that his counterintuitive statistic contained a powerful truth about the new economics of entertainment in the digital age. With unlimited supply, our assumptions about the relative roles of hits and niches were all wrong. Scarcity requires hits—if there are only a few slots on the shelves or the airwaves, it’s only sensible to fill them with the titles that will sell best. And if that’s all that’s available, that’s all people will buy.

  But what if there are infinite slots? Maybe hits are the wrong way to look at the business. There are, after all, a lot more non-hits than hits, and now both are equally available. What if the non-hits—from healthy niche product to outright misses—all together added up to a market as big as, if not bigger than, the hits themselves? The answer to that was clear: It would radically transform some of the largest markets in the world.

  And so I embarked on a research project that was to take me to all the leaders in the emerging digital entertainment industry, from Amazon to iTunes. Everywhere I went the story was the same: Hits are great, but niches are emerging as the big new market. The 98 Percent Rule turned out to be nearly universal. Apple said that every one of the then 1 million tracks in iTunes had sold at least once (now its inventory is twice that). Netflix reckoned that 95 percent of its 25,000 DVDs (that’s now 90,000) rented at least once a quarter. Amazon didn’t give out an exact number, but independent academic research on its book sales suggested that 98 percent of its top 100,000 books sold at least once a quarter, too. And so it went, from company to company.

  Each company was impressed by the demand they were seeing in categories that had been previously dismissed as beneath the economic fringe, from the British television series DVDs that are proving surprisingly popular at Netflix to the back-catalog music that’s big on iTunes. I realized that, for the first time, I was looking at the true shape of demand in our culture, unfiltered by the economics of scarcity.

  That shape is, to be clear, really, really weird. To think that basically everything you put out there finds demand is just odd. The reason it’s odd is that we don’t typically think in terms of one unit per quarter. When we think about traditional retail, we think about what’s going to sell a lot. You’re not much interested in the occasional sale, because in traditional retail a CD that sells only one unit a quarter consumes exactly the same half-inch of shelf space as a CD that sells 1,000 units a quarter. There’s a value to that space—rent, overhead, staffing costs, etc.—that has to be paid back by a certain number of inventory turns per month. In other words, the onesies and twosies waste space.

  However, when that space doesn’t cost anything, suddenly you can look at those infrequent sellers again, and they begin to have value. This was the insight that led to Amazon, Netflix, and all the other companies I was talking to. All of them realized that where the economics of traditional retail ran out of steam, the economics of online retail kept going. The onesies and twosies were still only selling in small numbers, but there were so, so many of them that in aggregate they added up to a big business.

  Throughout the first half of 2004 I fleshed out this research in speeches, the thesis advancing with each talk. Originally the speech was called “The 98 Percent Rule.” Then it was “New Rules for the New Entertainment Economy” (not one of my better naming moments).

  But by then I had some hard data, thanks to Rhapsody, which is one of the online music companies. They had given me a month’s worth of customer usage data, and when I graphed it out, I realized that the curve was unlike anything I’d seen before.

  It started like any other demand curve, ranked by popularity. A few hits were downloaded a huge number of times at the head of the curve, and then it fell off steeply with less popular tracks. But the interesting thing was that it never fell to
zero. I’d go to the 100,000th track, zoom in, and the downloads per month were still in the thousands. And the curve just kept going: 200,000, 300,000, 400,000 tracks—no store could ever carry this much music. Yet as far as I looked, there was still demand. Way out at the end of the curve, tracks were being downloaded just four or five times a month, but the curve still wasn’t at zero.

  In statistics, curves like that are called “long-tailed distributions,” because the tail of the curve is very long relative to the head. So all I did was focus on the tail itself, turn it into a proper noun, and “The Long Tail” was born. It started life as slide 20 of one of my “New Rules” presentations. I think it was Reed Hastings, the CEO of Netflix, who convinced me that I was burying my lead. By the summer of 2004 “The Long Tail” was not just the title of my speeches; I was nearly finished with an article of the same name for my own magazine.

  When “The Long Tail” was published in Wired in October 2004, it quickly became the most cited article the magazine had ever run. The three main observations—(1) the tail of available variety is far longer than we realize; (2) it’s now within reach economically; (3) all those niches, when aggregated, can make up a significant market—seemed indisputable, especially backed up with heretofore unseen data.

  TAILS EVERYWHERE

  One of the most encouraging aspects of the overwhelming response to the original article was the breadth of industries in which it resonated. The article originated as an analysis of the new economics of the entertainment and media industries, and I only expanded it a bit to mention in passing that companies such as eBay (with used goods) and Google (with small advertisers) were also Long Tail businesses. Readers, however, saw the Long Tail everywhere, from politics to public relations, and from sheet music to college sports.

  What people intuitively grasped was that new efficiencies in distribution, manufacturing, and marketing were changing the definition of what was commercially viable across the board. The best way to describe these forces is that they are turning unprofitable customers, products, and markets into profitable ones. Although this phenomenon is most obvious in entertainment and media, it’s an easy leap to eBay to see it at work more broadly, from cars to crafts.

  Seen broadly, it’s clear that the story of the Long Tail is really about the economics of abundance—what happens when the bottlenecks that stand between supply and demand in our culture start to disappear and everything becomes available to everyone.

  People often ask me to name some product category that does not lend itself to Long Tail economics. My usual answer is that it would be in some undifferentiated commodity, where variety is not only absent but unwanted. Like, for instance, flour, which I remembered being sold in the supermarket in a big bag labeled “Flour.” Then I happened to step inside our local Whole Foods grocery and realized how wrong I was: Today the grocery carries more than twenty different types of flour, ranging from such basics as whole wheat and organic varieties to exotics such as amaranth and blue cornmeal. There is, amazingly enough, already a Long Tail in flour.

  Our growing affluence has allowed us to shift from being bargain shoppers buying branded (or even unbranded) commodities to becoming mini-connoisseurs, flexing our taste with a thousand little indulgences that set us apart from others. We now engage in a host of new consumer behaviors that are described with intentionally oxymoronic terms: “massclusivity,” “slivercasting,” “mass customization.” They all point in the same direction: more Long Tails.

  A PREVIEW OF TWENTY-FIRST-CENTURY ECONOMICS

  This book is partly an economic research project, with the help and involvement of students and professors from the Stanford, MIT, and Harvard business schools. It’s partly the fruit of more than a hundred speeches, brainstorming sessions, and site visits with companies and industry groups that see the Long Tail changing their world. And it’s partly a collaboration with the dozens of companies and executives who shared many megabytes of internal data, giving me an unprecedented view on the emerging micro-economics of markets in the online age.

  What’s fascinating about this moment is that the economics of the twenty-first century are already evident in outline form in the databases of the Googles, Amazons, Netflixes, and iTunes of the world. In those many terabytes of user behavior data is a clue to how consumers will behave in markets of infinite choice, a question that hadn’t been meaningful until recently but has now become essential to understand.

  Surprisingly, very few economists are looking at this data, mostly because they haven’t asked (most of the academics I worked with are in business schools, only a few of them are economists). There are some exceptions—University of California Berkeley economist Hal Varian works part-time at Google, and auction-theory economists unsurprisingly love eBay—but they’re rare. Some of the data in this book has never before seen the light of day.

  Given the uncharted waters, I solicited a lot of help from experts in all corners. As an experiment, I worked through many of the trickier conceptual and articulation issues in public, on my blog at thelongtail.com. The usual process would go like this: I’d post a half-baked effort at explaining how the 80/20 Rule is changing, for instance, and then dozens of smart readers would write comments, emails, or their own blog posts to suggest ways to improve it. Somehow this wonky public brainstorming managed to attract an average of more than 5,000 readers a day.

  In software, developers release early (“beta”) versions of their code to their most avid users. In exchange for the privileged early look at the program, these users test it on their own machines, in their own way, and find errors that the developer missed. Such beta-testing is essential to creating robust software applications. My hope is that the same process—stress-testing many of my ideas in public—has led to a better, or at least sounder, book.

  I should note here the difference between beta-testing ideas in public and actually writing a book in public. Although many have tried to do the latter—posting draft chapters online and sometimes even opening the text to collective editing—I chose to use the blog mostly as a public diary of my research in progress. The actual writing of the book, and most of the words in the following pages, I did offline.

  Finally, one more note on parentage. Although I coined the term “The Long Tail,” I can’t claim any credit for creating the concept of using the efficient economics of online retail to aggregate a large inventory of relatively low sellers. That would be Amazon’s Jeff Bezos, circa 1994. Most of what I’ve learned has come from talking to him, his counterparts at Netflix and Rhapsody, and others who have all been acting on this for years.

  Those entrepreneurs are the real inventors here. What I’ve tried to do is synthesize the results into a framework. That is, of course, what economics does: It seeks to find neat, easily understood frameworks that describe real-world phenomena. Coming up with the framework is an advance in itself, but it pales next to the original inventions of all those who discovered and acted on the phenomena in the first place.

  1

  THE LONG TAIL

  HOW TECHNOLOGY IS TURNING MASS MARKETS INTO MILLIONS OF NICHES

  In 1988, a British mountain climber named Joe Simpson wrote a book called Touching the Void, a harrowing account of near death in the Peruvian Andes. Though reviews for the book were good, it was only a modest success, and soon was largely forgotten. Then, a decade later, a strange thing happened. Jon Krakauer’s Into Thin Air, another book about a mountain-climbing tragedy, became a publishing sensation. Suddenly, Touching the Void started to sell again.

  Booksellers began promoting it next to their Into Thin Air displays, and sales continued to rise. In early 2004, IFC Films released a docudrama of the story, to good reviews. Shortly thereafter, HarperCollins released a revised paperback, which spent fourteen weeks on the New York Times best-seller list. By mid-2004, Touching the Void was outselling Into Thin Air more than two to one.

  What happened? Online word of mouth. When Into Thin Air first came out, a few readers
wrote reviews on Amazon.com that pointed out the similarities with the then lesser-known Touching the Void, which they praised effusively. Other shoppers read those reviews, checked out the older book, and added it to their shopping carts. Pretty soon the online bookseller’s software noted the patterns in buying behavior—“Readers who bought Into Thin Air also bought Touching the Void”—and started recommending the two as a pair. People took the suggestion, agreed wholeheartedly, wrote more rhapsodic reviews. More sales, more algorithm-fueled recommendations—and a powerful positive feedback loop kicked in.

  Particularly notable is that when Krakauer’s book hit shelves, Simpson’s was nearly out of print. A decade ago readers of Krakauer would never even have learned about Simpson’s book—and if they had, they wouldn’t have been able to find it. Online booksellers changed that. By combining infinite shelf space with real-time information about buying trends and public opinion, they created the entire Touching the Void phenomenon. The result: rising demand for an obscure book.

  This is not just a virtue of online booksellers; it is an example of an entirely new economic model for the media and entertainment industries, one just beginning to show its power. Unlimited selection is revealing truths about what consumers want and how they want to get it in service after service—from DVDs at the rental-by-mail firm Netflix to songs in the iTunes Music Store and Rhapsody. People are going deep into the catalog, down the long, long list of available titles, far past what’s available at Blockbuster Video and Tower Records. And the more they find, the more they like. As they wander farther from the beaten path, they discover their taste is not as mainstream as they thought (or as they had been led to believe by marketing, a hit-centric culture, and simply a lack of alternatives).