The Register has an article on a study on the long tail that puts an additional chink or two in the armor of the popular digital distribution theory. Economist Will Page (of the MCPS-PRS Alliance), his colleague Gary Eggleton and Mblox founder Andrew Bud looked at tens of millions of transactions and came to conclusions that are odds with Chris Anderson's long tail teachings.

The transactions had a log-normal distribution, not the power law distribution common in long tail theory. The trio found a relatively concentrated head and a "rather poverty stricken tail."

Research by Anita Elberse, an associate professor at Harvard Business School, has similar findings about the concentration of hit titles and consumers' lack of interest in the obscure. "Light users have a disproportionately strong preference for the more popular offerings," she wrote in a reaction to Anderson's rebuttal to her study.

"Is the 'future of business' really selling more of less?" asked Page. "Absolutely not. If you had Top of the Pops now, you'd feature the Top 14, not Top 40."

The key thing about research that contradicts Anderson's theories is that they come to conclusions at odds with the business strategies he laid out in "The Long Tail." The issue is not about the shape of the tail or what constitutes the curve's short head. "The Long Tail" told us consumers were losing their taste for hits, and it argued that making available online more obscure titles would level the playing field. Between this study and the research of Professor Elberse, we have evidence that popular titles now represent a greater share of sales and merely making songs available online does not put an artist at an advantage.

Posted by Glenn at 1:48 PM | Link
|Long Tail|

After 24/7 Entertainment's Frank Taubert said at the Popkomm conference that only 3 million out of the service's 4.5 million tracks had ever been downloaded, eMusic, the ultimate long tail retailer, went on the offensive. The company sent around a press release with a statement from the Managing Director of eMusic Europe, Madeleine Milne . "Three-quarters of eMusic's entire four million track catalogue sells at least once every year, or to put it another way, we sell more than 50% of our catalogue at least once every quarter," she said.

But Milne didn't exactly refute Taubert. Many of 24/7's clients offer mobile services, and mobile doesn't lend itself to music search and discovery as well as online resources. This was the point made by Jupiter's Mark Mulligan and the point I echoed in my post yesterday. Milne's statement echoes it as well. "Music discovery on mobile devices may not be supporting long tail sales," Mile said, "but the new digital music consumer is web savvy, and turns to social networks, blogs and the web to find out about new music."

OK, great. So we all agree on it. Mobile discovery on mobile devices do not lead to trips down the long end of the tail. eMusic may get all sorts of purchases of bedroom producers and unknown Indian sitar players, but mobile users stick with the more familiar. I imagine it's a function of screen size and the services a device can offer. If anybody ever figures out how to put a music service on a watch, it will be a boon to the Top 40.

If mobile's inability to foster music discovery continues, we will be left with a mobile paradox: Mobile devices are thought to be the future of music, and the future of music is all about ubiquity and selection, but mobile music leads to less choice and benefits the short head over the long tail. That doesn't sound like the future of music to me.

Posted by Glenn at 3:48 PM | Link
|Long Tail||eMusic|

To help people soak in recent academic research and its conclusions, I wrote a paper on re-thinking Christ Anderson's seminal book "The Long Tail." As you may recall, the July issue of the Harvard Business Review has an article written by a Harvard Business School associate professor. "Should Businesses Invest In The Long Tail?" comes to conclusions that are very different than some espoused by Anderson.

Anita Elberse, the author of the HBR article, wrote in a response to a response by Anderson that she hopes business managers' decisions "are grounded not in romantic notions of the impact of technology, but are based on empirical evidence of what is actually taking place." I fear some romantic notions have replaced more realistic notions of why consumers purchase what they purchase.

The main issue, as I see it, is whether or not consumers will use the availability of incredible amounts of content on the Internet to pursue niches more popular titles. Are consumers ditching the hits? Evidence points to a greater market share of hits and a lengthening and flattening of the tail (meaning niche titles are less popular than in years past).

Download: Rethink The Long Tail

Posted by Glenn at 2:05 PM | Link
|Coolfer Paper||Long Tail|

Anita Elberse, the Harvard Business School professor who recently wrote an article -- "Should You Invest in the Long Tail?" -- that questioned some of the assertions in Chris Anderson's "The Long Tail," has posted a reply to Anderson's response.

I find Elberse's research and response to be illuminating and concise. Anderson's response is a bit confusing -- it confused Elberse as well -- and not at all to the point of the HBR article. All three posts are great reading, though, and I hope you find the time to get through them all.

Posted by Glenn at 2:06 PM | Link
|Long Tail|

The Age has a good primer for the upcoming wave of "free is good" marketing advice. Chris Anderson (author of "The Long Tail") offered a preview of his upcoming book (out early 2009), and The Age article gives a lot of ink to Anderson. It points to a podcast I wasn't aware of, an interview with Anderson at EconTalk section of The Library of Economics and Liberty. (It lasts over an hour and gets painfully slow at times, but it's worth a listen.)

At the time, there are numerous examples that freemiums, or free samples of limited music, can have a positive impact on sales of a main product such as an album. Coldplay's free download and free concerts, for example, probably had a benefit on first-week album sales of Viva la Vida, although in measuring the impact one would have to correct for the tons of press the giveaways created.

With gasoline prices where they are, the free debate takes on even more importance. A free music model based on the resulting benefit to touring revenues hinges upon the cost of travel. There have many good articles on the topic -- today the Austin American-Statesman has one called "Bands Running on Empty" and The Oregonian's recent article, "Bands on the Run -- From High Gas Prices." Austin and Portland are cities that exemplify the problem bands face when touring the West. Cities, at least the ones that have markets to support touring bands, are separated by great distances. The less popular band that plays in small clubs for $15 or less per head, the very type that would get the most out of giving away music, the one that has limited physical distribution, is being hit the hardest. And there are a lot more less popular bands trying to make it than there are successful bands that can charge $20 or higher.

Anderson himself brought up a key question in the podcast interview: "Do people value something less because it's free?" Marketers will say that's the case, that a higher price can increase demand. That's true with items like luxury purses and university tuitions. Digital music is a different case.

Anderson also spoke about the difference between marginal cost and average cost, although he doesn't adequately address the issue in a music context. This is an important distinction. A Google or Yahoo can double or triple store space in its free email services because the marginal cost of doing so is zero or very close to zero. But music doesn't fit the server metaphor. While the digital distribution of one additional copy of an MP3 has a marginal cost of zero (or so close to zero it's immaterial), there is very much an average cost that needs to be taken into account. While the distribution is nearly costless, the creation, recording and marketing of that music require expenses. Distribution is but a small part of the value chain. Giving away one song (of an album's ten or 12 songs) is costless and may be a good idea in some cases. Giving away the entire product, on the other hand, may not make sense.

The debate over free music will be important. All parties will need to have their say, and hopefully they will experiment often to see what works and what doesn't. People's expectations that all digital music should be free will be shaped by the debate. Will people want to jump through hoops for free music? (There are costs -- time and frustration -- using services like Qtrax and SpiralFrog.) Will people expect musicians to subsist on other revenue streams to satisfy the public's expectation (some would say right) for free digital music? (Give me my free music and maybe I'll show up the next time you play in my town.) Should the fact that digital distribution is cost efficient take away from the creators their ability (some would say right) to set the prices for their finished products?

I am sure about two things: People will pay for digital music, and free music can be used in positive ways. The two are not exclusive. Finding strategies that embrace both will be a challenge that will take us beyond this decade.

Additional reading: "Free! Why $0.00 Is the Future of Business" by Chris Anderson from Wired magazine, February 2008.

Posted by Glenn at 7:03 PM | Link
|Long Tail|

Chris Anderson, editor of Wired and author of "The Long Tail," has a new article in Wired titled "Free! Why $0.00 Is the Future of Business." (Is "The Long Tail" already outdated? Two years ago the future of digital business was selling more of less. Now its future is giving away items for free. His next article should combine the two and try to explain how free can work with the paid models described in "The Long Tail." This article does not merge into a meaningful way for industries such as music or entertainment. There are only a few anecdotes thrown in to prove his point.)

Yes, Anderson has most certainly downed the Radiohead-Reznor Kool-Aid, but he's right on.

The rise of "freeconomics" is being driven by the underlying technologies that power the Web. Just as Moore's law dictates that a unit of processing power halves in price every 18 months, the price of bandwidth and storage is dropping even faster. Which is to say, the trend lines that determine the cost of doing business online all point the same way: to zero.

Before the Zero Marginal Cost Militia repeats its call for all music be priced at zero, let's think about the free model. Not all music is going to be free -- as in free MP3 downloads -- but free music can increase awareness, build relationships and encourage sales (or other revenue streams).

From the consumer's perspective, though, there is a huge difference between cheap and free. Give a product away and it can go viral. Charge a single cent for it and you're in an entirely different business, one of clawing and scratching for every customer. The psychology of "free" is powerful indeed, as any marketer will tell you. ... The huge psychological gap between "almost zero" and "zero" is why micropayments failed. It's why Google doesn't show up on your credit card. It's why modern Web companies don't charge their users anything. And it's why Yahoo gives away disk drive space. The question of infinite storage was not if but when. The winners made their stuff free first.

last.fm, for example, has free streaming from all four majors and a gaggle of indies. MySpace is rumored to be working on an ad-supported music portal -- it would presumably have free streams but could include paid downloads as well. Promotions such as PepsiStuff give away "free" downloads that were underwritten by Pepsi purchases. Lala.com's model was built upon the premise that free streaming music would be supported by revenue from downloads and CD sales. Free can work.

(Anderson stumbled, though, when he brought up the unrealistic expectation that free CDs and downloads can lead to more ticket sales. It's not a universal model since not all artists can or will perform live. He's going to dangerously raise expectations with that kind of talk.)

He ended with the idea that today's youth don't just want free, they expect free.

It took decades to shake off the assumption that computing was supposed to be rationed for the few, and we're only now starting to liberate bandwidth and storage from the same poverty of imagination. But a generation raised on the free Web is coming of age, and they will find entirely new ways to embrace waste, transforming the world in the process. Because free is what you want — and free, increasingly, is what you're going to get.

No doubt about it.

Posted by Glenn at 10:18 AM | Link
|Long Tail|

Somewhat music-related here...In a list of working papers at Harvard Business School's Working Knowledge, there is an abstract to a paper titled "A Taste for Obscurity: An Individual-Level Examination of 'Long Tail' Consumption." The author, Anita Elberse, a Assistant Professor of Business Administration at Harvard Business School, is looking at DVD rental data to see if interest in popular and niche titles are evenly distributed. Her findings show that renters of niche titles tend to rent hit titles in even greater numbers -- and appreciate the hits more than the niche titles.

"I find that a large share of consumers, particularly those who consume with a higher frequency and concentrate on a narrow selection of genres, regularly opt for obscure products likely not available in bricks-and-mortar stores. Casting doubt on the 'democratizing' nature of online channels, however, I also show that even for consumers who regularly choose the most obscure products, hit products typically constitute the lion's share of their choices. Moreover, reminiscent of the 'double jeopardy' concept, consumers of obscure products generally appreciate those products less than the more popular products."

That last line made me think of people who put a few super-underground titles in their year-ending Top 10 lists so their friends won't think they're just following the bandwagon. If the professor's findings hold true for music, what does that say about Internet radio and the long tail of digital distribution? It would be hard to imagine the "hits" (in parenthesis because the definition is a moving target) will never be overcome by a large collection of obscure titles.

There is no paper available for download at this time, but I'll keep an eye out for it when it surfaces.

Posted by Glenn at 10:13 AM | Link
|Long Tail|

• Apple and the major music groups will meet with the European Commission on September 19 and 20. The parties will discuss the Commission's antitrust accusations regarding territorial issues with iTunes. The Commission would like to see equal pricing across territories and wants shoppers in one territory to access an iTunes store of another territory. Refer to this April 2007 article in the Financial Times for more on the Commission's antitrust probe. (PC Advisor)

• Proving once again that the future of hip hop is marketing, Pharrel will hook up with Moet Hennessy USA for a marketing campaign. Pharrel will do more than lend a name to a brand, he will create the music for seven internet spots and three television commercials. (Billboard.biz)

• The IFPI is pleased as punch that a new Swedish government-commissioned report has recommended that ISPs take greater responsibility for removing infringing material from their networks. From the report: "It is proposed that the law be amended so that Internet Service Providers can be ordered, under penalty of a fine, to take action such as terminating the contract of a subscriber to prevent continued infringement using the Internet Service Provider’s services." (IFPI)

• Apple confirmed that the upcoming iPod touch will not download MP3s from web pages over the Internet. No biggie. Americans love to sideload. (Listening Post)

• Over at Michael Geist's blog, he has some comments on a Vancouver Sun article on the state of the Canadian recorded music industry and some rebuttals for comments by the CRIA's Graham Henderson. II'll add a comment to one of his points: The fact that Canada's digital sales increased 122% last year versus 65% in the U.S. does not speak to any relative health in Canada. Instead, it means Canada lags behind the U.S. digital growth is slowing. Less mature markets tend to have higher growth rates but lower overall numbers. Canada's rate of growth will drop just as growth dropped here. (Michael Geist)

• There is a big difference between television and music in terms of quantity of content available, but this quote from Tracey Scheppach, Starcom USA's SVP of video innovations, and the interview it's from is still worth examining: "...one lesson that I have learned from TiVo is when a consumer has control over what they want to watch and when, the long tail is just not that long." (Ad Age, via Techmeme)

• The Bay Area mourns the closing of Mill Valley's Village Music. (SF Chronicle)

Posted by Glenn at 7:32 AM | Link
|Apple||Long Tail|

Bear Stearns recently released a report on user-generated content (UGC) titled "A Longer Look at the Long Tail." It was a follow-up to a report issued last November, which I blogged about at the time. UGC is "not likely a fad" and that the increase in video content supply "could lead to lower user satisfaction." The report has 39 pages, if you have the time to read it. In short, Bear Stearns believes the long tail can indeed be monetized and that the value is shifting to the middle of the supply chain. Creators and users will get far less from long tail UGC that the portals will get.

At his Long Tail blog, Chris Anderson goes into the issues the report addresses. He pulled out this quote on the staying power of UGC, which I'll post as well:

"Some investors remain skeptical that UGC is more than a passing fad. If we define UGC as page views only from sites such as Myspace.com, Facebook.com, Youtube.com, Wikipedia.org, Blogger.com, and Digg.com (which is quite conservative), we estimate that UGC now accounts for 13% of total U.S. Internet traffic, up from 0%-1% in 2004. Based on these statistics, we submit that UGC is here to stay."

One point of contention: Bear Sterns and Anderson look at revenue of Hollywood studios' average box office receipts from 2000 through 2006. Their volatility, we're told, is proof that nobody has been able to constantly create quality content. I would argue that looking at year-to-year changes in such an industry is too small a time frame. It's like looking at just any one quarter for any other industry. (Consumer product companies are unlike companies in the entertainment business. The former develops a cash cow and rides it for sometimes decades. Entertainment companies are in the business of constantly launching new products, and their cash cows have a shorter life span.) Another point is that when looking at short head (the big studio movies) vs. long tail, it would be better to group all studio revenues together and not look at variations within the peer group. There may be just as many hits but they're distributed differently from year to year. The point is that big budget movies are getting made, and people are watching them.

The thrust of the paper is dead on, though. Content aggregators stand to benefit the most from UGC as well as the massive flood of music and video one professional notch above UGC.

Posted by Glenn at 1:15 PM | Link
|Long Tail|

Back in August the reviews of Chris Anderson's The Long Tail were popping up and the Wall Street Journal's Lee Gomes had a critical review of the hit book. Gomes called out the "current popularity of Web utopian fantasies about the way sales of niche products can rival those of hits." Looking at the economics of the long tail, I wrote that aggregators and distributors, not creators. Makes perfect sense. Somebody, something must handle the incredible volume of content. Per-transaction margins will be low, but volume will be high. Pennies to a creator are just pennies. Aggregators, though, will get mountains of pennies.

Now a Bear Sterns analyst has come out with a report that says the sweet spot of the content supply chain may be in the middle. The tile of the report is "The Long Tail: Why Aggregation & Context and Not (Necessarily) Content Are King In Entertainment." It tracks the changes in content distribution over the years (TV stations, cable, Internet).

"Most believe the balance of power is shifting to content owners," it says. "However, technology is altering economics of content creation as well." Lower costs and greater flexibility in the creation, storage and distribution of content have resulted in a proliferation of content. The result is what Bear Sterns calls an "economy of abundance." Filters are "required to connect users with content that appeals to their interests" and aggregators such as Google, AOL and Yahoo are tagged as viable competitors, but lack a detailed long tail strategy, says the report.

Posted by Glenn at 12:16 PM | Link
|Digital Distribution||Long Tail|

• Another 8,000 lawsuits were filed against file-sharers in 17 countries. Will it stop file-sharing? Not by a longshot. Will the trade groups stop suing file-sharers? Not a chance. Talk against piracy is cheap. Trade groups know they have to act. (Read Reuters article at CNNMoney)

• A judge dismissed Entercom Communications' motion for dismissel in a payola lawsuit. This lawsuit was Eliot Spitzer's first against a radio company. It now moves to the discovery phase. (Read article at Billboard.biz)

• Universal Music Group moved 250,000 units of 3,000 digital tracks that were released from the company's archives. The press is really eating this up. What if they did the math? I'll rephrase: Universal Music Group sold an average of 8.33 units each of a group of digital songs it released from its archives. Doesn't sound so spectular does it? And that was over seven months. That's just over one download per track per month. On an aggreggate level it's fine, but individual royalty checks won't change much. I've said it before: The Long Tail wil benefit aggreggators, not individuals. (Read article at News.com)

• Used CD 2.0 website Lala.com is currently selling all its CDs at wholesale cost. (Yes, Lala.com does sell brand new CDs in addition to facilitating the trading of used CDs.) It's been a while since Coolfer wrote about Lala.com. A revisit post is forthcoming.

Posted by Glenn at 7:36 AM | Link
|Legal||Long Tail||Payola||Universal Music Group|

• One thing that's all over the news today -- on the Internet, at least -- is MP3tunes' Oboe Free. Oboe is the company's online music locker that allows users to listen their files from anywhere. Oboe Free is a free version of Oboe, which costs $19.95 a year for 2,000 songs or $39.95 for unlimited songs. Users can store up to 1,000 songs at no cost with Oboe Free. If you're interested in signing up, go to the Oboe registration page and select the free version. Using Obie requires downloading a program that will sync up the music on your computer to your Oboe locker. It's a process that takes one hour for every 100 songs.

• David from Digital Audio Insider moderated a panel at the Bandwidth Conference and wrote about the long tail discussion at this blog. He asks the important question: "Is the pie getting bigger?" Much of the long tail discussion I've read assumes a zero-sum game in which the outer-laying artist takes from the superstar. What I'm waiting for is the new digital technology that will increase the size of the pie. Until then new technology will simply allow people to replace one behavior with another without increasing the overall value of the market. (Digital Audio Insider)

Posted by Glenn at 5:21 PM | Link
|Digital Music||Long Tail||Music 2.0|

• Rock band Keane (Interscope) has postponed its fall North American tour while its singer undergoes rehab. (Billboard.com)

• A vote for "flexible and variable pricing" from the chairman of Viacom and CBS, Sumner Redstone. (Digital Music News)

• RED Distribution has added two labels: Uprising, formerly with Koch, and I Surrender. (Billboard.biz)

• Since YouTube is always in the music press lately, here's one for you: Sony Corp placed its bet on the future of the Internet by purchasing Grouper, a website that hosts user-generated videos, for $65 million. Question: Will the company find a way to fold that into its overall strategy that could benefit its hardware and media divisions? (NY Times, via 12F)

• You don't know what you're missing, Los Angeles: In the spring survey, country music extended its streak as king of the top Arbitron markets. Spanish formats had the biggest gain. (Radio Ink)

• Something for you long tail nuts: A transcript of a Bandwidth Conference panel featuring "The Long Tail" author Chris Anderson and folks from Yahoo! Music and Rhapsody. (37signals, via Digital Audio Insider)

• Something for you Leslie Gore nuts: Digital Rights Agency has inked a deal with K-tel International to digitally distribute its songs worldwide. (Press Release)

• Worth reading: A thread on Ryko's recent conference in New Orleans and ADA's turnkey digital distribution solution the company will implement. (The Velvet Rope)

Posted by Glenn at 9:08 AM | Link
|Digital Distribution||Distribution||Long Tail||Radio||Universal Music Group|

Add music industry analyst and consultant Barry Sosnick to the list of long tail naysayers. Sosnick wrote a piece titled "Long Tail Will End In Heartbreak" for The Register that argues too broad a product assortment can confuse shoppers and that low barriers to entry demand either price competition or some other way to compete. He summarizes at the end:

"While 'Long Tail' is the newest catchy business phrase, but is just another name for a market segmentation using broad assortments. It is not the ultimate competitive advantage. In fact, reliance on broad assortments can confuse and alienate consumers and inadequately differentiate a retailer from its competitors."

For additional reading, go to an article Sosnick linked to in the article, "Searching For Seach Clues" at MIT Sloan Review. It argues that after the elimination of search costs "fierce price competition is likely to ensue, making brand largely irrelevant and driving prices close to marginal cost." In context of the music business, long tail search costs are far from zero. At some online stores, the search process is a miserable experience that requires much scrolling through lists of albums. Such search costs encourage a user to migrate toward more coherant content like recommendations and top sellers. Those needles in a haystack? Too hard to discover.

Posted by Glenn at 11:05 AM | Link
|Digital Distribution||Long Tail|

Working Knowledge has a post about The Long Tail by Harvard Business School professor James Heskett that poses some questions and asks readers for feedback.

"In the Long Tail, money is made by such things as avoiding inventory, producing to order, letting customers do the work, pricing creatively and flexibly to various customers, utilizing a variety of distribution methods, sharing information, trusting the market to do your job, and understanding the 'power of free' combined with money-making services or products. Does the Long Tail represent what some would call a 'paradigm shift'? Who will the Long Tail benefit most: consumers, producers, or intermediaries?"

Coolfer answered that particular question just the other day. (Intermediaries.)

There are some good comments thus far; the post is open for comment until August 31st, 2006.

Posted by Glenn at 2:30 PM | Link
|Long Tail|

The Wall Street Journal's Lee Gomes isn't buying the basic tenants of Chris Anderson's era-defining book The Long Tail, which is seen in some quarters as a guidebook for the digital music era. His latest piece about the book by the editor of Wired Magazine picks apart Anderson's arguments and shoots to bring down the "current popularity of Web utopian fantasies about the way sales of niche products can rival those of hits."

Though The Long Tail is heralded left and right as a defining book, Gomes points out many examples of how hits are still hits and consumers aren't giving any greater market power to small book publishers and record labels. One book publisher told him his company's sales to Amazon.com mimic its sales to brick-and-mortar stores. Netflix, he estimates, gets 30% of its rentals from just 0.8% of its titles. Indie songs accounted for 15% of Rhapsody streams, which is roughly the same as indies' share of the CD market. (That may not bode well for music recommendation services. There is a huge overestimate on the number of people who want to find new music. Most people want familiar music, not new music.)

Businesses, he continues, should be wary of expanding production in hopes of riding the long tail to greater sales. This makes perfect sense. To me, the long tail benefits aggregators and distributors, especially those who deal with digital goods, who have the scale to add titles at almost no incremental cost. The cost to a digital music store of adding one more song to the catalog is virtually zero. But what about content creators? The cost to create that song isn't as low as the cost of distribute -- and that's to say nothing about the cost of marketing. There are exceptions, such as if it's an old song that is seeing the digital world for the first time. (Even then there's a cost associated with converting from analog to digital and readying for sale.) For new songs the cost to bring to market is usually not as low. A distributor can afford to add songs and sell less of each, but creators shouldn't have the same goal.

Posted by Glenn at 12:54 PM | Link
|Books||Long Tail|