Excellent HBR piece challenging the Long Tail

Anita Elberse, a Harvard Business School associate professor, has a really interesting article in the new Harvard Business Review that analyzes some Long Tail data and challenges some of the theory's predictions. Based on Rhapsody music data and DVD rental data from an Australian Netflix clone called Quickflix, she concludes that the blockbusters are not losing share to the long tail of niche products in those markets; indeed, they're gaining it. She writes:

Although no one disputes the lengthening of the tail (clearly more obscure products are being made available for purchase every day), the tail is likely to be extremely flat and populated by titles that are mostly a diversion for consumers whose appetite for true blockbusters continues to grow.

That's surprising (not least to me), and now that I've had a chance to give the paper a quick read, let me jot down some quick thoughts on why Elberse (who I collaborated with on some of my research and respect highly) would come to such different conclusions than I do.

Let me start by saying that the paper looks rock solid and I'm sure her analysis is accurate. But there is a subtle difference in the way we define the Long Tail, especially in the definitions of "head" and "tail", that leads to very different results.

The best example of this is in what she describes as a growing "concentration" of sales around a relatively small number of blockbuster titles. In the Rhapsody data, she finds, the top 10% of titles (out of more than a million in that data sample) accounted for 78% of all plays, and the top 1% account for 32% of all plays. That sounds pretty concentrated around the head, until you reflect, as she notes, that "one percent of a million is still 10,000--[...]equal to the entire music inventory of a typical Wal-Mart store."

This is a good moment to remind everyone of the normal definition of "head" and "tail" in entertainment markets such as music. "Head" is the selection available in the largest bricks-and-mortar retailer in the market (that would be Wal-Mart in this case). "Tail" is everything else, most of which is only available online, where there is unlimited shelf space.

So in the data she cites, the head of the online music market represents 32% of the all plays, and the tail represents 68%.  That's certainly no challenge to the Long Tail theory; indeed, it's even more tail-heavy than the data I cited in my book (probably because I used a more generous estimate of 50,000 tracks for Wal-Mart's inventory).

She then looks at Quickflix data. Here the top 10% of DVDs accounts for 48% of all rentals, and the top 1% accounts for 18%. "The concentration [of sales around the blockbusters] is not as strong as Rhapsody, but it's still substantial," she writes.

But here, too, the use of percentages misleads. Quickflix had 18,000 titles at the time of the research, compared to the average Blockbuster's 3,000 titles--there's only a factor of six between their inventories, as opposed to a factor of 100 in the Wal-Mart/Rhapsody comparison. If you look at her chart, you'll see that the top 3,000 titles (ie, the amount equal to Blockbuster's inventory, or the "head") accounts for 70% of rentals and the "tail" accounts for just 30%, making it more concentrated on the head than Rhapsody, not less. (BTW, I calculated almost exactly the same split for Netflix in the book.)

My point is not to suggest that Elberse is wrong and that I'm right, it's only to point out that different definitions of what the Long Tail is, from "head" to "tail", will generate wildly different results.

Anyway, it's getting late and I just wanted to highlight a few other interesting data points and conclusions from her article:

• Much of the paper is about consumer satisfaction in the head vs tail. In the Quickflix data, she says, "customers give lower ratings to obscure titles...it is a myth that obscure books, films and songs are treasured. What consumers buy in Internet channels is much the same as what they have always bought." That may be true for the specific example of the Australian DVD data, but it is not clear from the paper why she feels able to extrapolate that to all Internet commerce.
• The heaviest DVD renters were the most likely to venture into the tail; light consumers largely concentrated on the hits.
• In music, of the 2.4 million digital tracks sold in 2007 in the US (most of them through iTunes) 24% sold only one copy and 91% sold fewer than 100 copies.

And there are pages and pages of other nuggets like this. It's an excellent article, and although I don't agree with all the conclusions, I'm delighted to see research of this rigor on the topic. Recommended.

Comments

[Jean-Marie Le Ray]

Well, it could be a good bet to Turning the Tail... http://adscriptum.blogspot.com/2008/06/turning-tail-from-long-tail-to-big-tail.html
Jean-Marie

[Jean-Marie Le Ray]

Sorry, Turning the Tail

[David Cushman]

Chris, Here's some illustrations I think explains why the 'Hit' MUST be worth (proportionally) less in a networked world.
Here's a link to the slide deck:
http://fasterfuture.blogspot.com/2008/06/why-hits-must-have-less-value-in.html

[Amal Dorai]

I agree that the analysis is flawed because it is based on percentages rather than on absolute numbers.

The same technological progress that has enabled the Long Tail has also lowered the cost of producing media entities such as songs and movies. I recently had a few ideas for an electronic music beat; twenty years ago, I would have had to borrow or buy a Roland drum machine to test it out, but instead I just downloaded a program called FruityLoops and had a song going in under two hours.

Thus, Long Tail markets are simply much bigger today in terms of number of members, and the top 1% will constitute a much bigger number of songs. Yet the memory capacity of humans, and the number of memes or acts who can be considered "pop culture" has not changed nearly as dramatically. The Long Tail is real; the Madonnas and Michael Jacksons of the 80's give way to the Tay Zondays and Obama Girls of the 2000's.

Elberse's analysis breaks because it mathematically incorporates what is culturally irrelevant; the large mountain of unknown garbage at the bottom of the distribution, which nobody knows about or consumes.

[nordsieck]

"Much of the paper is about consumer satisfaction in the head vs tail. In the Quickflix data, she says, "customers give lower ratings to obscure titles...it is a myth that obscure books, films and songs are treasured. What consumers buy in Internet channels is much the same as what they have always bought." That may be true for the specific example of the Australian DVD data, but it is not clear from the paper why she feels able to extrapolate that to all Internet commerce."

This is particularly true for products, part of whose value is novelty, like wedding dresses, women's jewelry or any other type of fashion.

[Los Angeles Real Estate]

I agree. The information seems flawed.

[randomwalker]

I think you're being way too generous on the author! It shouldn't even need pointing out that looking at percentages misses the entire point of what the long tail is about. In markets where the barrier to entry for content producers drops to zero, percentages are not a meaningful way to measure *anything*, let alone head and tail.

[parocks]

"Hit products are also liked better than obscure products. It is a myth that obscure books, films, and songs are treasured."

or

"Products that are liked better are more likely to be hit products than products that are liked worse. It is a myth that poorly liked books, films, and songs are popular."

Which makes more sense?

Are these books, films, songs permanently stuck in their decile?

A completely indie release of a film, record comes out with no advertising budget and goes right to the end of the long tail. However, those who like niche products, obsurities, etc., view it, hear it, buy it, and give it good reviews, if its good. Others respond to the good reviews and view, hear, buy the product. If it's good, they like it, they give it good reviews. After this happens a few times, The product becomes a hit, no longer obscure. Why? Because people were driven to consume the product by the good reviews.

Bad reviews cause obscurity, obscurity does not cause bad reviews.

[Gustavus]

This debate is only becoming more interesting (and relevant). For anyone curious to know how the Long Tail theory can be applied to the writing biz, I've just posted on its economics and utility when it comes to authoring history books (itself a bit of a Long Tail-y niche subject . . . )

http://historyman.wordpress.com/2008/06/29/how-to-write-a-history-book-part-2-special-long-tail-edition/

[moss]

Fascinating:

Will the market crash or rally?

http://barrons.com/story/market_crash_or_rally?eder44e.htm

The Saudis have already said they believe current demand is being met, despite the high prices — suggesting that

high prices alone might not be enough to warrant the increase in supply.

[Biofuelsimon]

The only way to tell which view is right (and odds are both are not entirely wrong right) will be to live the change. I've always worked in a niche. The internet has made it more visible. Its also made competitive niches more visible. In the short term we're working like made to make our niche stand out. Will it be enough to enable our business to transition from paper to online fully?

[oggy]

It is still not easy for consumers to tell which tail is gold for them but not junk. A more sophisticated tool that provides the solution would dramatically change the result of this kind of study.

[Erik]

When you do a 'statistical' analysis, you should never forget about the context of the data collection. In the case of data for product purchases, you should never forget that, in most cases, the data collection process is biased by the quality of the recommendation engine (and measuring this quality in a formal way is a very tricky and open problem). A poor quality recommendation engine will allow only 'hard core' users to find out what they are looking for, and, guess what, statistical counts will show you that only hard core and heavy users buy both blockbusters and niche products. If your start using a better (choose your metrics here...) recommendation engine, you may very well see that the relative weight of the long tail increases, and the purchases distribution will be shifted to the tail.
So, in a nutshell, this article, starting from the same data, could conclude that the recommendation engines of the two proposed examples could be improved...
Besides the fact that I am a firm believer that going long tail without a proper recommendation engine is useless (and potentially harmful), what I wanted to insist on is that, when you want to confirm, through the numbers, a statistical hypothesis, you must show that your hypothesis is better that other alternatives that could explain equally well the effects, and this has not be done in the article.
The other remark is that drawing general rules by looking at two examples is (statistically?) very questionable (dangerous?)...
Finally, all the remarks on the economics of the long tail are OK (cost of niche products should be minimized, etc.), but they are not new.

[Joe Chiang]

I am working on a project whereby I am making and posting a new paper toy every day for a year, and providing the PDF for anyone who wants to make the toy to download for free.

Here's the data I collected since I started the project more than a month ago:

Number of items: 40
Total downloads: 30,263
Total downloads for the top 10% (top 4): 8,976 (which translates to close to 30%)

I will update it at the end of the one-year project and see if the long-tail theory works in my case.

[Chris Skinner]

Chris

Since your book came out, I’ve been gathering thoughts on the long tail in banking and spotted Anita Elberse's thoughts, so blogged a quick note on Sunday about it: http://www.thefinanser.co.uk/2008/06/arguing-about-t.html

Result is that I was asked to explain the long tail in banking which you can find here:
http://www.thefinanser.co.uk/2008/07/the-long-tail-o.html

Bottom-line: I also disagree with Anita and believe the idea of ‘free’ and ‘long tail’ will fundamentally change banking and finance as much as any other industry or market that depends upon technology and the internet for its business model.

[Denis Hancock]

Interesting discussion.

I agree with the majority here in that the percentages aren't neccesarily the best way to approach the problem in markets where production and distribution costs are approaching nil (i.e. digital music). At the same time, I'm not sure the other definition of "head" - tied to availability in bricks and mortar outlets - makes that much sense anymore either.

I was kicking around ideas for a bit until a rather obvious sounding idea jumped out at me - what lies between the "head" and "long tail" is the "body". Could this body, at least at a conceptual level, be the most important of all?

For example, in the Rhapsody data, the top 1% account for 32% of the sales, and the bottom 90% 22% - leaving 46% for that 9% jammed in the middle. Might this group be interesting to look at in their own right?

I realize there's still the same % problem, and I'm not sure how you would come up with the proper dividing lines, but there's something compelling to me in regards to carving out a middle ground.

I posted more on this if you follow the URL link below.

[gregory]

now is when you need an itegral spiral dynamics model

she tracked lower consciousnesses, you wrote about higher

pretty simple .. the hard part is embracing the fact that not all humans are on the same level

[Michael F. Martin]

Chris,

Had you considered that the demand function could be modeled as a cumulative distribution function corresponding to a poisson distribution in the frequency of consumption?

On my view, supply and demand can be modeled as cycles, and consumption of information is higher-frequency, but still poissonian in shape. Hence, there is not algebraic, but exponential decay at the long tail.

http://brokensymmetry.typepad.com/broken_symmetry/2008/07/the-long-tail-r.html

[Phil Leigh]

The Harvard seems to ignore the point that consumers are naturally attracted to stores with a large inventory. Moreover, this is not an Internet-era phenomenon. For example, consider how the "supermarket" led to obsolesce of the neighborhood grocery store. Another example is the impact of Barnes & Noble and Borders on the neighborhood book store.

Essentially consumers are drawn to the stores with a wider variety (longer tail) of merchandise even though the bulk of the money they spend might be on the most popular items. They choose to go to the store with the bigger inventory because it permits them to also get the hard-to-find items. This applies to Internet retailing as well. We may spend most of our money on the popular books at Amazon.com, but we choose the store (partly) because we know they will have most any book that might be on our shopping list.

http://insidedigitalmedia.com/a-new-twist-on-the-long-tail/#more-127

[Greg]

Elberse's response to Chris is up:
http://conversationstarter.hbsp.com/2008/07/the_long_tail_debate_continues.html

[Mike]

she really addresses it point by point. what is the defnition of the long tail? the largest retailer?

[Chris Anderson]

@Mike:

Yes, I've always used the inventory of the largest retailer as the definition of "head". But to be honest any absolute number that's tied to traditional blockbuster channels will do, as long as you hold it constant. What will NOT do is to use percentages or deciles (which is the same thing), because that means the dividing line between head and tails moves as the number of products in the market grows, making it impossible to do any head vs. tail analysis over time.

It's been the one constant problem with her research, and I'm afraid it tends to obscure her many other excellent points and data.

[Mike]

i think she is pretty convincing. Lee Gomes had a good piece this morning in the WSJ too.

[Chris Anderson]

@Mike

Could you explain which part you found convincing? (Or, alternatively, which part of my answer you found unconvincing?) Happy to drill down to help explain the differences.

[dJ Chang]

Chris,

The economic premise of the Long Tail is that the Internet allows more people an opportunity to distribute their product, service, or content -rather than the conventional wisdom hypothesis of increased concentration among top brands. Quantified success of top brands on the Internet does not negate the positive impact on the Long Tail.

Ironically, is Ms. Elberse not a member of the Long Tail? Has she used the Internet to gain distribution for her content? Is the discussion about the Long Tail, an example of discussions that may never see the light of day without the Internet?

Newspapers and magazines have had difficulty maintaining share on the Internet. Social networks and informal blogging continues to claim increasing share.