"Long Tail Effect" Doesn't Work As Advertised, Say Wharton Researchers 82
Death Metal writes "In a working paper titled, 'Is Tom Cruise Threatened? Using Netflix Prize Data to Examine the Long Tail of Electronic Commerce,' Wharton Operations and Information Management professor Serguei Netessine and doctoral student Tom F. Tan pull information from the movie rental company Netflix to explore consumer demand for smash hits and lesser-known films. Netflix made its data available as part of a $1 million prize competition to encourage the development of new ways that will improve its ability to introduce customers to lesser-known titles they might find appealing." In short, the researchers say that the Long Tail effect described by Chris Anderson is much less important in the real world than popularly held. Says the article: "The key difference between the opinion of [Anderson's] book and the study by Wharton researchers is how they define 'hits' and 'niches.' In the book, Anderson focuses on the definition of hits in absolute terms such as the top 10 or top 1,000 products, while Netessine and Tan argue that, to take growing product variety into account, one has to define popularity in relative terms, such as the top 1% or top 10% of products, to properly assess the presence or absence of the Long Tail."
Missing the point (Score:4, Insightful)
The long tail doesn't threaten those at the top any more than it isolates those at the bottom. It only describes the shape of the market which necessarily has only a few specific market products which are used by the majority and the rest of the products with very few customers in the "long tail". It's a market definition, not a competition definition.
You can cut the tail off of a gecko at any point, but it doesn't mean that somehow the tail can exist without a fat end and a thin end. Since the tail is simply the appendage attached to the abdomen, wherever it is attached defines its fat end, and where it ends is the thin end. Even if you cut the tail off completely, all that you've done is stimulated the tail regrowth reflex.
This is completely moronic (Score:3, Insightful)
Re:Missing the point (Score:5, Insightful)
But the shape of the market is exactly the point. In a competitive market profit margins are very thin and a relatively small difference may mean life or death to a company. In the entertainment industry we often see an effect where the biggest productions often seem to struggle to break even, while relatively small investments may bring huge profits.
Defining a "hit" as one of the top ten or top 1000 or any absolute number is stupid. It reminds me of a political joke in the Soviet Union, where the result of a race between two athletes, a Russian and an American, was reported in the press as "the Russian came in second while the American was next to the last". In electrical and electronics engineering threshold values are often defined as the point where the power is one half of the maximum, the so-called "-3 dB" points.
Long Tail Wikipedia Link (Score:3, Insightful)
I just want to thank the submitter/editor for providing the link to Wikipedia for those of us who don't know what's meant by the Long Tail. As it happens, I do know what the "long tail" is, but one of the more tiring aspects of SlashDot is the number of narrow articles that hit the front page that wholly lack any sort of description.
Re:This is completely moronic (Score:1, Insightful)
If you add an insignificant product to the end of the tail, it obviously increases the proportion of market share of the first X% of products. That's simple math!* If your model of the "long tail" completely fails in the simple case of adding a once-purchased product, maybe your model sucks and Chris Anderson's model was more useful.
But that DOES matter if you are buying into CA's conclusion that the "Long Tail" is important from a business perspective. Nobody I've read disputes that the total sales of more obscure DVDs (for example) may be quite significant in aggregate compared to the sales of blockbuster titles. However in order to apply a "Long Tail" business model you are going to have a truly vast quantity of inventory that will mostly move very, very slowly (maybe one or two copies per year of each SKU). You will have much greater warehousing and finance costs between the date you purchase any particular item and when it gets sold.
In order to make a successful, profitable business it is important that the "Long Tail" not only exist but be sufficiently "heavy" to offset these increased costs. Talking about the relative market share of the top X% and the rest is relevant to what sort of business model you would want to choose. Talking about the "top 10" vs the rest is irrelevant if 10 is not a realistic number in that marketplace.
Re:This is completely moronic (Score:2, Insightful)
Please don't use DVDs as an example of long tail economics. Having a physical copy of every item in the tail will in most cases completely negate any kind of profit. The whole point is that with virtual products (or print on demand) you can have much, much higher storage density with limited cost. Only then can you afford to only sell two items a year.
Re:Missing the point (Score:4, Insightful)
Defining a "hit" as one of the top ten or top 1000 or any absolute number is stupid. It reminds me of a political joke in the Soviet Union, where the result of a race between two athletes, a Russian and an American, was reported in the press as "the Russian came in second while the American was next to the last". In electrical and electronics engineering threshold values are often defined as the point where the power is one half of the maximum, the so-called "-3 dB" points.
There are 17,770 movies in the Netflix Prize training set. 1000 movies account for 5.63% of movies for the entire dataset. This 5% account for 63% of all rentals. If you use your threshold of "half of the maximum," then you have the top 100 movies. More to the point, your threshold definition using decibels is predicated on the data being normally distributed, and this data conforms to a power distribution, most likely Zip-Ian.
Getting back to your point of defining a "hit" based on profitability, that too is poor way of defining it. It's much easier for a very cheap film to make multiples of it's budget in revenue, but still no one sees it. Number of viewers has always been the traditional way of defining a hit. Revenue is just proxy for that.
Re:Heavy tail distributions are dangerous beasts (Score:3, Insightful)
Generally they're talking about Pareto or power-law distributions, which aren't quite as degenerate as, say, the Cauchy distribution with no higher-order moments. You're right that they should check for the existence of the moments they care about; I presume they do, but I don't know.