Michael Rabjohns sent me a note informing me of an article in the July Harvard Business Review written by Anita Elberse.  Elberse is an associate professor of business administration in the marketing department at Harvard Business School.  Her article leads off with a portrayal of Grand Central Publishing, a company that lists 275-300 books each year in its catalog and identifies two (my emphasis) for which it will pull out all the stops in marketing.  Grand Central pursues a blockbuster strategy for which Elberse gives credit to economists Robert Frank and Phillip Cook (an economist at my alma mater, Duke University) for endorsing in their 1995 book, The Winner-Take-All Society. 

Elberse states that this strategy conflicts with the long tail theory espoused by Chris Anderson.  She reviewed sales data from Rhapsody (used by Anderson) and Quickflix (an Australian DVD rental company).  She states that the data shows that seldom rented DVD’s shift into the tail and their usage doubles over a five year period of time.  However, the number of DVD’s with no usage quadruples over the same period of time.  She states that the tail is becoming longer and flatter, not bulkier.  During the same period of time, the number of titles in the top 10% dropped by 50% which demonstrates that the “winner-take-all” strategy is winning.  Elberse’s data for music tracks confirms a similar trend where the tail lengthens, but grows flatter and fewer tracks are in the top 10%.

Elberse concludes her article with recommendations to producers and retailers.  The only finding where she directly supports the long tail is in her recommendation to retailers stating that if the goal is to cater to the heaviest volume customers, include a diverse assortment of titles since her research indicates that the buyers of the obscure titles tend to be the heaviest users of the online services.

Anderson doesn’t dispute Elberse’s research; in fact, he praises it.  But he claims that Elberse’s portrayal of the “head” of the tail at 32% and the tail at 68% confirms his theory about the size of the tail.  Elberse responds to Anderson’s comments by stating that he is inconsistent about the definition of size of the tail and head and that consistency in definition would verify her findings that the tail is bigger but flatter and that analysis has repercussions for producers and retailers in the strategy that they should pursue.

All in all, this is a great debate and I thank a reader for pointing this out to me.  I’m sure the dialogue will continue.  I think Anderson and Elberse are accurate in their claims and I like Elberse’s caution to certain businesses, particularly producers.

Chris Anderson, Editor-in-Chief of Wired magazine, published The Long Tail: Why the Future of Business is Selling Less of More in 2006 after writing an article in Wired of a similar title back in 2004. This is another “must read” book that provided me with a number of insights into the scalability of businesses on the internet. Using a few examples, like Amazon.com and Rhapsody, Anderson demonstrates how the value proposition in retail business shifts when retailers utilize the internet. Typically, stores stock the items most likely to turnover (sell) the fastest since physical shelf space is at a premium. The better your selection, the higher your gross revenues. Anderson shows that the long end of the traditional demand curve is longer than many realize. An example of contrasts is the demand for music downloads.

The largest music retailer in America is Wal-Mart. The average Wal-Mart stocks 4,500 unique CD’s. The top 200 albums account for 90% of Wal-Mart’s CD sales. Rhapsody has approximately 2 million tracks available. The top 25,000 tracks on Rhapsody are the equivalent of Wal-Mart’s 4,500 CD’s. Sales of the next 75,000 tracks (from 25,001 to 100,000) are nearly 25% of Rhapsody’s downloads and there are sales beyond that. From 100,000 to 800,000, the downloads are about 15% of Rhapsody’s totals. According to Anderson, every time Rhapsody adds a track, there’s at least one download a month indicating that someone in the world is interested in purchasing even the most esoteric track. Rhapsody doesn’t have to give up shelf space to add these tracks. Once they’re digitized and added to the virtual inventory, they’re available to the consumer.

Anderson concludes by saying that successful internet companies will adhere to two rules: 1) make everything available and 2) help me find it. I keep his book on the shelf nearest my desk and I subscribed to his blog’s RSS feed.