Recently I wrote about how killing a product can be the best choice for an efficient portfolio management. The idea of investing heavily on the blockbusters and limiting costs and effort on the niche products is consistent with the Pareto 80/20 rule and is the foundation for a lean product management approach.
That said,some schools of thought go in the opposite direction. When Chris Anderson formulated his Long Tail theory, back in 2004 as a post on Wired, which became a book 4 years later, he proposed a controversial idea. The idea is selling less of more products. At the time this was a completely new fascinating paradigm. In brick and mortar retail, shelf space is limited, therefore shops must focus on a small number of as big as possible hits. They stock few products and they sell significant quantities of all of them. If anything does not sell it gets quickly replaced by another product to avoid wasting precious shelf space. This is commonly known as the blockbuster strategy. In a nutshell, it is relying on a small number of carefully selected products and maximising sales for all of them.
This strategy has worked for centuries without substantial changes until the advent of digital retail. Online stores offer virtually unlimited shelf space and carrying unimaginable levels of stocks becomes suddenly feasible. This is particularly valid for digital content (movies, software, e-books, etc.) rather than physical goods, for which warehousing still carries some cost burden. However, while storing digital products is undeniably cheap these days, e-commerce also allows efficient distribution and warehousing, inaccessible to normal retail shops. This obviously drives down the cost of keeping stock. Moreover, modern online marketing strategies make possible to give proper visibility to a big number of product like never before.
In this novel scenario the Long Tail becomes fascinating and disruptive. The elimination of shelf space and stock keeping constraints make a completely new strategy entirely possible. You can now effectively sell a much bigger number of products. Even if each one contributes to a small revenue percentage, the virtually unlimited scalability of online businesses makes the concept extremely attractive. So attractive that the Long Tail made it to Business Week’s “Best Ideas of 2005”. Since then the idea got undeniable traction. Businesses like Netflix, iTunes, Amazon Prime Video and Spotify leverage the strength of their immense catalogs. It looks like the Long Tail can finally defeat the Pareto rule. Or does it?
Digital products are the natural habitat for the Long Tail, for the above mentioned supply chain and stock keeping concepts. With moderate costs, the tail can be of infinite length adding massively to the total revenues even if pretty thin. If it wasn’t for the fact that this does not seem to happen consistently. In an eye-opening article on HBR, Anita Elberse published her research on digital music and video consumption. The expectation was a flatter sales distribution curve, with top products accounting for a decreasing proportion of total sales. What the study findings suggest instead is the contrary. Numbers speak for themselves, 15-20% of products account for 80% or more of total sales. The longer the tail, the more successful the top products become. This is a winner-take-all scenario, not what postulated by the Long Tail theory.
Citing from the article, “rather than bulking up, the tail is becoming much longer and flatter”. The Long Tail seems to make niche product even more niche and blockbuster even more blockbuster. Interestingly, what the long tail does is attracting what Anita defines as “heavy customers”, after a wide assortment of products. Heavy customers are attracted by the broad offering, however they seem to contribute for a negligible push in revenue generated by the tail. As Wharton researchers put it in a study on Netflix published in 2009, “since only a small fraction of consumers constitute heavy movie watchers, it is not surprising that there is weak evidence of the Long Tail effect”.
The data is leading to questioning the Long Tail theory’s fundamentals. Citing from Wharton’s study, “whether or not the Long Tail exists is a fundamental question for decision makers in marketing, operations and finance who face the prospect of further penetration of the Internet channel, which offers expanding product variety and new recommendation systems to help manage it”.
Leaving any philosophical considerations about the theory on the side, I’d like to go back to a concept that emerges from Anita’s article on HBR. In the final remarks, she makes a very important reference to development cots: “strictly manage the costs of offering products that will rarely sell. If possible, use online networks to construct creative models in which you incur no costs unless the customer actually initiates a transaction”.
Shifting the focus from digital to physical products, this becomes even more critical. Assuming shelf space is still an insignificant cost, development and marketing do not come for free. Investing in products that won’t sell is a waste. Moreover removing resources from the more profitable ones is an additional opportunity cost. Even if the Long Tail effect exists, it deals with revenue, not necessarily profitability. For that you need to look at costs and they can kill you. Maybe it’s better to kill a few products instead.