Blockbuster leaves legacy of vital data mining techniques


  • Author: Carlos Alberto Gómez Grajales
  • Date: 15 November 2013
  • Copyright: Image appears courtesy of iStock Photo

Just a few days back, a chapter closed in America's business history, a chapter that had almost 30 years in writing. On Wednesday 6th, Dish Network announced that Blockbuster’s three hundred remaining U.S. retail stores would close. Their stock will be slowly liquidated, leading to an early 2014 closing, but the company’s DVD mailing service, Blockbuster By Mail, will see twilight in about a month.

thumbnail image: Blockbuster leaves legacy of vital data mining techniques

Right now, media and business websites are filled with retrospective stories of how Blockbuster could not adapt to the changes of the market and blew its business away. But here, the point is not the focus on what went wrong the last couple of years within the company, but instead, to focus on what they did extremely right in the beginning, which was in fact, using statistics (as you may have guessed, just because I'm writing about it).

Blockbuster Entertainment Corp. was created by a computer system entrepreneur, David Cook, who taught himself to write computer programs for business applications. Moreover, the video rental giant traces its roots to a former failed company named Cook Data Services (We're already seeing data here!). In the late 70's the power of computers was the way to go for innovative companies, and Cook used his expertise to develop real estate programs in 1978 and then a more refined program to manage inventories for the oil industry, which was booming at the time. After his data company went public, in fact just weeks after the offering, the Organization of Petroleum Exporting Countries halved the price of oil worldwide, an event that had a cataclysmic effect on the oil industry... and on Cook's business.

Cook had then some money from the offering, but his former industries were not being profitable at the time, which is why he looked into some new business territory, in particular the movie rental territory. Due to a recommendation from an acquaintance, Cook invested in a local video store chain, with the thought of creating a franchise. The novel idea was to have a store, with an immense film catalogue and far more offers for their clients, thus having a clear advantage over the competition. Yet the big idea that would catapult Blockbuster came from Cook's previous endeavours.

Cook Data Services then adapted their software for a video rental chain. This meant that an important effort was devoted to create a highly detailed computer system that would manage and improve the store's catalogue. The system used a bar code scanner to read key data from each rental tape and from the member's card. The inventory control system generated some valuable statistics, such as the number of times each tape was rented, and a daily summary report by store. Even more, the system analysed data on the demographics of Blockbuster customers and segmented the data based on what tapes they rented. At the beginning, with only a few stores open, the system was used mostly for account management, but as Blockbuster started to grow, this rudimentary form of Data Mining (#BigData in the 80's, anyone?) would become crucial for Blockbuster's future and helped develop their own personalized recommendations.

Cook invested $3 million on a global distribution centre, which stored thousands of films from distributors, repackaged them and sent them to the appropriate location. It was in this distribution centre that the statistical work started to generate work for the company. Every film was coded and categorized, by genre and demographic appeal. Wait, demographic appeal? That's right, using statistics gathered at the local stores, the distribution centre managed to ship a personalized cargo for each of the stores. For instance, a local store in Florida may get forty copies of the last Family friendly flick, considering that the locals love those kinds of films, according to the rental statistics. A Blockbuster in Alabama could get only 5 copies of that same movie but much more copies of a new horror hit, which was estimated to be popular among the teenager costumers of that store. In practice, Blockbuster started filling their stores with personalized suggestions, making the film catalogue attractive to the demographic of their customers, at the time that they ensured that the latest new releases were available everywhere. Even more, the distribution centre could ship all the inventory for a Blockbuster in less than 24 hours, thus assuring that any new store would have the most rented movies and some personalized titles based on the expected consumer's demographics. This was to become key to the amazing expansion that the store experienced in the 90's.

David Cook left Blockbuster in the late 80's, but his data mining legacy changed the movie rental industry. Many other outlets started using a similar system to ship their stores, with a catalogue based on local preferences. Even today, Netflix's online recommendation system is just a sophisticated version of Cook's initial ideas (For the trivia, Blockbuster passed on buying Netflix for fifty million dollars in 2000).

Sadly, when the main advantage Blockbuster had over their competition was lost, the grim future for the company started. Now, after almost thirty years since its inception, Blockbuster will soon be off the market, yet their most important contribution to the movie rental business, the use of intensive statistics and data mining techniques, will live way beyond a Blue and Yellow logotype.


Related Topics

Related Publications

Related Content

Site Footer


This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.