Following from our previous article, “5 Reasons why Big Data Projects Fail,” here is a look at five major enterprises which tried to deploy big data systems without fully understanding and analyzing their business case and failed. These case studies are perfect examples of what not to do when working with Big Data and how to avert such corporate disasters.

big data blog, big data, big data failures, case studies, big data projects



The best known case of big data failure, Google Flu Trends was created in 2008 to predict flu outbreaks in 25 countries by analyzing region-wise Google search queries about flu. In 2009 GFT failed to adequately warn about the H1N1 outbreak and 2013 saw the service crash and fail with an error margin of 140 percent mainly because their data collection did not take into account factors like searching for “cold” and “fever” did not necessarily indicate flu.


Sears could have easily written a successful Big Data story had it done things a little differently. An ineffective and negative MDM strategy across suppliers, products, and customers affected their procurement systems. In the confusion between their POS (point of sale) and CRM (customer relationship management) systems, customers were left feeling that their data was sorely mismanaged.


For companies that have access to big data, it is important to check the data before running analytics tools on them in order to avoid a fiasco. Amazon faced the storm when it published T-shirts on its site with offensive messages such as “Keep Calm and Rape A Lot” and “Keep Calm and Punch Her.” Though the seller company, Solid Gold Bomb, blamed poor programming and analytics for automatically generating the phrases via a scripted computer program, it was an unprecedented blunder to put up the t-shirts without first checking them.


Target almost paid a heavy price for injudicious use of predictive results. The retail-giant analyzed vast amounts of customer information, shopping trends and personal information and created a targeted sales approach for expectant mothers offering them personalized pre- and post-natal items. But when an irate father complained about such pregnancy-related promotional emails being sent to his teenage daughter, it came out that the girl was indeed pregnant and hiding it from her parents. The incident put a black mark on Target’s record for violation of privacy and insensitive marketing tactics; they lost face as they compromised the client’s trust.


In 2000 Blockbuster was at the top of its game when Netflix proposed a strategic brand partnership where they wanted to run Blockbuster’s online brand in exchange for the latter promoting Netflix in their retail chains. Netflix proposed the use of data analysis to gather insights about customers and the market; Blockbuster refused. A decade later, Blockbuster, which refused to change with the times, filed for bankruptcy while Netflix gained a competitive advantage by using data analysis to identify customer preferences and tailor their offers.

While there are a number of such failures, there are companies that have found the right way to tap into their most important asset – data. These agile enterprises have taken a progressive approach to Big Data projects by doing away with legacy systems and ideas and deploying global data governance policies under a strong leadership.

We will talk about how to make big data projects succeed in our next article “7 Steps for Big Data Success,” so log into next week.

Reach out to us for more information on how you can make your Big Data pr