By: Jesse Cryderman
“The price of light is less than the cost of darkness.”
--Arthur C. Nielsen, Sr., Founder, ACNielsen
Arthur C. Nielsen, the market research pioneer, is probably best known for illuminating trends related to radio and television broadcasts. In pursuit of this particular “light,” in 1948 his son,
Arthur C Nielsen, Jr., convinced ACNielsen to invest $150,000 in the UNIVAC, the first commercially available business computer. Under the leadership of Arthur Jr., the company subsequently grew
from $4 million in annual revenues to nearly $700 million.
Today, companies around the world are investing like never before in tools that shed new light on their customers, operations, systems, and potential market opportunities. Most of these tools can
be grouped under one heading: big data.
With so much buzz surrounding this concept, and so many solutions being re-labeled as “big data” solutions, it might be helpful to advance an easy to apply, scalable definition. Big data is
simply a collection of data sets that are so large and complex that they strain the limits of legacy data management systems. In other words, when the size of the data becomes part of the
problem, it’s a big data challenge. In the days of the UNIVAC, a single megabyte of data would have constituted a big data challenge. In 2013, Twitter and Facebook typically generate a combined
30 terabytes of structured and unstructured data every 24 hours. (Try managing that with a traditional relational database!)
According to Gartner’s latest deep dive, big data will drive $34 billion in worldwide IT spending in 2013, with more than 40 percent of global firms actively investing in big data. There is
no such thing as a one-size-fits-all big data solution, however, and each industry and each company has unique challenges that must be addressed. A rental car company, for example, requires a
substantially different big data solution than a mobile network operator, and yet some solution providers essentially re-brand the same big data solution for multiple verticals.
Chances are your company is already investigating big data, so now is a good time to ask: does your big data strategy have any blind spots?
What you can’t see can hurt you
Instead of attempting to predict the outcome of a presidential election based on early polling data (a task the UNIVAC successfully completed on live television in 1952), mobile network
operators (MNOs) are analyzing myriad data streams to address fundamental business challenges: reducing OPEX, increasing revenues, and maintaining and enhancing the end user experience.
In the current environment, there are several complex questions that MNOs need to address. When network congestion in metropolitan Chicago impacts application performance, are iPhone 5 users
impacted in the same way as Samsung Galaxy S4 users, and if so, do the network irregularities affect these customers’ likelihood to churn? Which of these customers are likely to be influencers in
their social networks, and should receive priority attention? Of these influencers, which contact channels and promotions are most effective?