December 1, 2013
By Marshall Lager, founder and managing principal, Third Idea Consulting; contributor, CRM magazine

New Year, New Strategies

Big data and friends

Humanity has continued to generate a truly staggering amount of data, and this trend will only continue and intensify—it has been estimated that 90 percent of all the world's data was generated in just the past two years. We're creating 1,200 exabytes (that's 1.2 billion gigabytes) of data every year. Businesses must come to terms with data hordes measured in petabytes, exabytes, zettabytes, and yottabytes—most people haven't even heard of those measurements.

"Big data is typically defined as a volume of data that exceeds the technology to handle it—which isn't really too hard to grasp as a concept," says Paul Greenberg, president of consultancy The 56 Group. "It's seen now as not only the amount of data, but the velocity of the data and the structure or lack thereof, as part of what defines it."

Coping with this massive information overload is a headache, but it's also an opportunity. As computing power expands, the ability to quickly process and analyze this data improves. Crunching the numbers to find trends and whatnot, using more factors than ever before, is the realm of big data. It's been part of the discussion for the past few years, but 2014 is likely to be more intense, as more developers come up with ways to use big data to affect business and everyday life.

"Big data is probably in a similar boat to CRM, in the sense that every company has (or should have) [their] own definition of what it truly means to them," Leary says. "You'll see a similar path for big data that CRM has traveled—and is still traveling. We're still [in the] early days, and so I think companies are still trying to figure out what to capture, how to aggregate it, how to analyze it, and how to use the insights found—in a programmatic process that can be repeated."

"Big data isn't new; it's an idea that is simply coming around again," Pombriant says. "Dealing with it as a pure technological issue has given us a mania about storage without showing much value. It's a cultural issue of teaching people how to make better decisions based on knowledge derived from information, which is extracted from big data. We need to enable ourselves to think about the meaning of data, and that's not at all common yet, according to my research."

It's interesting to note that the acceleration of data creation has happened in step with the growth of social media. That's no accident—communicating and sharing are strong human drives, and technology has allowed us to make those instincts into more or less full-time jobs. We'll come back to discuss the social connection shortly, but few would doubt that it has fueled the data explosion.

In the end, it's not how much you have but how you use it. "There are more refined aspects of [big data's] interpretation, such as the context of the data, and a lot is driven by the conversations occurring on the social Web," Greenberg says. "But companies like Oracle and SAP, with their varying versions of in-memory computing, and Infor, with the infinite capacity of Epiphany to scale and analyze and interpret the data, are merely examples of technologies that can handle so-called big data."

Big data will continue to get bigger as more companies adopt social technology, and machines begin speaking directly to other machines—the so-called Internet of Things. "Companies need to get cracking on this now," Leary says. "When the Internet of Things kicks in, and machine-to-machine data from sensors exponentially blows up the amount of data available, you have to have some kind of plan and process to help organize things, or it will be completely overwhelming—leaving many companies paralyzed and unable to get started."

Using what we know

One of the flagship cases for using big data in a practical way is predictive analytics. High-powered computers sort the torrent of information to trace root causes, spot hidden trends, and essentially predict the future for large populations as well as individuals.

"One of the most important differentiators in this era of customer engagement is to understand the customers that you have, and to be able to anticipate their behavior," Greenberg says. "Predictive analytics is the way to do this, and social data gives you the rich information you need to do it."

Accessing the power of predictive analytics isn't just about RAM and multithreading processors. Users have to adopt a new way of gathering information. "If you ask closed-ended questions about what happened in the past, you will not get anything interesting to analyze, so you need to ask open-ended and aspirational questions to do well with predictive analytics," Pombriant says. "Without that kind of data, you might be able to calculate next best offers or stack rank preferences, but the answers to those types of questions evolve all the time, rendering the predictive model obsolete over a short time. Gotta keep asking."

Previous Page Next Page