A survey of over 1100 executives by the IBM Center for Applied Insights showed that organizations making extensive use of analytics experienced up to 1.6x the revenue growth, 2.0x EBITDA growth, and a 2.5x stock price appreciation compared to their peers. And what they are analyzing is Big Data, a combination of structured data found in conventional relational databases and unstructured data pouring in from widely varied sources.
Big Data is growing fast. By 2015 the digital universe, as forecast by IDC, will hit 8 zettabytes (ZB). (1ZB = 1021 bytes, one sextillion bytes). Adding to the sheer volume is the remarkable velocity at which data is created. Every minute 600 new blog posts are published and 34,000 Twitter tweets are sent. If some of that data is about your organization, brand, products, customers, competitors, or employees wouldn’t you want to know?
Big data involves both structured and unstructured data. Traditional systems contain predominantly structured data. Unstructured data comes from general files; from smart phones and mobile devices; from social media like Twitter, Facebook, and others; from RFID tags and other sensors and meters; and even from video cameras. All can be valuable to organizations in particular contexts.
Large organizations, of course, can benefit from Big Data, but midsize and small businesses can too. A small chain of pizza shops needs to know the consumer buzz about their pizza as much as Domino’s.
IBM describes a 4-step process for tapping the value of Big Data: align, anticipate, act, and learn. The goal is to make the right decision at the point of maximum impact. That might be when the customer is on the phone with a sales agent or when the CFO is about to negotiate the details of an acquisition.
Align addresses the need to identify your data sources and plan how you are going to collect and organize the data. It will involve your structured databases as well as the wide range of enterprise content from unstructured sources. Anticipate addresses data analytics and business intelligence with the goal of predicting and shaping outcome. It focuses on identifying and analyzing trends, making hypotheses, and testing predictions. Act is the part where you put the data into action, whether it is making the best decision or taking advantage of a new pattern you have uncovered. But it doesn’t stop there. Another payoff from Big Data comes from the ability to learn, for the purpose of refining your analytics and identifying new patterns based on subsequent data.
Big Data needs to be accompanied by appropriate tools and technology. Earlier this month, IBM introduced three task-specific Smarter Analytics Signature Solutions. The first addresses anti-fraud, waste, and abuse by using sophisticated analytics to recommend the most effective remedy for each case. For example it might recommend a different letter requesting payment in one case but suggest a full criminal investigation in another.
The second Signature Solution focuses on next-best-action. This looks at the various data uses real-time analytics to predict customer behavior and preferences and recommend the next best action to take with regard to a customer, such as to reduce churn or up-sell.
The third Signature Solution, dubbed CFO Performance Insight, works on a collection of complex and cross-referenced internal and external data sets using predictive analytics to deliver increased visibility and control of financial performance along with predictive insights and root-cause analyses. These are delivered via an executive-style dashboard.
IBM isn’t the only vendorr to jump on the Big Data bandwagon. EMC has put a stake into this market. Oracle, which has been stalking IBM for years, also latched onto Big Data through Exalytics, its in-memory analytics product similar to IBM’s Netezza. Of course, small players like Cloudera, which early on staked out Hadoop, the key open source component of Big Data, also offer related products and services.
Big Data analytics will continue as an important issue for some years to come. This blog will return to it time and again.