Organizations need to find ways to bring seas of data under control and use methodologies capable of identifying and harvesting the information that they have a need to know.
Three years ago, an IDC study sponsored by EMC predicted a 45-fold growth in digital information in the decade of 2010-2020. Three years later, little seems to stand in the way of the prognostication fulfilling itself.
On the heels of this, researcher McKinsey & Company reports that big data IT talent will be in under-supply, with the U.S. facing a potential shortage of 140,000-190,000 analytics specialists by 2018-and a shortfall of 1.1 million managers and analysts with the background and know-how to plumb data with great questions that produce the kinds of competitive-edge answers enterprises want.
Enter in this discussion IBM's Watson, a system capable of answering questions in natural language that debuted as a Jeopardy! Winner in 2011, and that continues to grow its value-added analytics capabilities in industry verticals like healthcare, financial services and telecommunications.
Last week, IBM gave its latest Watson presentation to industry analysts. I couldn't help but be struck by one of the presentation's' catch phrases: "Dying of Thirst in an Ocean of Data."
IBM's own research reveals that 90 percent of the world's data was created in the last two years, that 80 percent of all data is now unstructured (and therefore, not able to be processed in traditional transaction processing systems), and that we now have one trillion connected devices generating 2.5 quintillion bytes of data per day. The survey also reveals that one out of every two business leaders say that they don't have access to the data that they need
Just what do these business leaders want to know?
They want to know what their traditional daily, weekly, monthly, quarterly and annual reports can't give them-like how their business is doing at this very moment in time, and whether they are winning or losing with consumers and clients. They also want the agility to respond proactively to situations as they need to.
Customer fulfillment and satisfaction is a prime example.
If you are a bookseller and can see from your Website or a call center dialogue with your customer that he is interested in fly fishing, you are in a position to present him several books that fit the bill and broaden your chances of gaining the sale. He likewise goes away fulfilled because he believes that you have listened to him. In fact, you have even aided him in his quest by finding the titles he was searching for.
From an IT standpoint, unsecured data flowing in from the Web (or even voice-based information coming into the call center) has to be instantaneously analyzed and interpreted for results if business are to get to this point.
How important is this?
In its discussion with industry analysts, IBM reported 270 billion calls cost call centers 600 billion dollars each year; that half of these calls require escalation or go unresolved; that 61 percent of the calls could have been resolved with better information; and that companies could gain a potential improvement of 4.6 percent of market share for their wares if they could improve customer satisfaction by a single percent.
This brings us back to drowning of thirst in a sea of data-because no organization can get to this point unless it finds ways to bring these seas of data under control and use methodologies capable of identifying and harvesting the information that they have a need to know. The challenge can't be solved with hardware and software alone. It also requires identifying the right data to analyze, and determinations about which data should be archived or discarded. This is where many enterprises today find themselves. They don't know which data to keep and which to discard-and they are looking for a saving methodology.