Predictive Analytics: Businesses’ Crystal Ball

Stephen DeAngelis

August 3, 2016

“What if you could predict what products or services your customers wanted before they did,” asks Dr. James Canton (@futureguru), CEO of the Institute for Global Futures. “What if you knew how your competitors were thinking about the market? What if you could anticipate with accuracy the next game-changing business strategy that would give you a strategic advantage?”[1] Good questions. Are answers to those questions possible? Maybe. The synergistic rise of big data, artificial intelligence, and predictive analytics has given the business world the equivalent of a crystal ball. To be completely honest, the predictive analytics crystal ball is not crystal clear. It’s closer to the biblical phrase, “Now we see through a glass, darkly.”[2] Per L. Bylund (@PerBylund), an Assistant Professor of Entrepreneurship at Oklahoma State University, reminds us that human behavior is not completely predictable. “Prediction,” he writes, “requires that we are able to accurately exclude all but one or a couple [of] highly probable outcomes. And we have to be able to rely on that these predictions turn out to be true. Otherwise we’re just playing games, and so we’re making guesses. Sure, they’re educated guesses (because we’ve excluded the impossible and almost-impossible), but they’re still games and guesses. … Guessing with access to huge amounts of data is easier, at least if the data is reliable and relevant. But a good guess is not the same thing as a prediction; it is still a guess, and it can be wrong. Winning every time requires luck.”[3] Nevertheless, analysts constantly lament the current state of business forecasting and they are looking for anything with the potential for improving the situation.

 

Murali Nadarajah, Head of Data and Analytics at Xchanging, explains that new advances in analytics have moved businesses closer to being able to predict future outcomes. “The science of analytics,” he writes, “has continued to evolve. We are able to analyze huge amounts of data today — and with the advent of new tools and methodologies — are no longer restricted to analyzing only what has happened in the past. We now have the power to be predictive.”[4] He adds, “Predictive analytics is essentially the interrogation of massive amounts of historical data, from various sources and in various forms, to identify causal relationships between data points and thus predict future outcomes.” Are predictive analytics perfect? No. Are they better than anything businesses previously have had? Yes. Like many people, you may be confused about the differences between forecasting and predictive analytics. Steve Banker (@steve_scm), Service Director, SCM at ARC Advisory Group, places himself in the confused category. “I’m not sure we have a good understanding in the field of how predictive analytics differs from forecasting,” he writes. “Here is how I would differentiate the two. All forecasting involves making a prediction, but not all forecasts are based on Big Data. Using historical data to make a regional forecast for most companies would not require Big Data. Forecasting demand at the store level probably would. The more inputs, the more granular the data, the more granular the forecast, the more we are in the realm of Big Data analytics. In short, when I hear predictive analytics I am apt to think this is a forecast based upon Big Data.”[5] In other words, Banker accepts Bylund’s notion that forecasting is about a good guesses and predictive analytics is about a great guesses.

 

As Banker notes, predictive analytics provide a more granular look at potential behaviors. Nadarajah explains how this can be very useful to a business in areas like churn analysis. He writes, “Churn analysis is another way companies are using big data to predict customer behavior. By tracking the behavior of past customers, analyzing that data and applying that analysis to the current customer base, a company can determine which of its current customers are going to leave or drop-off the system. That insight can then be provided to the appropriate department to take action and retain the customer.” What really distinguishes today’s analytics from traditional analytics is the emergence of artificial intelligence (especially cognitive computing). I define cognitive computing as the combination of semantic intelligence (machine learning and natural language processing) and computational intelligence (advance mathematics). Cognitive computing can ingest and analyze much more data and deal with many more variables than has been previously possible. This results in the kind of granular analysis described by Banker. Rich Wagner notes, “Today, companies are hungrier than ever to utilize data to gain greater insight into their business performance, with a strong emphasis on the variety of data used. From traffic and weather patterns to changes in consumer sentiment and volatility in Asian markets, companies are dealing with more complex problems and are turning to external Big Data to for the answers. No longer is data isolated in the IT department, and executives are looking to Big Data to provide big answers. They want to discern which external factors will impact the sales and demand of a particular product in the next six months — and by how much. They want to accurately determine which markets their company should exit and which markets are poised for growth. The demand for such answers — in real time, no less — is bringing about three distinct trends in today’s predictive analytics process.”[6] He continues:

“The diversity of data sets and sheer amount of external data continues to grow with the speed of technology. Global companies certainly recognize its power, but only now are they beginning to find ways to glean real business value from the insights this data can provide. From implementing more seamless processes for gathering and correlating external data to finding flexible solutions that enable hypothesis testing and analysis of various data sets, companies looking to fully leverage Big Data to solve big problems must embrace this new era of predictive analytics.”

Charlie Dai (@CharlieKunDai), an analyst with Forrester Research, asserts, “Predictive analytics has become the key to helping businesses … create differentiated, individualized customer experiences and make better decisions.”[7] He adds, “Organizations require predictable insights into customer behaviors and business operations. You must implement predictive analytics solutions and deliver value to customers throughout their life cycle to differentiate your customer experience and sustain business growth. You should also realize the importance of business stakeholders and define effective mechanisms for translating their business knowledge into predictive algorithm inputs to optimize predictive models faster and generate deeper customer insights.” For me, the bottom line is this: Companies have access to a lot of data and they are wasting that data if they don’t take advantage of predictive analytic technologies that are rapidly maturing. Predictive analytics may still be a cloudy crystal ball, but the picture is clearing every year.

 

Footnotes
[1] James Canton, “The Predictive Enterprise: Five Trends Shaping the Future of Business,” Huffington Post The Blog, 26 November 2015.
[2] 1 Corinthians 13:12.
[3] Per L. Bylund, “No, ‘Big Data’ Can’t Predict the Future,” Mises Daily Articles, 7 December 2015.
[4] Murali Nadarajah, “The Power of Being Predictive — Big Data Gets Smarter than Ever,” insideBIGDATA, 28 August 2015.
[5] Steve Banker, “Predictive Analytics Comes to the Logistics Industry,” Logistics Viewpoints, 29 February 2016.
[6] Rich Wagner, “A New Era of Predictive Analytics: 2016 Trends to Watch,” Information Management, 12 February 2016.
[7] Charlie Dai, “Architect Your Predictive Analytics Capability To Drive Digital Business,” Information Management, 22 December 2015.