First Tennessee Bank had a problem.
Every month, it would send out a mass mailing designed to get customers to purchase a bank service they were not already using, such as a checking account or CD that had limited success.
By implementing a predictive analytics solution, First Tennessee was able to determine the likelihood of each individual customer to purchase each product in its portfolio. This helped the bank pinpoint product clusters and focus on the programs that delivered the best “sweet spots” for cross-selling opportunities. In just two months, First Tennessee received a 642 percent payback. And, that was just by using analytics in one area of its bank. (Read more here.)
First Tennessee saw the value in leveraging analytics to help guide better decision making.
“Other organizations,” according to Ian Ayres, best-selling author of Super Crunchers and Carrots and Sticks, “are presumptively screwing up if they’re not using predictive analytics.”
That is one of the themes that Ayres will be discussing at the upcoming IBM Financial Services Summit being held in Toronto (Sept. 25) and New York City (Sept. 27). Ayres will be speaking to leaders from the banking and insurance industries who are interested in improving marketing, customer service or sales performance, or who are responsible for risk and financial management.
“Predictive analytics is really powerful and helps organizations find what works in the world,” said Ayres.
Ayres breaks down predictive analytics into two key elements: regressions and randomizations. Regression refers to a variety of techniques to crunch data to find underlying drivers that will impact results and outcomes, and randomizations as a way to run experiments when you want to find the impact of decisions when historical data is lacking.
For example, Infinity Insurance used these techniques to “score” its claims so it could provide a more efficient and accurate way to pinpoint fraud. As they gathered new information from customers, they could immediately determine whether claims were legitimate and qualified for immediate approval or were potentially fraudulent and should be further investigated – ultimately improving productivity and accuracy of the entire claims-handling process. (Read more here.)
Since writing Super Crunchers, Ayres admits that while the analytics revolution is moving forward, it is still incomplete.
“Most organizations are still not doing analytics with sufficient power,” said Ayres. “They need to learn how to streamline their consumption of predictive analytics – not only the initial decision to use the technology, but the ability to pull the trigger and change behaviors based on the results of the analysis.”
Ayres gives an example of organizations that say they have jumped on the big data bandwagon.
“You can always find businesses and leaders who say they have a big data set and are always analyzing it,” said Ayres. “But, it's harder to find organizations that say they have both analyzed it and changed any behavior because of it. The real value is to follow it through all the way to ROI.”
Obviously, changing behaviors inside organizations starts at the top and successful analytics-driven organizations need leaders who embrace the technology instead of being interested in “analytics window dressing.”
But, that’s only part of it, argues Ayres. Organizations also need the right staff with the necessary analytical skills to succeed.
“The U.S. would do well to make introductory statistics a central part of high school mathematics,” said Ayres. “After all, I'd guess that less than 3 percent of students actually use calculus in their daily lives compared to maybe 80 percent of students who use statistics. I don’t mean that those 80 percent are going to be hard-core number crunchers, but they will be consuming and conversant with statistics as part of their current job responsibilities.”
To Ayres' point, an increasing number of universities are beginning to have quantitative statistical courses as prerequisites for graduation, while other universities are launching Masters in Analytics programs, such as Michigan State University, Northwestern University and others.
In fact, Ayres was quick to point out the collaboration his current employer, Yale University, has with IBM to train the next generation of number crunchers.
“The word is out at Yale that those who possess a strong acumen in analytics have a leg up in the competitive job market,” said Ayres. “There is such clear evidence that money can be made for those with these skills.”
And speaking of making money, Ayres says he has a treat for those attending the IBM Financial Services Summit. He plans to share a new statistically-based approach to help customers better determine which 401k mutual funds will perform best. You might be surprised by his findings.
For more information on the event and to register, click here for Toronto and here for New York City.