Predictive Analytics World Boston, Sept 27-Oct 1, 2015, we asked Philip O’Brien, MIS and Portfolio Manager at Paychex, a few questions about his work in predictive analytics.
Q: In your work with predictive analytics, what behavior do your models predict?
A: Our models are targeted to four main portfolios: Our Operations Portfolio focuses primarily on retention and sensitivity to price, while our Sales Portfolio concentrates on product upsell potential. The Risk Portfolio is focused on customer default and resolution. Lastly, our Strategy Portfolio is a variety of models designed to influence the strategic decision making at a corporate level, and contains the employee model we will be discussing in Boston.
Q: How does predictive analytics deliver value at your organization? What is one specific way in which it actively drives decisions?
A: The value of modeling can be summarized by a quote from Shao Ming Lo, “Strategy in fact is the art of focusing.” The separation our models create allow targeted decisions to be applied to cross sections of our clients. Through this application of predictive analytics you can make the greatest impact with a strategy at a reduced cost.
Q: Can you describe a successful result, such as the predictive lift of your model or the ROI of an analytics initiative?
A: The value derived from our models can be demonstrated in a trio of retention models in our Operations Portfolio. Those models focus on controllable losses, uncontrollable losses, and price sensitivity, which we have presented on at previous PAW conferences. What is most impactful about these models is the way they are used across the organization, from discounting and retention to exception processing – these models help shape the way we view a significant portion of interactions with our clients.
Q: What surprising discovery have you unearthed in your data?
A: I continue to be amazed at how predictive analytics positively impacts every aspect of business. Surprisingly, even with consistently inconsistent data, there is value in putting a model to it. Once you have identified the data that signals the behavior you are going after, simply start applying the science to uncover valuable insight that can change the way you do business.
Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.
A: In dealing with customers, no matter the business, we understand the inherent value the relationship with the customer has on their longevity with your business. However, in looking at employee data, we found evidence that the inverse is also true: the relationship employees have with your customers impacts their employment longevity as well.
Don’t miss Philip’s conference presentation, Predicting Employee Churn with Anonymity on Tuesday, September 29, 2015 at 11:15am to 12:00pm at Predictive Analytics World Boston. Click here to register to attend.
By: Eric Siegel, Founder, Predictive Analytics World