Predictive Analytics Times
Predictive Analytics Times
EXCLUSIVE HIGHLIGHTS
SHARE THIS:

2 weeks ago
Wise Practitioner – Predictive Analytics Interview Series: Tom Warden at Employers

 

By: Eric Siegel, Program Chair, Predictive Analytics World for Business

In anticipation of his upcoming conference presentation at Predictive Analytics World for Business Las Vegas, June 16-20, 2019, we asked Tom Warden, SVP, Chief Data and Analytics Officer at EMPLOYERS, a few questions about incorporating predictive analytics into business. Catch a glimpse of his presentation, Engage Everyone Who Will Touch the Analytic Model in the Development Process, and see what’s in store at the PAW Business conference in Las Vegas.

Q: In your work with predictive analytics, what behavior or outcome do your models predict?

A: Currently we are focused on a wide range of predictive analytics, from highly-segmented pricing models, claims triage and management, to the intelligence that enables automation of many of the decisions made in the life cycle of a policy. So we are predicting what the appropriate price for a policy is, when a claim is made how severe it will become and whether a specialist is needed to make a specific decision rather than having an algorithm do it.

Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?

A:  One way we are creating value is by making our operations more efficient. There are many ways we do this but the overarching principle is to focus human intelligence on decisions that are complex and benefit the most from domain knowledge while standardizing and routinizing simple decisions with automation. Over time, we expect machines to enhance our human intelligence as well.

Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?

A: I can but my CEO would shoot me. That aside, measuring realized lift is difficult unless one creates a true test and control experiment. We do our best to quantify the impact our models have by looking at how results change pre and post implementation, but in a business like ours there are too many variables to control for to truly measure realized lift with precision. The old axiom that for every 10 basis points of lift a business program creates, the sum of the contributions to it by each business function involved will total 50 basis points, is as true as it ever was.

Q: What surprising discovery or insight have you unearthed in your data?

A: One challenge we have in insurance is when we don’t write a policy we’ve quoted we don’t know what price the “winning” company wrote it for and whether we’ve actually won or lost in the bargain. We get a very incomplete view of price elasticity and whether giving up premium to write more policies is appropriate. All that said, I think there are insights to exploit from analyzing whether at renewal we keep or lose policies we’ve already written based on the price we’ve offered.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

A: That as predictive modelers and data scientists we need to adopt a holistic view of what we are developing in order to maximize the realized business benefit.

—————————–

Don’t miss Tom’s presentation, Engage Everyone Who Will Touch the Analytic Model in the Development Process, at PAW Business on Tuesday, June 18, 2019 from 11:45 AM to 12:05 PM. Click here to register for attendance.

By: Eric Siegel, Conference Chair, Predictive Analytics World for Business

Leave a Reply