Predictive Analytics Times
Predictive Analytics Times

6 months ago
Wise Practitioner – Predictive Analytics Interview Series: Dr. Gary Anderberg at Gallagher Bassett


By: Eric Siegel, Program Co-Chair, Predictive Analytics World for Financial

In anticipation of his upcoming conference presentation at Predictive Analytics World for Financial Las Vegas, June 16-20, 2019, we asked Dr. Gary Anderberg, SVP of Claim Analytics Product Manager at
Gallagher Bassett, a few questions about incorporating predictive analytics into finance. Catch a glimpse of his co-presentation, I’m Tired of All Those Predictive Alerts, and see what’s in store at the PAW Financial conference in Las Vegas.

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

A: We get the most satisfaction when our models accomplish 2 objectives:  1 – identify what the human claim adjuster misses, 2 – identify the event as soon as possible. These core objectives cover our breadth of models, from determining the ultimate cost of a workers’ compensation claim, to determining which claims are likely to be litigated and mitigating that event. The behaviors we are predicting are complex and multi-party and involve actions (or inactions) on the part of the claimant, the medical providers and occasionally the employer. The ultimate goal is to predict the potential for actions which will delay the resolution of the claim and drive up costs unnecessarily.

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

A: Our predictive models assist our resolution managers in their ability to make the best decisions in an efficient, timely manner. One example is uncovering claims with a high potential to minimize time off work by having a nurse dedicated to the injured employee’s recovery process. Perhaps the most obvious output in terms of financial performance is helping the adjuster to identify claims with third party liability, thus improving our ability to recover often substantial portions of claims costs.

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

A: When we developed our subrogation model, we saw the potential to increase recoveries almost 3-fold thru more efficient identification and selection of claims likely to have been caused by third parties at fault. Note that GB is a third party administrator handling claims on the behalf of clients, so the ultimate ROI will be an incremental reduction in overall claim costs.

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

A:  Mostly confirmation of what we know from experience but had a difficult time quantifying accurately, for instance the extent of the relationship between the initial denial of a claim and the likelihood that the claimant will retain a lawyer to resolve their claim after it is accepted. Perhaps the most interesting realization is that predictive factors can vary greatly depending on external circumstances such as the state of jurisdiction. On the other hand, one universal confounding factor is simply this—the older the claimant, the more likely the claim is to be difficult to manage in one or more dimensions.

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

A: Our presentation will provide an outline of how to prioritize multiple model alerts going to the same audience. One possible surprise — in many regards the hardest part of developing and successfully implementing a sophisticated predictive system is the social engineering of output integration with efficient operations in a large, complex administrative organization. The best predictive lift is useless if the operations people don’t understand how to use it correctly.


Don’t miss Gary’s presentation, I’m Tired of All Those Predictive Alerts, at PAW Financial on Wednesday, June 19, 2019 from 2:15 to 2:35 PM. Click here to register for attendance.

By: Eric Siegel, Program Co-Chair, Predictive Analytics World for Financial

Leave a Reply

Pin It on Pinterest

Share This