Tech Trends: Joe Burgess of CHSI on the Power of Predictive Analytics

“Predictive analytics” as a tech tool has been a part of the insurance business conversation for some time now. The idea behind the technology is to aid underwriters and others to make the best pricing decisions possible based on targeted analysis of claims data.

Predictive analytics was addressed briefly in the first part of this series on technology and its relationship the millennial workforce. Below, in the second part of Louisiana Comp Blog’s “Tech Trends” series, Joe Burgess, business development expert with Nevada-based CHSI Technologies, offers his perspective on how predictive analytics can benefit small workers’ comp companies.

Comp Blog: Please explain your background and your work with CHSI.

Burgess: CHSI Technologies formed as a company in 2010. I am a shareholder and provide business development support for the company. The origin of our “Connections” software was with the services company, CHSI of Nevada, a company founded in 1996 that developed and managed workers’ comp self-insured groups. I was invited to join CHSI by founder Jim Leftwich at the beginning of 1997, and my focus was on business development for self-insured group and risk management support for members. I had worked with Jim as a service provider on improving post-loss response for individually self-insured companies. I became an officer of CHSI in subsequent years.

In 2000 CHSI hired an IT Director and, in 2003, a developer, to begin building a software for group administration, based on our needs. We were developing a new California program that was going to place heavy demands on our administrative capabilities, and handling it with spreadsheets and Microsoft Access was going to be tough. Ultimately, we invested substantial portions of our profit from administration into the development of “Connections.”

I have also been on the Workers’ Compensation Committee for the Self-Insurance Institute of America (SIIA) since about 2007.

Comp Blog: What is “predictive analytics” and around when did this technology become a talking point for the industry?

Burgess: Predictive analytics has been used in workers’ comp programs in some fashion for a long time. Old school underwriters might have been using their intuition but of course they were attempting to base their prediction on their own trending analysis of historical data, and actuarial analysis. What evolved was the ability, with technology, to sharpen analysis of large volumes of claims data and to bring greater prediction and insights to the process that was more than “seat of the pants” decision-making.

Greater predictability has always been a fundamental of the law of large numbers in insurance – emerging technology has been able to refine the detail of that prediction. Technology is moving quickly – the “buzz” on the field has been building for at least five years.

Comp Blog: How does CHSI incorporate predictive analytics into “Connections?”

Burgess: “Connections” supports predictive analytics through the design with clients of specific reports that will present data in forms that will be analyzed most effectively by the predictive analytics software in use. As the name implies, “Connections” integrates with other software. The planning is also that “Connections” will have an integrated analytics package in 2017.

Comp Blog: How does predictive analytics work for the day-to-day operations of a company? How about a self-insured group in particular?

Burgess: The day-to-day application of predictive analytics will vary widely by program.  Improving accuracy of underwriting is a big opportunity. A further exciting opportunity is to improve post-claims response – identifying claims very early that are a high risk for high development, to improve response in the critical first week.

Often, systems are very slow to call out the claims that blow up – they are identified when the writing’s already on the wall, when effective early intervention is already in the rearview. In these cases, predictive analytics can also drive better responses – when effective involvement is occurring early, the outcomes can be better for the employee.

For self-insured groups, they are beginning to use predictive analytics to increase engagement of members – helping them understand risk and what drives loss to develop prevention strategies. Funds can create financial incentives for that increased engagement, and of course members will be incentivized by better performance.

Comp Blog: Where do you see predictive analytics, and more generally, workers’ comp software solutions, going in the future?

Burgess: Predictive analytics will continue to be refined, enhanced, and integrated into workers’ comp operations. As that integration continues, it will be less of a buzzword and more just a part of operations. To this point, one recent topic of discussion in comp Linkedin groups has been the decreasing importance of experience modification factors in insurance pricing. As insurers use predictive analytics and go to credits and debits to get the pricing right they can be more accurate and competitive in pricing business. Programs will have to incorporate analytics to be fully competitive.

Leave a Comment

Your email address will not be published.