Following a growing trend in analytics, North American P&C insurers have increased predictive modeling use across the majority of business lines, according to research from Towers Watson. The company polled P&C insurers to examine their use of predictive modeling in order to improve claim management, underwriting, pricing and other core business functions.
Eighty-three percent of respondents in personal lines rated predictive modeling as essential or very important. Seventy-nine percent of small- to mid-market commercial lines, and 56% of large accounts and specialty lines, claimed the same.
“We see carriers continue to try and increase the sophistication of their overall ratings structures, looking for new variables that might be interesting,” says Klayton Southwood, senior consultant in the P&C insurance practice at Towers Watson, of the marked increase in predictive modeling use.
Many insurers that employ modeling began using it to cut costs, Southwood explains, which is why few have a company-wide approach for predictive modeling across all core functions. “The motivation around there is expense saving,” he says. “They’re trying to look at predictive analytics and automate as much of that as possible to drive savings in underwriting.”
However, Southwood notes many other ways to employ predictive modeling for various business needs. In addition to cutting costs, companies may use it to take a holistic approach to claims management, make underwriting decisions, focus their marketing efforts, and determine placement and compensation for agents.
Survey results indicate that overall modeling usage varies according to company size and line of business. Large insurers, for example, generally have modeling ingrained in their operations and are beginning to apply it to other company functions. They have greater experience with modeling in relation to pricing and more frequently use it to boost the financial success of their claims departments. Larger carriers also experience more positive top- and bottom-line benefits from predictive modeling.
Small carriers, in contrast, aim to distinguish themselves through claims and service rather than with predictive modeling. “Smaller carriers struggled early on with whether they were going to model,” Southwood says. They were so focused on trying to apply modeling to pricing, he says, that it has taken them longer to expand their focus to other divisions. Some smaller carriers are also concerned that modeling will result in negative effects on their top line related to retaining business and market share.
In regards to lines of business, personal lines is ahead of the game. It got the head start on predictive modeling when Progressive began to use it for personal auto coverage, Southwood explains. Because many other personal lines companies followed suit, that sector now has more advanced strategies for using modeling and applying it across the enterprise.
Personal lines has given the commercial sector has an example to follow. “[Modeling] is clearly coming now in commercial lines,” Southwood notes. The trend is quickly progressing among these insurers, which have a better understanding of data cleansing and other modeling practices after monitoring the success of personal lines.
The auto insurance sector, with the rising adoption of usage-based insurance, is an especially popular area for predictive modeling. In 2012, there weren’t any insurers with UBI programs and only 4% had plans to create them. According to survey data, nearly half of person lines auto carriers have formal UBI plans, up from just a third last year, and some carriers have made progress in executing them.
“It has undergone a big jump, and I think it’s really going to accelerate,” says Southwood of UBI adoption. “There is a big learning curve there that carriers have had to go through. A lot of carriers have been able to overcome hurdles and work with service providers to get the data.”
[To hear about how insurance companies and financial firms are managing their complex data architectures, attend the Future of the Financial Services Data Center panel at Interop 2014 in Las Vegas, March 31-April 4.
You can also REGISTER FOR INTEROP HERE.]
Kelly Sheridan is Associate Editor at Dark Reading. She started her career in business tech journalism at Insurance & Technology and most recently reported for InformationWeek, where she covered Microsoft and business IT. Sheridan earned her BA at Villanova University. View Full Bio