Commercial property/casualty insurance companies are making steady progress in applying predictive analytics to their operational processes, according to a report released Tuesday by New York-based Towers Watson & Co.

The “2012 Predictive Modeling Benchmarking Survey” is based on an October 2012 survey of 72 property/casualty North American insurance executives and finds advances in implementation by insurers.

“As carriers become more data-driven, the capture and transformation of data into useful information has become a critical differentiator of performance in today's P&C insurance marketplace,” the report says. “Leading carriers are leveraging predictive modeling to establish a competitive advantage starting with efforts to outperform competitors in risk selection and pricing.”

Nonetheless, in areas such as claims, the report found that insurers' use of predictive modeling varies by line of business, noting that personal lines insurers are much more likely to use predictive modeling to detect fraud, while commercial lines insurers are more likely to use it to triage claims for initial adjuster assignment or to evaluate claims for litigation potential.

The report also found varying opinions regarding analytics according to the size of the insurance company. “Larger insurers, given their scale and greater resources, are using predictive modeling for risk selection and pricing to a greater extent for all lines, while smaller carriers have been slower to adopt predictive modeling applications,” the report states.

%%BREAK%%

But this dynamic may be beginning to change as the report notes that 39% of small to mid-market commercial insurers surveyed in 2012 said use of predictive models was essential, while only 22% said so in the same survey a year prior.

Klayton Southwood, a senior consultant at Towers Watson, cautioned that commercial insurance companies newly implementing predictive analytics will take time to get a return on investment.

“The lead time to realize both top- and bottom-line impacts tends to be much longer in commercial lines,” he said in a statement. “It will be interesting to see whether these times will shorten as carriers continue to feel more comfortable with predictive modeling efforts and become better versed in implementation.”

The report is available here.