The depth and breadth of predictive modeling applications are growing as insurance companies become more comfortable with this core function in modern property/casualty business. However, few carriers have fully explored the potential of predictive models and data-driven analytics to favorably impact their top and bottom lines.
Executive SummaryWhile the use of predictive models rapidly has become standard operating procedure for pricing and risk selection in major lines of business, their application remains sporadic outside these functions and in specialty lines and large-account pricing. In addition, insurers are still far from leveraging the full gamut of sophisticated modeling techniques, even in pricing, Towers Watson reports.
While the use of predictive models to develop more sophisticated rating structures rapidly has become standard operating procedure in major lines of business, their application in specialty lines or large-account pricing remains sporadic. Furthermore, applications outside pricing—e.g., claims, distribution management, loss control or premium audit—while growing, are still not commonplace across the industry.
Predictive modeling’s impact on profitability has increased over time, suggesting that predictive modeling applications offer greater value and that pricing implementation works. Yet application lacks uniformity and differs by line of business and carrier size. For instance, larger carriers are actively deploying predictive modeling in areas beyond risk selection and pricing, while smaller carriers are lagging. Personal lines carriers are leading the way, but commercial carriers are closing the gap.