Insurance-Canada.ca - Where Insurance & Technology Meet

Predictive Modeling Gaining Momentum with Insurers

There is increasing evidence that P&C insurers in the US and Canada are seeking, and finding benefits with the use of predictive analytics and modeling in underwriting, and claims.

National Underwriter On-line reported that a recent Towers Watson survey found that 90% of US insurers surveyed indicated that the use of predictive analytics enhanced rate accuracy in 2010, up  from 70% of companies surveyed in 2009.  In addition, in  2010, 76% of the companies surveyed realized an improvement in loss ratio in 2010.

Brian Stoll, Towers Watson senior consultant and one of the survey’s authors, is quoted in the article as concluding: “Effective implementation of predictive modeling enhances risk selection and pricing, leading to greater insurer profitability and the potential for growth in market share.”

The theme of momentum was reinforced at the 2011 Insurance-Canada Technology Conference.  As noted in our last post,  Kimberly Harris-Ferrante, 2011 Insurance-Canada.ca Technology ConferenceVP and Distinguished Analyst at Gartner reported on a recent survey of 53 Canadian Insurers and 42 Canadian Insurance Brokers.  One key finding was the congruence of opinion between two groups on the priority of Business Intelligence and Data Management – second highest IT priority among insurers, and highest IT priority among brokers.

Turning to claims, at the same conference, Christina Colby, Vice President, Global Business Intelligence Leader for Insurance, Capgemini noted that there is a trend for insurers to pursue efforts to use predictive analytics to attack fraud.  Ms. Colby suggested that this looks good on the surface, but can become very complex in the implementation.  She indicated that some insurers are starting with simpler, low hanging claims fruit – like recoveries and litigation management – which can demonstrate success and provide funding for the more complex, but potentially more rewarding initiatives.

The interesting point throughout all the information  is that discussions about predictive analytics and modeling is no longer focusing on whether to proceed, but how.