Increasingly, businesses in almost every industry are investing in predictive analytics to better anticipate and mitigate risks. To determine how insurance companies are using predictive analytics, Earnix and ISO conducted a joint industry survey in September titled “2013 Insurance Predictive Modeling Survey.” With feedback from 269 insurance professionals in Canada and the United States, the survey provides insights on how insurers are successfully using predictive analytics, resources that enable their efforts, and where and how carriers are applying those initiatives to overall business performance.
Click through the slide show below to see how predictive modeling is used throughout the insurance industry and what the future holds for advanced analytics.
Predictive analytics in insurance is becoming widespread as a way to gain a competitive advantage, with 82 percent of insurance professionals using it in one or more lines of business. It is most widely used in personal automotive today, with homeowners a close second.1
The most common use of predictive modeling is in pricing, where 81 percent of insurance professionals use it always or frequently. In highly competitive and automated lines of business such as personal auto, accurate pricing can give insurers an edge.2
Not surprisingly, companies with more than $1 billion gross written premium (GWP) rely almost exclusively on internal resources to develop their predictive models, while smaller companies rely more on external resources.3
All of the large companies (more than $1 billion GWP) surveyed have resources dedicated to predictive modeling. Larger companies can distribute resource costs across lines of business and also reap the rewards of applying analytics to a greater number of policies, making investment in internal predictive analytics resources a high ROI proposition.4
Nearly all insurers using predictive analytics — 91 percent — must go outside the company to acquire additional data elements for their predictive analytics programs.5
Of the companies that use external data, the vast majority use insurance score or raw credit attributes, but there is increasing emphasis on catastrophe and weather data as events such as Superstorm Sandy continue to plague policyholders throughout the country.6
The primary challenge for large companies is lack of usable data, while smaller companies cite insufficient data overall.7
The two biggest obstacles in adopting predictive analytics are a lack of know-how/technical expertise and cost. Data inefficiencies, scarcity of analytic talent, and the cost of that talent can also hold companies back from expanding the use of predictive analytics.8
Profitability is the predominant benefit insurance professionals get from using predictive analytics. Benefits such as speed, efficiency, effectiveness, and growth often contradict each other and create a management challenge for insurers trying to implement an effective predictive modeling program.9
Successful predictive modeling takes skill and agility due to its fast-paced evolution and often mind-bending scientific approach. According to 49 percent of insurance professionals, personal auto and homeowners will continue to be the primary focus of new predictive modeling initiatives in the next two to three years, but there is increasing interest in applying predictive analytics to commercial lines as well.10
Pricing and underwriting will be the two functional areas at the focus of predictive modeling initiatives in the next two to three years.11