Providing the underwriter with predictive models with thousands of synthesized variables may prove more profitable.

In today’s soft market and challenging economic times, carriers continually search for innovative ways to meet profit goals. One area getting greater attention is the underwriting operation, since underwriting profitability correlates so closely with a carrier’s ability to accurately price risk.

While predictive analytics technologies—to enhance claims reserving, fraud detection and personal lines rating—have been around the industry for some time, we are just starting to see some traction in the application of predictive analytics for underwriting.

Carriers, especially their underwriters, have long argued that underwriting is an art that relies on years of experience and carefully honed judgment. Though I respect the value of the underwriter and their respective skill, predictive analytics offers a new opportunity to complement the art with science, and support the underwriter’s skill to make underwriting a much more profitable endeavor.

Even the most effective underwriters make decisions based on a small, finite set of variables, combined with their judgment. But, consider the possibilities if you provide the underwriter with predictive models that have synthesized hundreds and even thousands of variables, i.e., a highly multivariate environment.

Data is the fuel of predictive analytics—the more data the better—and insurers have an abundance of data. Through the process of predictive analytics, the data from as many sources as possible is analyzed to provide analytic models designed to predict a future outcome. Armed with this insight into the future, underwriters can make the most informed risk and pricing decisions.

Predictive analytics underwriting systems build carrier-specific risk models based on the company’s historical policy, claims, underwriting, billing and agency data, but also can incorporate more data from external sources. Underwriters can use predictive models to aid in a variety of underwriting decisions, including appetite and risk selection, tier placement, precision rating and schedule rating.

Most carriers rarely debit, and generally apply robust and sometimes indiscriminate credits. Faced with a competitive market, many carriers have dropped base rates across the board.

However, carriers using predictive analytics for underwriting are able to pinpoint the risks that truly deserve aggressive credits. Knowing which risks to target in a soft market can provide a dramatic competitive advantage. If a carrier knows that a certain class of business is apt to be more profitable, it can aggressively target that class of business with better pricing, which translates to competitive advantage and increased market share.


All business is not necessarily “good” business. With predictive analytics, a carrier has the opportunity to avoid adverse selection by identifying classes of inadequately priced business, prompting the carrier to reduce (or eliminate) the credits given or debit accounts in the riskier class. Facing a higher, albeit more appropriate, premium, these higher-risk insureds often will look for better pricing and switch to those carriers with lower inadequate pricing.

When applied to schedule rating, predictive analytics are an ideal complement to an experienced underwriter. A carrier’s manual rate is designed to set an adequate price for a specific class of business; then, experienced underwriters use their best judgment to apply scheduled credits and debits, further refining the price. Using predictive analytics during the schedule rating process, underwriters are better able to apply scheduled credits and debits objectively at the individual policy level to better match premium and risk.

With predictive analytics, insurers can reliably fast-flow already adequately priced policies through the process. As a result, underwriting expense is lowered while ease of doing business between carriers and agents is improved. It’s a win for all policy stakeholders—from the carrier to the agent to the insured.

The benefits of predictive analytics are not limited to those lines of business that are experiencing soft market conditions. A predictive model can help underwriters better understand where to focus their attention, and help them to understand where they should not focus their attention.

While every insurer may not yet be ready to take on the complexity of predictive analytics technologies, as adoption rates increase, every insurer should begin their education process to understand and evaluate the potential of these powerful tools.

Dax Craig is president and CEO of Denver-based Valen Technologies Inc.

(c) 2008 Insurance Networking News and SourceMedia, Inc. All Rights Reserved.

Register or login for access to this item and much more

All Digital Insurance content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access