Without Good Data, the Outcome Is Wrong

Rules-based underwriting systems are designed to reflect the logic of the best and most experienced underwriters. In this way, they can provide agents and customer service representatives with the appropriate questions to ask applicants, and enable the company to more accurately and efficiently assess risk. But what if the data the applicant provides is wrong?"The problem with technologies such as rules-based systems or efforts to streamline information flow to agents is the matter of the quality of the information," says Daniel Finnegan, Ph.D., and president and founder of Quality Planning Corp. (QPC), San Francisco. "You can make all the rules you want, but if you don't get the data right, the outcome is wrong."

QPC analyzes auto insurers' books of business for rating errors-and also provides a call-center service to obtain the correct information from policyholders. QPC's proprietary method, called RISK:check, applies more than 100 audit checks to rating factors, including commute distance, annual mileage, vehicle garaging territory, unlisted drivers, business use and marital status.

It matches underwriting data to billions of public records in about 300 databases, and uses pattern analysis to identify questionable areas. For example, Finnegan says, "we know all the prison addresses in the United States. So if somebody's address is prison, we're really concerned about who's driving their car."

After reviewing California State Automobile Association's book of business, QPC pointed out that the San Francisco-based auto club was really good at underwriting the kind of risk that did not perform well, but not as good at underwriting the kind of risk that did perform well, says Kevin Coley, manager of analysis and performance management, auto insurance product management at CSAA. CSAA, which insures more than 2 million vehicles in Northern California and Nevada, began using QPC's auditing program five years ago.

QPC discovered that underreporting of mileage was a significant problem at CSAA, and missing drivers and business use of vehicles were also risk factors with significant errors, according to Coley.

In response to the audit, CSAA made adjustments to improve its underwriting practices. For example, the company began paying closer attention to new business. "Prior to that, we were just spot-checking new business, and errors got through," Coley says. CSAA also changed its flat rating plan to one that did a better job of differentiating between risk profiles.

"We've really gotten more discipline in our underwriting process as a result of working with QPC," Coley says. What's more, the auto club is reaping a 4:1 return on investment from using the auditing services, he says. "That's just in premium loss. We haven't even measured the impact of the program from a loss point of view."

For reprint and licensing requests for this article, click here.
Security risk Policy adminstration Data security
MORE FROM DIGITAL INSURANCE