Infinity Automates Claims Management With SPSS

After merging six stand-alone companies into one, Infinity Property & Casualty Corp. ($1.92 billion in total assets) found that claims processing had become a bottleneck. Although claims submission was completely digitized, distributing and managing claims among more than 1,000 adjusters remained largely manual, according to Bill Dibble, the Birmingham, Ala.-based carrier's SVP of claims.

"Some carriers were adopting software to identify special investigative unit [SIU] claims," explains Dibble. "But I envisioned using predictive analytics to initiate a far more comprehensive strategy." During summer 2006 a cross-functional team hammered out an idea for a computerized scoring system to automatically evaluate claims and route them to the appropriate adjusters.

The team evaluated vendors through the first half of 2007. "Although none of them offered an insurance-specific solution beyond SIU, SPSS ... had just started implementing systems at telecoms and brokerages similar to what we wanted," Dibble recalls.

After purchasing the SPSS (Chicago) predictive analytics solution, which uses standard interfaces to harvest policy and claims information from Infinity's Oracle (Redwood Shores, Calif.)-based data warehouse, the insurer added two application servers to its mixed mainframe and Sun (Santa Clara, Calif.) Solaris LAN environment for analyzing and scoring the data. By fall 2007, a three-phase implementation was mapped out, Dibble relates. "We started with SIU and subrogation claims because the rules were well established and in use manually, offering us the fastest return," he notes.

From October to February, SPSS developers collaborated with Infinity to build out Phase One, which went live in March 2008, Dibble relates. "For the scoring system that forms our solution's core, SPSS worked with us rather than imposing a structure," he says. "We're already identifying double the rate of claims with fraudulent components. In addition, referral time to SIU is down from about 14 days to less than 24 hours. We're also identifying and addressing subrogation claims twice as fast, in 10 to 15 days rather than 26 days."

While implementation issues have been minimal, according to Dibble, an unexpected challenge was increased demand for certain claims expertise. This resulted in additional training for 10 percent to 20 percent of Infinity's adjusters, he says. "Going forward, we'll proactively align staffing using the claims trending data that SPSS generates," Dibble adds.

Upon completion of Phase Two, anticipated for late Q2 2008, the solution will differentiate relatively simple claims, such as glass replacement, and route them (through Infinity's existing claims processing application) to in-house agents rather than field adjusters, Dibble relates. Meanwhile, he says, Infinity is transforming its 100-person call center into a licensed, junior-level adjustment force. "Instead of 4 percent of claims staying in-house, we expect to drive that up to 20 percent," Dibble explains.

Phase Three involves increasing the granularity of data gathering during claim initiation, Dibble explains. Slated to finish in early 2009, this will improve claim routing accuracy and speed of settlement while providing customer service prompts, such as assisting with temporary transportation, he says. "Every time a customer is transferred, or we place a follow-up call to get more information, it's a pain point," Dibble stresses. "We want the first human point of contact to be efficient and the only one a customer makes."

Case Study Profile

company:

Infinity Property & Casualty Corp. (Birmingham, Ala.; $1.92 billion in total assets).

lines of business:

Property and casualty.

vendor/technology:

SPSS (Chicago) predictive analytics solution.

challenge:

Automate internal distribution and management of claims.

Anne Rawland Gabriel is a technology writer and marketing communications consultant based in the Minneapolis/St. Paul metro area. Among other projects, she's a regular contributor to UBM Tech's Bank Systems & Technology, Insurance & Technology and Wall Street & Technology ... View Full Bio