Nearly eight months after Hurricane Katrina, many Gulf Coast residents and businesses remain displaced and are still looking to their insurance carriers to help them weather the proverbial storm. But just like many of their policyholders, some insurance companies were unprepared for the magnitude of the 2005 hurricane season.

Insurance companies are determined not to be caught off-guard again. They're re-examining IT strategies and operations in anticipation of the fast-approaching hurricane season. That includes improving the accuracy of predictive analysis tools and adding mobile devices and network security to their priority lists.

"Meteorologists are projecting that the severity of these storms will be an ongoing pattern this year and in the future," says Misty Crawford, enterprise program manager for Computer Sciences Corp. "With the 2006 hurricane season less than four months away, insurance companies are focusing on their IT strategies from all angles."

	Stuck with a mess after the levees broke in New Orleans Photo by UPI

Carriers are looking to boost the performance of their predictive modeling tools by ensuring that their data sources are clean, accurate, and accessible. This is even more important after widely used catastrophic risk models came under fire in the devastating 2005 hurricane season.

Following perceived failures of the models, some insurers claimed technology providers' modeling tools delivered outdated and ineffective potential loss predictions. Technology providers' rebuttal was that only accurate data produces accurate results.

To avoid similar pitfalls, insurance companies should regard risk models only as input in their decision-making processes, and not depend too heavily on them for predicting exact results. Models "represent solid thinking on a potential path and impact for the catastrophe," says Michael Costonis, a managing director at Accenture. "However, the insurer must contingency-plan around that and continually probe around the edges of the models."

Vendors report that some insurers are taking the right steps toward cleaner data for better accuracy. "Companies need to look at their data warehouses and interfacing systems and ensure that they're populated correctly," says Mark Vanston, a senior adviser at Hewlett-Packard. "Clean, accurate data will provide more-precise statistics and analysis."

Equally important, good predictive analysis will let insurers better strategize on workforce deployment. "Carriers can efficiently get adjusters to the right location in an efficient manner and service the policyholders," says CSC's Crawford. "Katrina and Rita just reinforced the importance of these predictions."