Finding the right answers to these questions is becomingcritical for most insurers, resulting in insurers searching for newtools and strategies. One tool garnering a lot of attention andgaining acceptance in the marketplace today is predictiveanalytics. Used effectively, predictive analytics is able to assistinsurers in gaining and maintaining a competitive edge--adifferentiator that is becoming increasingly difficult to achievein today's market.

In the earliest stages of adoption, predictive analytics waspredominantly used in claims within the commercial property andcasualty space. Today, it is quite common to see insurers usepredictive analytics to support claims reserving issues as well asto identify fraudulent activities within active claims. Much of theimpetus for this predictive analytics adoption has been due to thecontinual rise of health care costs. Rising claim costs, coupledwith declining investment income and an economic meltdown, has puttremendous pressure on insurers to secure operating profits fromcore business activities.

Encouraged by the early success in claims, more insurers are nowturning to predictive analytics to support decision making focuseddirectly at underwriting disciplines in an effort to generate afair return from the business written.

The focus of insurers as they look at underwriting and policymanagement operations are primarily issues of understanding risk,pricing properly, and collecting the premium due. As companiescontinue to take a progressive approach to business (or at aminimum work to maintain the business that they currently have),predictive analytics is now often being factored into thestrategy.

Today, more and more insurers are using predictive analytics toprofitably grow market share, retain existing clients, minimizepremium leakage, flag potentially misclassified risk, price withbetter precision, and provide protection against adverseselection.

Those companies willing to invest in predictive analytics earlyon, when it was still considered new and unproven, are seeingdramatic results, which encourage them to expand its use andcontinue to look for new opportunities for predictive analyticsapplications. The use of credit scores closely aligned with the useof predictive analytics is one common and easily understoodexample.

In our own industry, the use of predictive analytics has playeda significant role in the dramatic increase of market share in thepersonal auto insurance vertical. In personal auto, wherepredictive analytics was first implemented in insurance, it's nocoincidence that the top 10 carriers grew their market share from56 percent to 67 percent between 1995 and 2009.

There is no question that predictive analytics will become a keycomponent and a major strategic factor for insurers competing inthe commercial insurance space. The question is "to whatextent?"

As we investigate the potential for predictive analytics inunderwriting, the options vary dramatically. Progressive insurersare using validated and sophisticated predictive analytics toexpand the dimensions of straight through processing capabilities.Additionally, the same insurers are using predictive analytics toscore policies, taking into account associated risks, and usingthese scores in conjunction with defined business rules to setprices, apply schedule credits, or place policies in the correctrate tiers and/or companies.

While measuring the ROI of predictive analytics is often morequalitative and subjective than quantitative, the benefits appearto be significant. And an attractive byproduct of the usepredictive analytics is the ability to avoid adverse selection bydeploying pricing strategies that are not easily recognizable bycompetitors.

If the personal auto insurance market is any indicator of whatwill happen in the commercial insurance space, it is likely thatthose insurers who invest early in predictive analytics will end upwith the lion's share of the available business. More importantly,they will improve profitability because they better understand therisks they are writing, price accordingly, and avoid unprofitablebusiness.

(Jim Haley is CMO of Valen Technologies and can be reachedat either [email protected] or (303)350-3730.)

For more information on how Pinnacol Assurance uses thepredictive analytics tool from Valen, click on this TechDecisions article.