With the global economy spiraling downward amid some expertswarning that a recovery won't begin any earlier than mid-to-late2009, insurance providers worldwide continue to adjust theirbusiness processes accordingly to ensure that they not only arefiscally responsible, but also remain competitive in theseturbulent times.

Making the burden even heavier for insurance carriers is thealarming rate in which fraudulent claims continue to escalate. TheInsurance Information Institute projects fraud in the U.S. alonealready accounts for 10 percent of the property and casualtyinsurance industry's incurred losses and loss adjustment expenses-- or about $30 billion annually.

While insurance providers have placed a huge emphasis on effortsto combat fraudulent activity, they also must remain committed toother equally important business functions, such as marketingoptimization, claim-process efficiency, minimized underwritingcosts, and new opportunities for revenue growth. Given the currentstate of the economy, insurance companies will need to continueimproving their business practices across the enterprise, includingenhancing their technology investments. That's not an easytask.

Predicting the Future?

By embedding and automating predictive analytics across theenterprise, insurance companies acquire the ability to betterunderstand and predict -- in real time -- their customers' futurebehaviors by analyzing, modeling, and scoring demographic andtransactional data from operational systems, as well as attitudinaldata gathered through customer feedback and surveys.

The convergence of predictive analytics, business processes, andIT architecture signals a transformation in the way insurancecompanies serve their customers. It marks a fundamental shift inthe way value is measured, from an emphasis on products andsolutions to actual experiences.

TowerGroup, a leading research and advisory services firmfocused exclusively on the global financial services industry,believes predictive analytics must play an integral role in the wayinsurance companies now do business. In a recent report on theinsurance industry, TowerGroup revealed that "data management andpredictive analytics are no longer merely 'nice-to-have' technologyinitiatives. Carriers that fail to recognize this fact will seesignificant deterioration in their results, as well as plummetingloss of competitive position."

Predictive analytics empowers claim handlers, underwriters,insurance brokers, customer service representatives, and other keystakeholders across the enterprise with the predictive insight tocreate value from underutilized data from every channel to improvecustomer interactions.

With a new focus on data about people and added importance beingplaced on achieving customer intimacy, insurance firms putthemselves on a track to becoming a predictive enterprise in whichbusiness objectives are interconnected.

With the ability to capture this vast supply of customer data --both structured and unstructured -- insurance companies are able totruly know who their customers are and dynamically improveinteractions from underwriting to claims so that they never losesight of the customer's wants and needs.

Industry analyst firm Gartner, Inc., also recommends thatinsurers seriously assess their current IT investments. In herreport entitled, "Hype Cycle for P&C Insurance, 2008," vicepresident and analyst Kimberly Harris-Ferrante said, "For success,property and casualty insurers must continue to invest intechnologies to stay competitive and to differentiate themselves --especially in claims and customer service.

"Technologies such as claim management solutions, predictivemodeling solutions, GIS, and wireless claim applications will helpinsurers to improve these tasks and, thus, reduce costs and improvecustomer service quality and retention. Quality claim processesthat are efficient, rapid, and better managed, including real-timefraud identification and streamlined processes, will be a criticalsuccess factor by 2010."

Taking a Bite

Fraud remains the number one concern of insurance carriers. Astudy released in November 2008 by the Insurance Research Councilestimated that fraudulent claims and buildup added between $4.8billion and $6.8 billion in excess payments to auto injuryinsurance claims that closed with payment in 2007.

That's only one example of how prevalent fraud has become.Common forms of this illegal activity can also be seen in healthcare, property and life insurance claims, among others, and include"padding" or inflating actual claims, misrepresenting facts on aninsurance application, submitting claims for injuries or damagethat never occurred, and staging accidents.

Insurance carriers are doing their part to not only uncoverexisting instances of fraud, waste, and abuse, but also preventfuture occurrences. They are integrating predictive analytics intotheir business rules based on industry best practices to capturekey insights that can then be used to detect and minimizefraudulent claims -- in real time -- at each stage of the claimlifecycle.

A claim handler can not only extract value from standard yes/noanswers with a customer, but also receive detailed information andsentiments behind those answers to instantly determine those claimsthat qualify for immediate approval and those that are flagged assuspicious and require follow up.

Alabama-based Infinity Property & Casualty Corporation(IPACC), a provider of personal automobile insurance with anemphasis on non-standard auto insurance, deployed predictiveanalytics software and solutions to reduce its payments onfraudulent claims, while improving its ability to collect paymentsfrom other insurance companies.

Until predictive analytics was deployed, the identification ofpotentially fraudulent claims had previously been theresponsibility of claim adjusters who had varying degrees oftraining and used inconsistent practices. As a result, data relatedto suspicious claims was typically not gathered rapidly or fullydetailed. That was a major obstacle, as speed of investigation andearly gathering of key data are extremely important for claiminvestigators.

Predictive analytics significantly improved effectiveness ofthat process by automating the workflows and data gathering relatedto fraudulent and subrogated claims. With the technology nowdriving this critical step for the adjusters, IPAAC is able toquickly spot suspicious claims and forward them to investigators,who can then begin their investigations within days of the originalclaim being filed. They also have more accurate data with which towork.

Optimizing Claim Processing

It's estimated that claim management costs can be as high as 20percent of an insurance firm's operational expenses.

Without risk-aware customer profiles that include predicted lossratios and fraud-risk scores, insurance carriers not only exposethemselves to such losses from current customers, but also thepotential for acquiring new high-risk customers.

Predictive analytics addresses those concerns by combining datafrom acquisition campaigns with proactive claim risk optimizationto filter out prospects that match predetermined claim riskprofiles.

When new claims enter the system -- from any channel -- they areimmediately analyzed against risk profiles and either approved forprocessing or flagged for investigation. With this additional step,the productivity and accuracy of the entire claim-handling processis improved, from first notification of loss to settlement.

In essence, the claim process incorporates the same predictivelogic and automatic procedures used to identify possible fraudulentclaims. Predictive analytics pinpoints those claims that are mostlikely legitimate and should be processed quickly. This process isautomatic, so claim handlers don't even realize that they are usingpredictive analytic technology.

ALKA, one of the top five insurance companies in Denmark, relieson predictive analytics to fight fraud and enhance customerservice. The technology determines during the claim notificationprocess whether an incoming claim qualifies for immediate approvalwhile providing the call center agent with the appropriate actionto take.

This ability increases customer satisfaction levels throughfaster processing of payments and improved subrogation results byenabling an insurance provider to focus on claims that are morelikely to be paid back. It also ensures that honest and loyalcustomers are consistently presented with the best prices.

Becoming a Predictive Enterprise

Just how long the economy will remain in recessionary mode isanyone's guess. What is important is that the insurance industryacknowledges the impact it is having and continues to seek out waysto make the best use of current resources. That means embracing theconvergence of predictive analytics, business processes, and ITarchitecture. Doing so means that insurance organizations havebegun their journeys to becoming a predictive enterprise whereprocesses are intertwined and personalized relationships withmillions of customers are created.