The property and casualty industry has not had a return onequity (ROE) better than the Fortune 500 since 1987. The last timesit was even close were 1991 and 1993. Even if the Fortune 500 isnot considered the right benchmark, the industry average ROE hasdeclined from a high of 12.7 percent in 2006 to an estimated 3.9percent in 2011.

Even more relevant, the industry ROE has been measurably lowerthan the average cost of capital since 2008. Running a businessreturning less than its cost of capital is rarely a sustainablemodel, as evidenced by the number of property & casualtyimpairments occurring since 2007.

Looking closer, private passenger auto (PPA) has been theconsistent premium giant, responsible for over one-third of theindustry's premium. PPA has been running basically at “break even”with a combined ratio hovering around 100 for several years. Withthe exception of inland marine and medical malpractice—both ofwhich have combined ratios notably below 100—the rest of theproduct lines are experiencing the same or worse in terms ofunderwriting results.

For those familiar with the industry, this is not surprising asunderwriting has accumulated losses of nearly $500 billion since1975. Industry profitability has been consistently coming frominvestment income, which unfortunately is at an unexpected all-timelow shaped by ten-year U.S. Treasury Notes running below fourpercent since 2008 and around two percent now.

Looking further at industry profitability quickly shows thatclaims, including loss adjusting expense (LAE), represent roughly80 percent of an insurance company's costs. Expense reductions areunlikely to provide enough savings, likely will impact service, andrepresent an often irreplaceable loss of expertise.

Rate increases are occurring and will continue, but givenintense competition, are likely to be managed rigorously. Thatleaves claims, a tremendous offset to revenue where the potentialfor improvement is rapidly gaining attention. Consider thefollowing, many fueled by the difficult economic conditions facedby consumers as well:

• The National Insurance Crime Bureau estimates insurance fraudhas escalated by 19 percent since 2009.

• The Coalition Against Insurance Fraud estimates insurancefraud costs Americans $30 billion a year.

• Insurance industry studies indicate 10 percent or more ofproperty & casualty insurance claims are fraudulent.

• One out of every three bodily injury claims from car accidentsinvolves fraud.

• U.S. insurance fraud runs an estimated 10 to 20 percent ofpremiums; in other markets like Brazil 25 to 30 percent.

• One in 10 Americans says they would commit insurance fraud ifthey knew they could get away with it.

• Fraud is costing the average household an estimated additional$300 per year

• Missed subrogation opportunities could represent up to $15billion annually just in the U.S.

Cost of Business

Historically, fraud has been difficult to identify and catchamidst the hundreds of thousands of submitted claims, particularlywith the pressures for shorter cycle times, higher customersatisfaction, and lower staff expenses. As a result, it is oftenconsidered a cost of doing business to be pursued responsibly.

Given the industry's current financial situation, the bestsolution to the ROE problem would seem to rest in claims, combiningaggressive pursuit of fraud with close operational management. Thatsaid, the focus cannot be reducing the number of legitimate claimspaid. Claims payments represent the delivery on the promise, aninsurer's most important opportunity to achieve customersatisfaction and competitive differentiation.

Fast, easy and efficient should be the mantra of theforward-thinking claims operation. This is especially true intoday's world of easily accessible choices and the amplifiedmessaging provided by social media.

Technological advances including predictive analytics and textmining have contributed to an insurer's ability to more efficientlyand effectively process claims. More specifically, whether dealingwith dishonest body shops, unscrupulous providers, ethicallychallenged consumers, construction scams driven by the manycatastrophic events or well-coordinated, cross-country schemes, theuse of predictive analytics to detect and escalate potential fraudis providing significant improvement.

The results include fewer false positives as well. As acontrast, consider how traditional rules-based systems would tagclaims that met set criteria for review. Once a “hole” in the ruleswas found by payment of a fraudulent claim, that “hole” would getaggressively exploited until a new rule was put in place.

Predictive analytics operate much more intelligently. Instead ofstatic rules, claims are evaluated based on all available historyfor all involved parties and service providers along with thestored outcomes, and then enhanced by correlating millions ofsimilar claims. As the number of claims decisions, involvedparties, service providers and characteristics increase, predictiveanalytics learn and adjust future outcomes.

Another advantage to this automatic adaptation is the ability toretrospectively revalidate claims, potentially identifyingpreviously undetected cases of fraud based on links to the newcase. Fraud schemes involving “cash for crash” and intentionalcollisions are more quickly surfaced via this type of linkanalysis. Some of the obvious immediate benefits include:

• A more efficient and wider “net” that quickly identifies morefraudulent claims with fewer false positives.

• Faster escalation of questionable claims, reducing oftenunrecoverable interim payments.

• Dynamic adjustment to new fraud schemes with the ability toretrospectively filter approved claims.

• Focused and more efficient use of SIU expertise, measured byhigher savings per salary dollar.

• Lower unrecoverable payments by prioritizing investigationsbased on probability and financial impact.

• Improved customer satisfaction resulting from significantlyfaster payment of “clean” claims.

Real Benefits

According to estimates from FICO, the functionality representedby these new tools could find up to 50 percent more fraud than thetraditional rules-based tools. Based on the cost of fraud mentionedearlier, and the fact that claims and LAE represent 80 percent ofrevenue, these kinds of results are clearly a materialimprovement.

As an added benefit, insurers that have successfully integratedpredictive analytics into their claims operation are also seeing anincrease in the consistency of how claims are handled—always achallenge in large and/or geographically dispersed centers—and evenmore notable is an increase in accuracy of loss reserves. Givenboth the impact of error and the resulting attention given to lossreserving, improved accuracy as a side benefit to claims analyticsfurther strengthens the importance to rapid adoption.

Claims subrogation and recovery represents another area wherepredictive analytics has been successfully deployed with meaningfulpositive results. Often, these opportunities are buried undermounds of data and claims payment activities. Equally as likely iswhen the indicators of opportunity are in handwritten notes in thefile.

These challenges are compounded by the complexity and fluidnature of the rules and regulations that guide these efforts. Theresult is usually either not enough resources to pursue, a missedidentification of the opportunity, or a long, drawn-out manualprocess that ends up taking up more time and effort than therecovery brings in.

By combining predictive analytics with text mining tools, theseissues can be greatly reduced and the value of recovery effortssignificantly increased. By outlining, and updating as necessary,the rules surrounding subrogation and recovery as well as thetriggering conditions, cases fitting recovery parameters areidentified early and efficiently for marking or discounting.

Late stage opportunities are found via text mining that searchesthe specific note fields for trigger words, bringing cases forwardthat fit the definitions. A feedback loop ensures that bothmechanisms—predictive analytics and data mining—constantly updatetheir search criteria based on successful discoveries.

Once discovery is profiled, predictive analytics can provideinsights into which path and vendors will likely lead to thequickest recovery of the most funds. Much of the tedious manuallabor expended in looking for the opportunities, browsing thenotes, determining the applicable options based on regulatoryparameters, and selecting the right vendor(s) is eliminated,leading to increased recoveries and reduced total claims costs.

Be Efficient

Improving operational efficiency represents an opportunity forleveraging predictive analytics. The size and cost of the claimsdepartment is typically a significant investment in extremelyimportant intellectual capital. Optimizing the use of thisexpertise will not only increase employee satisfaction but willimprove productivity, consistency, and customer satisfaction.

The application of analytics tools forresource scheduling and work allocation has been proven to have apositive measurable impact on cycle time, staff-to-claims ratios,and average staff cost. The predictive element comes into play inextrapolating not only the arrival and complexity pattern for FNOL,but for subsequent documentation, field and customer follow-upcontacts, and escalation patterns to name the main keyvariables.

Staff skill sets, schedules, productivity “sweet spots”, andsecondary expertise are then integrated with the projected workpatterns to calculate the optimal allocation of individuals andskill sets by time-slice. Practical rules have to be incorporatedto prevent every-other-hour schedules, while at the same timeallowing for split-shifts and shared part-time as allowed by thecompany.

While there are a number of call center products that model andmonitor call arrival patterns across skill-sets, the intention hereis to take that concept and extend it across a broader range ofmixed complexity, alternative source work activities. The morecomplex the model, the more efficient the eventual use ofresources; and as an indirect benefit, the more satisfied thosesame resources, and their customers, are as a result of the closerpairing of time and need to availability and skill.

Additional noteworthy advances are being made in claimsanalytics tools beyond fraud detection, claims recovery, andschedule optimization. Text mining is one in particular that seemsto be gaining the fastest traction. Simply put, text mininginvolves searching unstructured data like handwritten notes,translating findings into data that can be meaningfullymanipulated, and then either operating against the results togenerate predictive outcomes or incorporating them into theapplicable database.

In March, Text Analytics World was held with an impressive rangeof topics and real-life applications for both text analytics andthe expanding field of sentiment analysis. Specific to insuranceclaims was Accident Fund's presentation on the use of textanalytics to extract information about claimantco-morbidities—particularly obesity and diabetes—from workers'compensation adjuster notes.

Here, the data did not exist in digital form, but it had beenproven to correlate to claim outcomes and total costs. Given thatthe average combined ratio for the workers compensation industry isestimated to reach 120 this year—shored up by a continuing increasein medical claims costs—the ability to leverage text analytics indetermining and better managing these claims could prove anextremely valuable application.

Top Challenges

According to a 2011 study by Bloomberg, the number one challengepreventing companies from more fully adopting analytics wasexecutive concern over data quality, acquisition, and integration.Within the insurance industry, it is likely that this challenge isoften expressed as the “legacy systems” barrier, that hauntinglyfamiliar roadblock to innovation and transformation. While a validconcern for insurers who have yet to either start or complete theirtransition to new systems, it is worth noting that analytics toolscan operate against data warehouses. And data warehouses can be fedby a multitude of old, new, external, and in transitionsystems.

Unlike many transformations that are utterly dependent upon asuccessful migration to a new system or modular replacement of alegacy system, analytics can be implemented across diverseplatforms of varying age. A well planned roadmap can be designed tostart realizing benefits in as little as six months, as experiencedat Chartis in the case study presented at IBM's Financial ServicesSummit in September of last year. Despite having many businesssegments and a variety of systems, a repository was used to capturethe relevant data, analytics tools were engaged for a very specificpurpose to operate on that data, and the project's first of manystages was implemented on time with positive impact from day one.At that same conference similar examples were provided by AIG,Oppenheimer, and Chubb.

The three key criteria in common across the all-senior-executivepanel, as well as several other panels and presentations since, areas follows:

Have a focused roadmap that clearly outlines what resources willbe involved, specifically where analytics will be used and when,what tools will be used, and how will success be measured. Theroadmap may change with time and knowledge, but requires from dayone executive sponsorship and business buy-in on who, what, where,when and how.

Use clean data that is comprehensive, accurate, and current;this does not mean that 100 percent of the data has to meet thesecriteria, as some have used as little as 70 percent of availabledata. The bottom line is that the data that is used should berepresentative and should meet these criteria. This is wherestarting small and building big has the greatest advantage asrepresentative subsets of data can be used with proper thought.

Staff with talented and engaged people who completely understandthe business problem and are proficient with analytics. Note thatone person does not have to have both qualifications; it can be ateam where some are experts on the business problem and othersexperts in analytics. Both skill sets have to be represented.

According to an IDC study, the average ROI fora predictive analytics project was 145 percent and a nonpredictiveanalytics project was 89 percent.

According to research by the Aberdeen Group, insurersintegrating predictive analytics into their key decision processesachieved a one percent improvement in profit margin and a sixpercent improvement in year-on-year customer retention versus a twopercent drop in profit margins and a one percent decrease inyear-on-year customer retention for insurers who did not adoptpredictive analytics.

No matter how it is measured or what study is reviewed, thebottom line is that the use of predictive analytics, particularlyin the claims area for p&c companies, can provide near-termpositive impact despite the challenging economic times. Looking atthe field of those already engaged in the process, it is clear thattomorrow's market share leaders are being defined by today'sactions.