Most P&C insurers across the country are at various stages ofupgrading their core claims management systems. A fortunate fewhave already completed the journey of an information technology (IT) transformation.

From automated workflows and straight-through processing, tostrategic allocation of claims resources, these 21st century claimsmanagement systems are helping companies gain a competitive edgethrough the use of modern technology, enhanced key performance indicators, and keyperformance predictors. However, for many insurance companies, theclaims IT transformation journey is just beginning. Numerouschallenges can exist, such as gaining C-suite support for thesignificant investment, unraveling complicated legacy systems,competing IT resource priorities, the complexity ofresearching/selecting a processing system vendor, and determininghow to connect with the cloud computing evolution.    

At the same time, it is important to note that a number ofcommercial and personal insurers have effectively leveragedpredictive modeling on the underwriting side of the house toenhance their pricing and risk segmentation. For claims, however, asmaller percentage of insurers have been able to leveragepredictive modeling to enhance their ability to identify therequired claim for triage, the required resource for assignment,and the required time for targeted intervention (at first notice ofloss). To the extent that, claims organizations have the "hood" ofthe claims IT car open. It is vital for them to begin consideringhow advanced analytics can eventually play an important role in theclaims processing system transformation. This article discusses howinsurance companies can bring predictive analytics along for theride. Business Problem-Solving: Buildingfor the Future
Insurers can start the journey by thinking about the businessproblems the organization would like to solve today, in threeyears, and possibly five to 10 years down the road. Although an organization maynot be ready to implement a claim predictive modeling solutionimmediately, it is important to consider the impact that advancedanalytics could have on the claims organization and the claimshandling process in the future. As analytical capabilities grow andcontinue to be adopted, market forces and competition will likelymake predictive modeling table stakes on the claimsside. 

It is important to fully vet and understand the analytic horsepower of the organization today, and where the organization wouldlike to be in the future. The same current state and future stategap analysis should be performed on data collection; the metricsmonitored and claim handling processes. A clear vision helps lay afoundation for analytics early on in the technology integrationprocess.  This is especially important as the mosteffective time for planning and design is at the start of thetransformational process. Done effectively, analytics can have agame changing impact on various aspects of the claims lifecycle:claims assignment, special investigative unit (SIU) referral,medical case management, litigation, subrogation, escalation and,ultimately, claim settlement.  

The Value of Data: Do it Once, Do itRight!
As claims organizations embark upon their respectivetransformations, there is no time like the present to address thedesired future business state and the data capture necessary tofacilitate analytics. With the appropriate mix of internal andexternal data, predictive modeling can help insurers targetreferrals, assign claims at first report/notice of injury/loss;identify the most appropriate resources; alert case managers tosignificant deviations in treatment; decrease SIU referral time;and ultimately reduce claim duration and loss costs. From aninternal data capture perspective, the following questions areimportant:       

• What data can currently be accessed electronically?
• What data is currently paper-based, but should be capturedelectronically?
• What additional data should be captured going forward?
• What performance metrics could be used differently to help runthe business better? 
• What new or future performance metrics should be captured goingforward?

These questions should be considered during pre-planning andintegration design for a new claims management solution, which willtypically be able to invoke business rules based on captured dataand in part, further facilitate claims managementcapabilities. 

It is imperative to consider how data will be used whiledesigning the future state operating model and claims businessprocesses. The good news is claims organizations already have muchof the data they need to begin building effective claim predictivemodels. Deloitte's model building experience to date suggests thatup to three quarters of good predictive variables come frominternal insurance company data. The remaining predictive variablescome from a number of external data sources.

The key is to know which data and what combinations can driveoutcomes, and to know this information as early as possible in theclaims management process. Supplementingtraditional claims and other internal data with new external datacan help to provide more effective claims segmentation. The amountand quality of external data including zip code level, census blockgroup level, and household level information continues to growrapidly. From data about a claimant's lifestyle, to demographic andsocio-economic data, external vendors and data aggregators arecapturing much of what is now publicly available. 

By capturing a few more demonstrated data fields and integratingthis information into models sooner rather than later, a model canprovide greater predictive power by leveraging 50 to 100 variablestaken from background information about the employee/claimant, theinjury, external public databases, medical data and other externalsources. In the end, predictive models can essentially act asdecision support "eye glasses" for the claims adjuster'smind. 

One other consideration involves data quality and readiness.There is no such thing as perfect data, and spending too much timeand resources to pursue 100 percent data quality perfection ismisguided. The more important focus is to find the most applicabledata and put it to work in the claim process as soon as possible.The saying still holds true, "80 percent of something is betterthan 100 percent of nothing."

Analytical Capabilities
In any future state, an insurance company leveraging advancedanalytics may need to demonstrate the ability to apply statisticalprinciples and repeatable processes to deliver high-impact claimspredictive modeling solutions. 

For some organizations, this can initially require leveragingthe experience of others who specialize in advanced analytics andpredictive modeling. Over time, however, some insurers will developa formalized data mining team, with defined job roles andresponsibilities. Roles may include a data mining department leaderresponsible for the organization's most critical analyticsinitiatives, communication to the C-suite, and the ability to leada cross-functional team; a development leader with the ability totranslate business problems into model design specifications;credentialed statisticians for developing model inputs andevaluating multiple candidate models; and programmers to help bringthe model scores and corresponding business reason codes to life.To the extent the organization has already implemented underwritingpredictive models, the deep statistical experience of theseresources can be leveraged to build out the claims predictivemodeling team.

In addition to model development, the claims organization shouldestablish the performance metrics and business intelligence toolsnecessary to measure the impact of the data mining and predictiveanalytics solutions. 

Once the claims models are fully integrated within the claimsbusiness process, performance metrics can be used to monitor theimpact that data mining is having on the claims life cycle (forinstance, SIU referral time, loss cost reduction, adjusterefficiency, re-assignment rates, and so forth).

Integration Scope & Planning
As is the case in most major technology initiatives, data designand implementation planning can be critical early elements toachieving the desired results. Data design and integration planningshould be looked at multi-dimensionally to endeavor to provide agreater return on investment. 

Design should consider specific elements supporting an enhancedclaims management solution and advanced analytics. Quite honestly,it is hard to consider one without the other in today's rapidlychanging world. While the proverbial "hood is up" and exposing theclaims engine, organizations should confirm parts needed toeffectively leverage the power of data, facilitate core claimbusiness processes and bring new analytical insights to claimhandlers (identifying the right resource at the right time) areinstalled. These new insights should be considered early in theclaim process, after the initial party contact, and ultimatelythroughout the full lifecycle of a claim. Critical data and analytical considerations should be at theforefront of discussions with the specific technology stakeholders,claims professionals and analytics/business intelligenceleadership. If approached from a multi-purpose point of view, thenclaims data and integration planning should serve many businesspurposes. With the proper planning, organizations can minimize therisk of incomplete or fragmented data requirements or partialsolutions, ultimately providing a greater return on investment forclaims technology initiatives.

Analytics Continuum
As recently as three to four years ago, the notion of advancedclaims analytics was largely conceptual in nature. Most claimsorganizations were considering internal analytics and performingcore legacy technology replacement as separate and distinctbusiness initiatives. Unfortunately, a number of companies weren'teffectively approaching the opportunity as an integrated program orportfolio of projects. When contemplating analytics enablement,some were primarily using core claims or basic claims data as theprimary input variables for establishing basic routing guidance andbusiness rules. At the time, this helped these companies buildbasic analytic capabilities, but likely left them short of wherethey should be to take full advantage of predictive modeling.

Instead, organizations should consider incrementally buildinganalytic capabilities while learning and adapting over time. Whilesignificantly improved over the past five years, a number ofleading claims management solutions being implemented today canlack true predictive analytics capabilities of their own. Insurersshould understand how advanced analytics can be integrated withthese new solutions. Now is the time to think about core claimsprocess enablement, more effective business metrics and how toleverage new core technology solutions to help analytics becomepart of the organization's DNA. The graphic displays an analyticscapability continuum. When embarking on the journey to replacecurrent claims management solutions, the topic of advancedanalytics should be 'top of mind,' as well as an important part ofthe early planning and design phases. Regardless of when aninsurance company actually implements advanced analytics in theclaims transformation life cycle, it is imperative to help set theorganization up for achieving the desired results. 

By fully vetting and understanding the analytics horsepower ofthe organization, data capture, and business metrics theorganization should be capturing to support a desired/requiredclaims future state, the company can lay a strong foundation foranalytics into the future as it changes its core claims technology.If approached correctly and from a holistic perspective, theinsurer can improve its planning and design efforts today, to helpbring analytics along for the ride into the future.

This publication contains generalinformation only and is based on the experiences and research ofDeloitte practitioners. Deloitte is not, by means of thispublication, rendering business, financial, investment, or otherprofessional advice or services. This publication is not asubstitute for such professional advice or services, nor should itbe used as a basis for any decision or action that may affect yourbusiness. Before making any decision or taking any action that mayaffect your business, you should consult a qualified professionaladvisor. Deloitte, its affiliates, and related entities shall notbe responsible for any loss sustained by any person who relies onthis publication.

As used in this document,"Deloitte" means Deloitte Consulting LLP, a subsidiary of DeloitteLLP.  Please see www.deloitte.com/us/about for a detaileddescription of the legal structure of Deloitte LLP and itssubsidiaries.

Copyright © 2011 DeloitteDevelopment LLC, All rights reserved.