Leveraging Analytic Technology

For those of us who have been in the claim business a decade or two, we have witnessed the introduction of technologies that have had a profound impact on managing claims. Every few years, a promising technology packaged in a practical claim application comes along, and rapidly becomes an industry-adopted “best practice.”

These new waves of technology not only alter how claims are managed, but permanently impact insurers' bottom-line results. Telephonic loss reporting, medical bill processing, and bodily injury evaluation tools are just a few of many examples.

We are now witnessing predictive analytic technologies becoming the latest breakthrough technology. Can insurance companies harness its power? Furthermore, what do claim executives need to know in order to fully benefit from its potential?

The Power of Predicting

An advanced form of “data mining,” predictive analytics applies sophisticated analysis techniques to enterprise data resulting in the discovery of meaningful patterns and relationships in data. Identification of those data patterns leads to actionable decision-making information and, in turn, become a self-learning set of rules that impact how a company conducts business.

Industries best served by predictive analytics have large quantities of data that hold the key to future customer behavior. Unlocking that information can help a company target the right product at the right price using the right business channel. Banks were early adopters, as in the case of Capital One. In the mid-1990s, the company revolutionized the credit-card industry by building its entire business around intelligence that was contained and mined from customer data. Using predictive modeling, the company was able to target distinct segments of the market with unique products that significantly helped increase Capital One's revenues.

The gaming industry also has discovered the value of predictive analytics. Ten years ago, Harrah's Casinos discovered that their “high rollers” accounted for about one-fifth of corporate revenues, yet incentive programs skewed highly towards these high-spending customers. As a result of implementing analytics, Harrah's altered their reward programs to non-high-roller clientele; a clear majority of their revenue base.

The insurance industry has collected and accumulated massive amounts of data on customers, policies, and claims. But it has been slow to embrace the use of predictive analytics throughout the enterprise. Actuaries have used various forms of predictive models for years for pricing purposes. Underwriters at most companies are rapidly implementing predictive models based on credit scores or other quantitative methodologies. But the most wide-ranging and powerful application of analytics in an insurance company is in claims, especially given some of the historical challenges of working with claim data.

Legacy Issues

In the 1990s, insurance companies began to realize that there was enormous untapped value embedded in their massive stores of data. In order to make some use of this data, insurers first had to be reassemble it into a usable format. As a result, data warehousing projects were instituted throughout the industry as companies attempted to consolidate decades of policy, claim, and customer information into a single physical location and format. Many of these projects collapsed under the enormity and complexity of the scope. For those who did create a usable data warehouse, there remained the problem of extracting actionable business intelligence it.

Despite these challenges, the urgency to find ways to leverage technology for better business decisions has increased. In 2006, a Gartner study found that chief executives expected their chief information officers to move beyond concerns focused solely on the IT department and to increase their emphasis on helping the business units grow revenue, reduce costs, and improve profitability. With that directive in mind, a recent survey found that the top technology priority of CIOs is “business process improvement through business intelligence applications.” Translation: CIOs are looking to improve a company's competitive advantage through the strategic and innovative use of information, business processes, and intelligence in products and services. The challenge is how to apply this in practical ways, and to do that, you must understand the different categories of claim data.

Structured vs. Unstructured

Structured data is standardized, easily entered and handled information, such as numbers or company-designated codes for financial information, lines-of-business abbreviations, causes of losses, and the like. Structured data fields have a defined range of values and a fixed length. This type of data rarely includes details of the claim and what was discovered during the investigative process.

Semi-structured data has some structured characteristics, but the data may be irregular or incomplete. It doesn't necessarily conform to a fixed schema. Web-related information, such as HTML formatting and markup, and fields where a brief description of loss or a street address may be entered, are examples of semi-structured material. Traditional predictive analytic programs usually provide inconsistent interpretations of this data, at best.

Unstructured data is the non-standardized, freeform, explanatory information. It exists in a majority of corporate data repositories today. Examples of unstructured data in the insurance industry include adjuster notes, imaged documents, and e-mail messages. This type of data is a goldmine of rich, but hidden, information. Unlocking it and extracting its actionable business value has been a daunting necessity. Historically, companies have tried to gain access to this data in manual-intensive ways, extracting hard copy files from departments and having teams of individuals review them for specific information. This is an extremely time-consuming and expensive task that leads to minimal actionable results.

Unique Characteristics of Claim Data

Claim data is unique in many ways when compared to enterprise data gathered elsewhere in an insurance company. What is the biggest differentiator? The proportion of unstructured data stored in claim systems. Research groups have estimated that up to 80 percent of a company's data is unstructured, and therefore, inaccessible or extremely difficult to gain access to.

As staggering as the 80 percent figure seems, other independent research on P&C claim data found that more than 97 percent of data in claim systems is unstructured. It's not only the percentage that is stunning, it's also the fact that the richest information resides in the unique set of words, acronyms, and abbreviations that comprise the adjuster's notes. But adjuster notes are just a part of a claim department's unstructured data inventory. There also are e-mails with attachments, imaged documents, case manager notes, web-based information, recorded statement transcriptions (or digital audio files), and digital photos. All of this valuable data content currently exists within insurance companies' data inventories. Yet it's safe to say that MIS reports available to claim executives are extracted exclusively from the structured data fields that only comprise a sliver of the available information.

Figuring out how to use technology to read and interpret unstructured data has always been the Achilles' heel of analytics efforts in the claim area. The technical and scientific skill needed for any individual insurance company to accomplish this has been overwhelming. Some traditional text-mining technologies use grammatical parsing as a main technique, but it requires correct spelling, no abbreviations or acronyms, proper punctuation, and complete sentence structure, all of which are rarely excluded in most claim files.

Applications for the Claim Executive

There is an explosion in the number of applications that are being developed and implemented by adopters of analytics technology in claims. The predictive nature of analytics is potent. These applications can improve the management of claims and loss costs in three general categories: mitigation, avoidance, and recovery. By understanding the underlying patterns in large and troublesome claims and identifying these types of claims weeks or months earlier, insurers can dictate a more effective triage of the claim, and consider more case strategies to influence the outcome.

Which claims are most likely to evolve into large, complex cases? Which claims are likely to show a reserve stair-stepping pattern over the life of the claim? Which claims will involve litigation, and what will be the nature of the litigation? What medical-only claims have the greatest propensity to eventually convert to lost-time claims? What claims are demonstrating enough suspicious characteristics to initiate an SIU referral? These questions are just the tip of the iceberg of solutions provided by predictive analytics technologies when properly and expertly applied. The number of new applications is increasing at an incredible rate.

Not only can analytic applications be predictive in nature, but they can have powerful discovery and forensic analytic applications, too. Simply put, discovery-based analytics can help shed more light on what happened, in ways that structured data analyses cannot provide. For instance, how many claims involved an incidence of road rage? How many claims have been handled where a particular auto body shop was utilized? How many claims have been paid that may have been caused by a laptop battery fire? There are infinite queries of this type that can be solved in real-time.

Forensic-based applications clarify why something happened. The focus is on seeing trends and deviations that are difficult to detect. For instance, is there an uptick in claims that resulted from a flaw of a particular brand of household appliance, indicating a previously undetected manufacturer's defect? Is there a disproportionate spike in claims where a specific partnering of providers has occurred (such as a particular chiropractor and lawyer frequently working together)? What will be the next asbestos or lead paint type of exposure that could disrupt an insurer's profitability or viability? Which insurers will utilize forensic analytics to notice this trend in their claims first, and tighten their coverage exclusions before their competitors?

Total Loss Cost Management

Predictive analytics technology is a potent fit within the claim function and particularly effective in managing loss costs. To harness the full power of analytics, companies must ensure that all of their claim data — structured, semi-structured, and unstructured — can be accessed, read, and analyzed. For too long, market pressures have short-changed claim executives by not providing them the right kind of actionable business intelligence. Now, predictive technologies have been adapted for the claim environment. As the technology is being rapidly deployed, claim executives are achieving unprecedented new levels of total loss cost management, similar to other waves of groundbreaking technologies deployed over the past several years.

Stephen Holcomb is the founder, president, and CEO of Full Capture Solutions and can be contacted at steve@fullcapture.com.