By Dax Craig, Co-founder and CEO, Valen Analytics
For a work comp claim, the multi-step journey from injury to payout, and even proper medical attention, varies across insurers. Yet, the process generally contains similar steps that should be optimized. On one hand, insurers must be diligent in understanding the nature of each claim, including variables such as subrogation potential and length of time an employee should be reasonably expected to miss. On the other hand, injured employees who need their income subsidized are rarely interested in waiting until a carrier is “sure” everything is copacetic.
For many reasons, the friction in the system is inherent. Insurers meeting state and national regulations must follow various procedures, and in many cases, initiatives focused on digital transformation are still nascent across the insurance industry. However, this isn’t to say that there’s no room for improvement. Analytically-driven decision making is enhancing the claims lifecycle and can make a significant difference in further protecting reserves, closing claims more rapidly and improving overall efficiency.
But claims data presents a unique challenge in terms of building a predictive model; a considerable portion of the data associated with a claim is unstructured. Data from handwritten notes about details of an accident through prescription information must be captured and utilized in order to present a “full picture.” This can create a hurdle for insurers.
Some have approached this issue by enlisting the help of external data entry services, whereas others have begun to use image recognition software for data capture. At Valen, we have a tradition of incorporating a wide range of data, both structured and unstructured, the latter of which captures a more robust set of human insights and nuance. This data is critical to understanding and offering a deeper analysis of claims.
Once the data input issue is solved, there are four instances in the claims process where analytics can make a dramatic impact.
- Incident Report
Upon ingestion of the data from the incident report, most insurers should be able to assign an early score to the likelihood of payouts, length of time missed, and potential for a claim to become a jumper. Many carriers will use this information to determine which adjusters will be receiving the assignment. Similar to the use of data analytics in underwriting, this approach allows an insurer to have their best and most experienced adjusters handle the most complex claims.
- Initial Contact
Once an adjuster has been given a claim, they initiate contact with the injured party. At this point, insurers can reevaluate the accuracy of their initial score, and decide whether the claim will be low touch, or require further review. For those claims that appear to be low touch, many insurers are creating a similar approach to straight through processing (STP), where the claim is quickly paid out with very little further adjuster involvement.
- Billing/Additional info
If a claim is determined to be high-touch, the adjuster will naturally look for additional information. This can include options for subrogation, adjustments to an insurer’s reserve pool, or even an examination of medicines prescribed to a patient. The info gathered here can be critical in identifying potential jumper claims.
- Strategy Evaluation
Data-driven organizations of all shapes and sizes can look at their decisions in hindsight to learn from mistakes and replicate their successes. This can be achieved through a wide range of data analysis and machine learning components. In the case of claims, being able to understand risks earlier is critical. It means better handling of medical attention, more efficiency throughout the process, and ultimately, the ability to get people back to work more rapidly.
Many insurers are beginning to understand the value of incorporating data analytics into the claims process. A recent study found that a 1% improvement in reducing claims costs for a $1 billion insurer is worth more than $7 million to the bottom line. Those who arm their adjusters with insights generated from data at each step in the claims lifecycle can reduce the severity of problem claims and out of control costs.
About Dax Craig
Dax Craig is the CEO and president of Valen Analytics®, an Insurity company, and provider of proprietary data, analytics and predictive modeling for P/C insurers. Valen leverages its large contributory data assets to help carriers price insurance policies more accurately and achieve lower loss ratios. Prior to founding Valen in 2004, Dax was founder and CEO of Xertex Technologies, which was acquired by a global leader in the wireless antenna industry, Centurion Wireless. Dax proceeded to serve as vice president of global business development at Centurion, where he was directly responsible for global business development including sales, market definition, market segmentation, market research, strategic planning, and market development. Dax graduated from the University of Tulsa with a bachelor’s degree in business administration and marketing. He earned his MBA in finance from the University of Colorado at Boulder.
About Valen Analytics
Valen Analytics, an Insurity company, provides proprietary data, analytics and predictive modeling for property and casualty insurers. We work with insurers who are actively looking to utilize modern approaches to pricing, risk selection, claims triage, and premium fraud. Our customers are focused on increasing competitive pressures, fighting adverse selection with innovative solutions, and raising awareness for the impending “experience gap” with initiatives such as Insurance Careers Movement. Our customers span many lines of business including Homeowners, Personal Auto, Workers’ Compensation, Commercial Auto, Commercial Package, Commercial Property, and BOP. Learn more about Valen at www.valen.com.