My presentation was titled “cREating Opportunities Amidst Change” and essentially I shared information with participants about predictive modeling and the powerful impact it can have on business results.

The presentation was very well received, and it got me thinking about how so few insurance businesses truly leverage the benefits of predictive analytics to enhance efficiencies, reduce risk and anticipate trends.

I thought it would be helpful to share the highlights of the presentation on our blog and I would be very interested in hearing your feedback and reaction in the comments section below.

Data can be overwhelming

Technology is on steroids in today’s world. Information is being collected and analyzed everywhere on a variety of platforms, in countless formats and for unlimited purposes.

But how do you make sense of it all? How do you process and interpret this data in a way that is meaningful and reliable?

What is Predictive Modeling?

Predictive modeling is a process whereby historic data is analyzed and relationships between predictive inputs and outcomes are identified and quantified.

The conclusions drawn from this process are then used to predict future outcomes of new cases.

Today, predictive models are used across industries to anticipate patterns related to:

• Credit scores
• Credit card fraud detection
• Mail sorting strategy
• Weather predictions
• Food preferences

Simply put – predictive modeling offers meaningful forecasting insights that can help drive business strategy.

Modern Predictive Modeling & the Insurance Industry

Needless to say, predictive modeling has dramatic implications for the insurance industry, if it’s leveraged properly.

It can offer significant competitive advantage, reduce costs, enhance efficiencies and improve bottom line business results.

But surprisingly, in spite of its prospective impact, few insurance companies embrace predictive modeling as a viable resource.

The insurance industry is conservative and resistant to change, but those companies that integrate predictive modeling into their business strategy will surely pull ahead of the pack.

There is no doubt, and here’s why.

Predictive modeling impact

Predictive models can be used seamlessly and strategically to forecast trends and behaviors relative to:

• Underwriting
• Fraud Detection
• Claims Management
• Target Marketing
• Experience Analysis

Furthermore, they can be used to measure the robustness of models and the reliability of data.

Your turn

Have you considered using predictive modeling for your insurance business? Contact us to set up a call, and don’t forget to share your questions and comments below.

The Insurance Canada Conference focuses on technology improvements in the insurance sector that facilitate:

• Better workflow
• Smoother processes
• Higher-quality data
• Timely access to transactions and information
• Improved decision-making and planning

The power of analytics

Ian Bridgman, VP Sales & Marketing at Claim Analytics, took this focus a step further and along with Randall Day, Head of Claim Solutions at FINEOS, a leading claims management software provider and Claim Analytics partner, addressed the topic “Utilizing Your Data to Positively Impact Claims Outcomes”.

The presentation revealed how the scoring of Workers’ Compensation, in conjunction with a flexible workflow system (like the one provided by FINEOS Claims), can be used to generate return-to-work plans for injured workers.

And this is a consideration of which any forward thinking insurance company should be mindful.

Impact your bottom line

The session highlighted how using predictive analytics, in conjunction with a modern claims platform, can produce actionable tasks to improve outcomes. The presenters also discussed real world examples of insurers using claims technology and claims scoring to impact their bottom line strategically.

For more information about how claim analytics can help you to save money and drive productivity, contact us today.

Claim Analytics recognizes the foundational value of benchmarking, and in 2011, we conducted the Group LTD Benchmarking Study, which compares LTD recovery experiences across several insurance companies.

So how do you measure up?

Performance was measured as the difference between a company’s expected 24-month recovery rate and the average of the expected 24-month recovery rates for all companies, as measured on a standardized portfolio of new Group LTD claims. Expected recovery rates were determined using predictive models that were calibrated using each company’s actual LTD claim experience.

There were essentially three key findings to the study:

• A wide range of performance between companies
• Differences in performance by diagnostic category
• Differences in 30-month performance

A surprisingly wide range of performance between companies

The difference in expected 24-month recovery rates between the best and worst performing companies was 21%.

That’s pretty significant.

Expected recovery rates were for a standardized portfolio of claims so that performance for each company was measured on a like basis.

Obviously, the next step would be to understand why this broad discrepancy exists. What are the better performing companies doing to optimize their practice?

Using diagnostic category to enhance claims practices

There were wide differences in performance by company for some, but not all, diagnostic categories. These were the diagnostic categories where claims management practices tend to have a significant impact on outcomes. Performance did not vary greatly by company for other diagnostic categories. These were the diagnostic categories where claims management practices have less impact on outcomes.

The benchmarking report can help companies to identify the diagnostic categories for which their experience is relatively strong or weak.

Consider, for example, the two most prevalent diagnostic categories: musculoskeletal and cancers:

• Musculoskeletal claims account for 30% of all claims analyzed. The range of performance by company for musculoskeletal claims is much wider than the range for overall results. This indicates that musculoskeletal claims are a diagnostic category where claims management practices can significantly influence results.
• Cancer claims, on the other hand, make up 17% of all claims analyzed. The range of expected outcomes by company for cancer claims is much narrower than the range for the overall results. This indicates that there is less opportunity for claims management practices to impact outcomes for cancer claims.

30-month performance

The benchmarking study also examined results for recoveries up to 30 months from the benefit date. On average, an additional 3.9% of claims were predicted to recover between months 24 and 30 with a predicted recovery range of 1.6% – 6.6%.

Companies with lower predicted recovery rates in the first 24 months tended to have higher predicted recoveries between months 24-30.

So how does your company measure up? What are your thoughts about this study and its findings? Share them in the comment section below, and if you found the post helpful, please take a moment to use the social sharing buttons and tweet and like!

You can read more about our benchmarking study here and learn what best practices you can implement to improve claim procedures and experiences and enhance bottom line results.

“This predictive modeling project with Claim Analytics has been a real eye-opening experience for us, it is amazing what is buried within our claims’ data” said Scott Brener, Senior Vice President and General Counsel at SFM Insurance.

“One of the surprises for us has been the value of using text mining on our claim case notes”, Brener went on to say.

Text mining is a technique used in advanced predictive models to enable data in free text field to be used in data analysis, something which has previously been difficult if not impossible. “We are using leading edge techniques here at Claim Analytics which are producing great results for worker compensation insurers” says Barry Senensky, President of Claim Analytics.

Claim Analytics has been analyzing SFM’s historical claims data to build predictive models which enable new claims to be scored on their severity at an early stage in the claim’s lifecycle. “This enables us to understand and quantify claim severity without lengthy discussion or debate and take appropriate action quickly” says Meg Kasting, Vice President of Claims at SFM Insurance.

“The way we handle a claim in the first few weeks can shape the outcome significantly.  Our objective is to get our claimants back to work quickly, and the Claim Analytics tool can help us do that.” Meg went onto say.

Claim Analytics understands the significance of this kind of unstructured data, and the importance of text mining in providing even greater accuracy in model building.   “Claim adjusters are a very rich source of information and their notes document their observations, actions and opinions. With appropriate data privacy safeguards, text mining provides an automated methodology to review adjuster notes and uncover valuable information that can be applied to improve a predictive model’s accuracy in measuring claim severity” says Senensky.

About SFM Insurance:
SFM is the leading workers’ compensation insurer in the Midwest, serving businesses of all sizes and types for more than 25 years. SFM provides unparalleled knowledge, guidance and resources to help employers maintain safe work environments, restore injured employees to health and productivity, and minimize the costs associated with workers’ compensation claims.

About Claim Analytics:
Claim Analytics is an innovator in the insurance industry providing in-depth expertise in advanced predictive modeling to major insurance companies to assist with claim scoring, pricing, reserving and fraud detection. As a result, the Claim Analytics team provides innovative solutions to clients’ problems that improve their competitiveness in the market.

Contact Information: For more information on SFM Insurance, please contact Scott Brener at scott.brener@sfmic.com
or on 952-838-4200, 3500 American Blvd W, Suite 700, Bloomington, MN 55431, www.sfmic.com.