Big data possibilities for insurers
Five opportunities for analyzing big data
Stuart Rose, Global Insurance Marketing Manager, SAS
Big data is changing the way companies go about their business, but not without a steep learning curve. Forty percent of financial services executives surveyed acknowledge the inability to turn data into valuable insights is a barrier that impedes data driven decision making. But from an insurers’ perspective what kind of insights are possible?
The goal depends very much on the insurance company, the market conditions and the strategic goals of a given carrier. It is not, however, simply a matter of capturing all the data. Analytics plays an important role in creating value from the data -- hence the growth in the term “big data analytics.”
Big data paired with analytics amps up the options for providing critical market advantages.
For insurers, the possibilities are quite intriguing. Insurers have always spent time and money crunching numbers to better understand who to insure, at what cost and how much to reserve for losses. Big data paired with analytics amps up the options for providing critical market advantages. Let’s look at a few areas in particular.
- Catastrophe modeling. A hurricane swirls onto land in coastal Florida. How will this impact your catastrophic modeling given that insurers must evaluate loss exposure and financial position to meet liquidity requirements, often in a real-time environment? By using data visualization, you can create geographical risk exposure reports by augmenting existing policy data with geospatial data to assess and monitor loss exposure by geographic region.
- Subrogation. It is estimated that, on average, 5 percent of claims that should go to subrogation don’t. By using data analytics and text mining techniques, you can minimize the number of missed recovery cases by recognizing known and unknown subrogation indicators in the claims information. One leading European insurer did this and was able to improve its recovery rate by more than 4 percent, representing millions of dollars per year added to its bottom line.
- Fraud detection. Fraudulent activities are on the rise. Unfortunately, if the fraudulent behavior is not discovered quickly, not only might it go undetected, but in the case of claims it might be too hard to recover the payment. High-performance analytics enables you to analyze data within you organizations to detect unusual behavior. You can also use analytics to look at external data, such as social media, to increase the likelihood of detecting fraudulent activities prior to a claim being settled.
- Telematics. Speeding, slamming on the brakes, harsh accelerating – all are associated with riskier driving. Telematics devices that measures these activities is gaining acceptance. But analyzing the volume of data to determine what is truly risky and what behaviors to reward isn’t easy. Data visualization software will let you visually explore billions of records/journey points and seek correlation on data sets to develop predictive models for accurately determining risk factors for pricing and claims. Hard brakes, for instance, was not known to be a risk flag until analysis of telematics data prove it.
- Product pricing. Actuaries have often relied on samples of historical data to run pricing models because the time it takes to prepare and run the models is too time-consuming. Today, actuarial departments are using advanced analytics like generalized linear modeling on the growing volumes of available data to enable more frequent variable exploration for finding subtle and non-intuitive relationships that can influence product pricing. For example, a regional workers’ compensation insurer found that account size and number of judgments or liens were very meaningful in terms of which companies would eventually file claims.
Organizations want more business value from big data and many insurers are realizing the potential for all the ideas listed above. But often they are frustrated by the technological hurdles. For more information on options for overcoming these hurdles, download the white paper, Return on Information – The new ROI.
Stuart Rose is the Global Insurance Marketing Manager for SAS. He began his career as an actuary and now has more than 25 years of experience in the insurance industry working for companies in the US, Europe and South Africa. Stuart has written many insurance-related articles and is also the co-author of Executive’s Guide to Solvency II.
- Learn how to get more out of your big data by reading the white paper, Return on Information – The new ROI.
- Find out how one insurer provides executives direct access to significant data volumes, so they can effortlessly analyze big data without IT assistance.