Granting insurance cover is a complex process of assessing risks and evaluating claims. Insurers have to sort through large volumes of data to assess the risk involved in a single proposal for insurance cover. At the time of claim, the insurer must ensure that the claim is genuine and this again requires sorting through a sea of data. Experienced underwriters and claim investigators rely on their past experience to underwrite proposals or assess claims. New insurers, however, do not have this advantage. Big data can come to the aid of the insurance industry to help them sort through information and use it to their advantage. A report from PricewaterhouseCoopers L.L.P entitled “The Insurance Industry in 2012,” predicts that big data will ultimately redefine the industry, as insurers become innovative and find new ways to attract and retain customers.
Fraud Detection: One of the reasons why insurers read through pages of data is to detect fraudulent claims. They do this not only when claims are registered but also when insurance policies are taken out at the onset and also when reviving lapsed policies. Insurers are frequently overburdened with several claims and proposals, and they do not have the time to gather information and sort through it. Fraud detection therefore relies more on regulatory practices rather than facts. Predictive analysis using big data can help insurers spot frauds quicker. Data analytics tools such as text mining, database searches, modeling, and exception reporting can be effectively used in combination with regulatory practices to identify fraudulent claims faster.
Enforce Subrogation: Insures have to read through pages of information and often get entangled in large volumes of text. As a result, they could overlook pointers such as margin-notes for initiating subrogation or third-party fraud proceedings. This results in a direct loss to the insurer. With the text-mining tool, large volumes of text can be quickly processed to identify potential subrogation situations and prompt action can be initiated saving the insurer a good deal in claims.
Settlement: Insurers are hard pressed for time and small claims are often settled without probing too deeply into the facts of the case. This is done with a view to saving settlement costs. However, if these claims are fraudulent, or if the insurer settles a claim for higher than the actual loss to the insured, it will reflect on the ROI. To avoid such over payments and frauds, big data analytics can be effectively used to analyze claims routinely, history of claims, and trending to identify claims that are over the top or fraudulent. Using data analytics will also help insurers quickly settle claims and garner better customer satisfaction.
Fund Balancing: Insurers maintain huge funds against potential claims. It is virtually impossible to determine the size of claims or predict when a claim will arise. Optimizing the reserve fund is, therefore, a herculean task that requires constant assessment of risk. Insurers and actuaries spend a good part of their time in these tasks. Trending and predictive analytic tools can relieve this burden considerably and help insurers make informed judgments about fund balancing and optimization.
Litigation Costs: A good part of the insurer’s earnings goes towards lawyer’s fees for handling litigations. These litigations are often a result of poor handling of the case by inexperienced personnel. Predictive analysis using big data analytics can help insurers identify difficult cases and assign them to experienced people who can possibly settle the claim for lower amounts without going to court. Insurers can thus save huge amounts spent on lawyers’ fees and redirect these resources to more productive areas.
Underwriting: A large amount of data is required for underwriting insurance policies, particularly corporate insurance. When assessing corporate data manually, the possibility of human error is high. This data can be directly obtained from the proposer in digital form, thus eliminating human error in underwriting. For example, when considering insurance for marine or vehicle fleets, data can be directly obtained from the telematics devices in vehicles. Calculations can be automated so that the underwriter is presented with data in the format that he requires. This can significantly speed up the processes of underwriting, besides making it error free.
Premium Rate Monitoring: Insurers strive to lower rates of premium in order to maintain a competitive advantage. Premium rates are calculated based on the probability of claims occurring. These are complex calculations that insurers need to perform periodically to monitor their premium rates and adjust them if required. Big data can help the insurer perform these calculations quickly and conduct analysis to predict future premium rates. Insurers can, therefore, maintain their market edge at low costs.
Customer Segmentation: Premium calculations are based on the claim experience of groups of customers. Customers are usually grouped by the type of insurance they seek, their age, sex, and possibly the term of insurance, but little else. Within a single group, customers may present varying levels of risk. When the same premium is applied to the whole group, low risk customers often have to compensate for high-risk customers. With big data analytics, insurers will be better able to group their customers and calculate risk-base premiums.
Value Added Service: Insurers can offer other services to their customers in addition to insurance. For example, “RiskMark” is a service offered by FM Global in USA. This service enables those seeking property insurance to assess the risk to their property before deciding on the amount of insurance. MetLife, on the other hand, has created a “Wall” on which customer information can be posted. Agents can view customer’s information at a glance, enabling them to suggest products for customers to purchase.
Big data is revolutionizing the insurance industry. New players in the market such as MetroMile, Oscar, and Climate Corporation now offer new products such as pay-as-you-go using the power of big data to analyze risks and fix premiums says Ellen Carney of Forrester Research. Big data makes microanalysis possible at cheaper costs. Understanding minute details related to a customer or object that is to be insured helps analyze risk and fix premium. For example, in property insurance, understanding usage patterns, average temperatures in the area, the plumbing, electric, and other systems as well as the risks these systems pose will help the insurer better assess the risk of insuring the property. The insurer can also suggest ways in which these risks may be lowered.
Says Carney, every business wants to grow but organic growth is slow. In order to attract and sustain customers, business intelligence and data analytics can help the insurance industry gain in their business and this translates directly into increased ROI.
Download Whitepaper - "Building Quack - A Social Discovery App: Case Study"