Big data and analytics are firmly entwined and they’re now part of everyday business vocabulary.
Data is the new business currency, the fuel that powers analytics to help insurers deliver improved processes and beneficial insights.
A recent report predicts that annual global Internet traffic will reach 3.3 ZB per year by 2021 or the equivalent of 278 exabytes (EB) per month. By comparison, in 2016 the annual run rate for global IP traffic was 1.2 ZB per year, or just 96 EB per month.
Traffic from wireless and mobile devices will account for more than 63% of total Internet traffic globally by 2021. Within the same period, the number of devices connected to IP networks will be more than three times the global population.
Without the requisite data integration tools and requisite skills, there cannot be any meaningful insight from all this traffic of information. This would mean insurers effectively conduct business blindfolded: lacking the ability to price risk effectively, check fraud waste and abuse, or the ability to retain customers, upsell, cross-sell and grow.
The scale of the problem
A 2012 piece of research by India Forensic concluded that life and non-life insurers in India lose a total $6.25 billion (or 8.5% of revenue) to frauds every year, and there is nothing to suggest that losses due to frauds have decreased. We at LexisNexis Risk Solutions have commented in a previous blog about the use of data analytics for curbing insurance fraud and mis-selling. But the use of analytics for spotting fraud signals across different datasets, whether industry contributory data or new data sources from Aadhaar or PAN, is only in the early stages.
The result is that underwriting profits are constantly weak across insurers due to over-competition and a fixation on prices. Genuine customers are made to bear the cost of insurers’ underwriting losses. According to one report, only one out of India’s 22 non-life insurers posted an underwriting profit in financial year 2016.
The fact is that no single insurer can have a complete understanding of a sales prospect, without having access to some elements of the individual’s relationships with competing insurers.
Other insurers could have had a good or a bad experience with the individual, which remains hidden unless the industry agrees to pool at least some elements of their data.
Pooled data: Give, to take back a lot in return
There is a need for the pooling the customer data – in the true insurance tradition of risk pooling – to create what is known as a contributory database. It is the way life insurers and general insurers come together in other market sectors around the world, to gain better visibility of their risk.
The same contributory principles apply across life, motor, health and other personal insurance lines.
Insurers mostly ingest data that is internally available today, basically the information customers provide (truthfully or otherwise) in the application form or claims form. Insurers accept the information provided by customers in good faith. The underwriting of risk, therefore, is done in good faith with just a little external verification, even in the case of existing customers.
So how would a contributory database help? What are the benefits of a contributory database and data standardisation for fraud, waste, and abuse detection?
Contributory databases collect a series of required data attributes on a real-time basis throughout the lifecycle of a policy. The datasets can include attributes such as claims, renewals and policy history, incidences of fraud or providing incorrect information, as well as gaps in cover and attributes specifically relating to the insured (the person, place and so on). All of this provides the basic raw material for analytics.
Applying analytics to a contributory database provides insurers with insights on sales prospects or customers in terms of quality of risk and the relationships across policies issued by all the insurers. For example, knowing about an individual’s motoring behaviour can be very helpful when underwriting for health or property insurance.
Insurers are then able to obtain the clearest possible picture of prospects and take faster and better decisions:
- Pooling data for fraud and waste identification broadens the scope of analysis to spot significant and revealing patterns only visible within large data sets
- The more data available to examine (such as claims history, information on life beneficiaries, nominees, instances of refusal of cover, gaps or cancellations, medical and prescription history), the more specific, accurate, and efficient will be the results in terms of pricing and meeting the customer’s needs
- For health insurance waste and abuse, contributory data gives the insurer an efficient, proactive means of identifying over-payments that are often otherwise hidden
- Contributory data, built on knowledge of multiple services and regions, can provide early warning signs of emerging fraud schemes in the market
- Contributory data can enhance already existing fraud investigations of insurance and health schemes
- Contributory data can contribute to a holistic view of the whole book of business, freeing up business capital for expansion that would otherwise be utilized for risk provision
- Using only claims data gives just a partial view. But joining of such data with the insurer’s own policy information and public records can give a holistic 360-degree view of individuals, health scheme providers or other specific areas of interest.
Healthcare is a particularly complex eco-system where contributory data is adept at detecting suspicious activity, such as instances where procedure codes reveal abnormal intervals between care, where suspicious patterns of sharing patients or patient identities exist, and where emerging at-risk providers are beginning to show anomalous activity.
A well-designed contributory database delivers geographical visual representations of clusters of abnormal activity and suspect providers. An outside investigative team can work on behalf of the insurers by gathering the data from them and approaching the health providers for any of the interested database community members.
LexisNexis Risk Insights for the life insurance sector
The current state of lack of symmetric information increases the possibility of adverse selection of risks. Contributory databases correct the information asymmetry to the largest extent. For the life insurance sector LexisNexis Risk Insights collates all the information received, and arranges it in meaningful formats for insurers to know the quantum of risk being underwritten and the correctness of the claim being settled.
It’s important to mention that a contributory database can bring additional system-wide benefits, in terms of data standardization and making data more powerful, scaleable and more usable across the whole industry.
Underwriting and claims information-gathering automatically improve in this process, as an information thread without even one required piece of information would have to be rejected, or reformatted, by the contributory database platform (according to the collective rules). This improves and standardizes data collection processes at insurers.
Insurers can find answers within insights derived from data, to provide more value for gaining trust of customers.
LexisNexis Risk Solutions’ contributory database platform fills the critical gap for an industry-wide customer database. Contributory databases make more analysis available to deliver a relevant customer experience, in terms of products, as well as services and distribution models. This ultimately helps insurers, their agents and bancassurance partners position themselves as trusted advisors.
Follow the link to the LexisNexis Risk Solutions India website to find out more about how we support insurers.