These days storing data is not an issue and the best use of such data is to analyse them for better decision making and strategising leading to more customer satisfaction and faster growth.
Data analytics has emerged as a major tool to influence various aspects of business in the 21st century. Digitalisation has not only facilitated business transactions with far greater speed and accuracy but has also led to generation of huge data of all the constituents of business and opening opportunities for analysing each of such stream of data for greater insight in respect of product design, product price, consumer’s taste and buying propensity and so many other issues impacting growth or profitability of a business entity. These days storing data is not an issue and the best use of such data is to analyse them for better decision making and strategising leading to more customer satisfaction and faster growth. In the business of insurance with each customer furnishing so much of data there is a huge inflow of information into the companies every day and later during tenure of a policy several transactions happen generating a pattern of claims or customer behaviour regarding repeat purchase or renewal.
But traditionally, the actuaries have been manufacturing products for the insurers on the basis of law of probability based on very broad data and they apply the past experience within the organisation or within a particular territory to the product development or the pricing process in a very generalised manner. The reason is not their inability to see through data but the fact that their vision always got impacted by lack of enough and credible data. Data analytics is a new discipline but the insurance industry has been a slow adopter of this new discipline of knowledge.
Data and motor insurance
Motor insurance is a classic portfolio in this context. This segment that contributes over 30% of premium to the non-life companies deserves to be hugely diversified on the basis of data analytics. In India, motor premium is determined on the basis of model, power of engine and age of the vehicle. As a result all the vehicle owners of similar model are made to pay the same premium irrespective of so many factors that may either increase risk or may dramatically reduce the risk.
For example, an owner driven car used within the city area only may be considered to be in a far safer position than a car driven by several drivers and often moving out on busy highways. The impression may be further strengthened on the basis of data regarding claims and the type of vehicles or drivers involved. Data analytics in this case may prove that there is a need to charge lower premium from those owners who belong to the former group.
This sounds too simplistic but data analytics expert would be able to decipher lots of messages within the data analysed in respect of age, life style and propensity to meet accidents. Data analysis even at superficial level may make us firmly conclude that a young man habituated to spend hours beyond midnight in restaurant and bar is more likely to crash his car on the streets than those youngsters who shun such habits.
You might also want to see this:
This example divides the customers into two groups, one which makes the company bleed more often and the other which passively keeps subsidising the benefit payable to the former. When applied to large data through analytical tools the preponderance of certain features lead to very precise conclusion about customer behaviour. A very large portion of motor vehicle claims are generally suspected to be fraudulent and they need to be very scientifically analysed so that the insurers develop mechanism to charge more premium from them or avoid to unduly charge heavy premium from those segments of customers who have been by and large honest while filing claims.
Data and life insurance
Life insurance also has immense possibilities to streamline its pricing methodologies by segmenting policyholders through analytics. Risk identification, potential fraud and economic profile of customers’ lapsing policies can be accurately mapped using such tools. Genetic profiling of policyholders may lead to vital information, more than health evaluation, which is the prevalent basis for determining premium. The loyalty cards issued by the airlines and retailers like Big Bazaar and Reliance Trends are tools for generating huge data everyday from the customers’ end which they utilise to design various offers and sales campaigns. Similarly, insurers can extend benefits to the policyholders in several ways if they move fast with analytics.
The writer is former MD & CEO, SUD Life, an Indo-Japanese JV.