Analytics in Banking and Insurance; Prospects and Challenges

By Prof Premchander

Professor Premchander Fellow IIMA: A visiting Professor at IIMA in the Finance and Accounting Area since 2009, he was a faculty of Finance and Control at IIMB from 1988 to 1997. Before and in between the above academic positions he has spent an equal amount of time in industry across SBI, Reliance, Accel Frontline, IL&FS, IL&FS Educational Service Ltd and lastly Mu Sigma Limited, a fast growing analytics company, as Vice President Operations. He has offered courses in management control systems, valuation, mergers and acquisitions and continues to have deep interest in the latter, financing large projects and venture capital. Prem Is also associated with a couple of schools where he volunteers his time at their management committees. In addition he is an independent Director at Yuken India Limited an engineering company.

As I walk into the ATM of any Bank and withdraw a sum of money I would have generated volumes of data for the bank. By identifying the location it is possible to map my travel pattern. By identifying the times of the day and my withdrawals it is possible to read patterns into my banking activities and spending habits. At another level the flow of customers through the ATM can in turn determine usage, waiting time and help make capacity decisions and also refilling decisions so that the ATM is never out of money and are efficiently located.

 Banking generates large volumes of data at a high velocity and in various often unrelated locations. A lot has been written about the potential for Big Data Analytics (BDA) in Banking. Already high fixed costs businesses, recent regulatory changes have pushed the fixed costs even higher making it all the more important to seek out profitable customers – that use the banks services and pay for it. In this search for customer’s banks no longer enjoy the luxury of traditional marketing paradigms of developing a product and searching for customers. The race is often to proactively assess what a customer wants and offer the service desired within the framework of the banks objectives, policies and the regulatory environment obtaining in that segment.

Driving revenue is not only about acquiring new customers but also about identifying new needs and crafting products to meet them. More often than not the customer is unable to articulate that need. A 360 degree view of the customer, personalized service, improved segmentation and targeting could be some of the benefits of BDA.. When a bank’s internal data is supported by third party data the potential could expand many fold.

Risk management has become far more critical in recent times. The events of the last decade have placed both regulatory and business pressures on comprehensive risk management policies. Risk modelling, predicting loan default, predicting fraud and identifying exposure to various segments are but a few of the areas where risk management could be critical. At a conceptual level all of them could use BDA to estimate and manage risk effectively. Banks are currently just taking baby steps in this area and the potential is huge. 

Research and strategy are yet another possible application that can grow out of BDA. At the level of individual customers analysis of data both internal and external could help identify high value assets and drive products towards them. 360 degree analysis and judicious use of external data could help reduce risk. (It could be possible to understand payment behaviour by buying data from telecom and utility companies). Such external data could embellish internal data.

 Banks with the storehouse of data that they possess would be well placed to provide a range of data based services for their clients. This could involve customer information, supply chain information and risk profiling of suppliers and customers on behalf of their clients.

With large opportunities, huge amount of data and pressure to manage risk and improve profitability one would expect greater penetration of BDA in the banking sector. Surveys show that while basic reporting tools are in place, in a majority of the banks, analytic tools are rudimentary and the application of predictive analytics is limited. In a recent survey a third of the bankers indicated that their organization did not even use analytics. 

It has to be recognised that in many situations there is often no one to one correspondence between the use of analytics and increase in revenue or reduction in costs. In addition with very tight operating budgets banks have little incentive to explore new operational technologies. Further, banking is a secretive business with much higher levels of security and reluctance to outsource.

 Banks may have to take the following steps to get the best out of BDA.

• Integrate the data being collected in various locations and coming in at high speed.
• Build internal resources in analytics to help interpret requirements from internal clients to external analytics service providers
• Look for small wins quickly to create a demonstration effect through an internal team.
  

The insurance sector, another financial sector, is in a slightly different stage of adoption and development. Traditionally, the insurance sector has used statistical tools to help in rate setting and risk profiling. Some of us were pleasantly surprised to read that hypertension and diabetes do not any more attract higher charges for health insurance. Such a measure could well also be arrived at from BDA. 

 

This sector has been slow in adopting predictive analytics, mainly because of the absence of integration in the large databases. As the level of adoption of data warehousing one should be looking to see even mid-size-insurance companies using more analytics in policy risk scoring, fraud detection, referral scoring etc. I am looking forward to an era when Indian insurance companies insure the driver rather than the vehicle. Of course that may need regulatory changes but careful drivers can hope to benefit.

 In the final analysis the financial services sector with high volumes of data and with data flying around at high velocity like banks and insurance companies are eminently positioned to benefit from the use of Big Data Analytics. The coming years, one can look forward to greater innovation in the use of Big Data Analytics in particular its integration with unstructured behavioural data and third party data, to get a 360 degree view of the business and customer.