Data, the new uranium for BFSI?

Companies in the BFSI space today face unprecedented challenges in the wake of the pandemic. The traditional way of doing business has been disrupted completely, forcing them to accelerate their digital transformation journeys. The digital shift has also given rise to an unprecedented data explosion that is expected to drive technologies such as AI, machine learning, and the Internet of Things (IoT).

This data can add value to the BFSI ecosystem by helping them make sharper decisions, but it can also overwhelm. With IDC predicting a ten-fold increase in data globally by 2025, Nilesh Kulkarni, Senior Manager at Qlik India, spoke to key business leaders in the BFSI space at Data Sessions, a recent ET Live Roundtable Powered by Qlik. Together, they discussed critical topics such as the need for real-time data, data literacy and data security, as well as the present and the future of data in driving business growth and decision-making.

First steps: From Data Strategy to Data-driven Value

While COVID-19 forced enterprises to find newer ways to work with data almost overnight, not many understood the right way to store it. According to Sanjivan S Shirke, Senior Vice President (Information Technology) at UTI Mutual Fund, encryption of data — while at rest, in use, and during transmission — is the primary and paramount part of data operations for BFSI companies. Businesses should also ensure that the data is cleaned and structured correctly to give them the input they need to work with Artificial Intelligence Markup Language (AIML).

“If you want to experience a Ferrari, you need a smooth road. So, if you are using AIML, but your roads (data) are not structured, you will not get the best experience,” Shirke explained. “People only concentrate on the sales aspect. Nobody looks at how the data is geographically scattered and structured, and how it can add value to back-office operations. A well-thought-out, data-driven AI strategy can help BFSI organizations transform how they approach critical decisions, such as fund management decisions and trade. Businesses need to take the entire ecosystem along to build a strategic data architecture. Any data deployment should have buy-in from all stakeholders across the board, from the management to the end-user.”

The next obvious step is knowing how to use this data effectively. Chakra Gampala, Head of Digital Transformation & Analytics at Tata AIG General Insurance, explained that, for his business, the digital transformation focuses on staying relevant, as tech companies globally (like Tesla) are starting to explore insurance. “We cannot just be a manufacturer. We also need to have the right experiences and engagements with our customers. Data helps us achieve that,” he said. From re-launching its website to building apps for inspection and risk management, TATA AIG has remained focused on building a data-driven digital strategy.

“We are using analytics to identify what kind of products we want to offer to our customers, what is the right product to offer to a customer, what is the right sum insured, etc.,” he explained. One COVID-related challenge for the company was reaching rural populations, which requires low bandwidth mechanisms like WhatsApp. “We’re working very hard to find ways of delivering those. It is the next billion people we want to target,” Gampala added.

Exploring use-cases and working towards self-servicing data queries

Meanwhile, for some BFSI organizations, like the National Payments Corporation of India (NPCI), data, AI, and machine learning are already playing an intricate role in day-to-day processes. “NPCI is a digital-born company that processes almost half a billion transactions a day. We get all the banks’ data. We started building data lakes in 2016 and make extensive use of AI and machine learning for risk management and for predicting fraudulent transactions in UPI, RuPay products, Aadhaar-enabled payments, etc. We even use AI and assisted technologies such as machine learning for marketing and investment decisions. Around 80% of this work is about data processing and engineering,” said Soujanya Aluri, Head of Digital Transformation at NPCI.

The company is also trying to use AI and machine learning to predict infrastructure failures and cybersecurity intrusions. Aluri said that the financial industry’s hesitation to move to cloud platforms that are providing cutting-edge algorithms could prove to be a bottleneck in the future. “We have to move into deep learning. We need to enable that capability and build talent that understands this technology as it is the way to go in the next two years. Cloud adoption will prove to be a fundamental shift for the industry,” she said.

Another BFSI company using data to drive innovation in everyday processes is HDFC Life. Its methods proved to be especially useful in tackling COVID-related interruptions. “We started using district-wise external data to plot sales and renewal collections against red, amber, and green districts. It gave us several insights and helped us plan our approach as per the need. For instance, if a location was in an amber or red zone, sales were doing well but not collections, or vice versa, this helped us understand what could the two teams learn from each other. Plotting internal business actions for those districts proved useful,” said the company’s Senior Vice President, Francis Rodrigues.

The organization recently completed a beta test for conversational analytics to allow sales teams to query the status of policies that they log in to the system. HDFC Life has used data analytics platform Qlik for a decade, making the product all-pervasive, with close to 15,000 people interacting with its dashboards every day. The digitization “helps turnaround time escalation and gives control of the data to employees. It can be used to build self-service modules, by teams like marketing, product, etc. as they like to extract their own insights, instead of (depending on) a central team that does business insights,” Rodriguez said. He added that data democratization is the future for all companies.

Back to the future: How data can provide a roadmap into the unknown

Leading BFSI companies are also trying to gauge the role data will play in the future. In terms of opportunities, Rukesh Patel, Chief Technology Officer – Credit at Edelweiss Financial Services, said his focus was on dynamic algorithms — “using data with machine learning to build the correlation that humans or business intelligence could not.”

Patel said the company has two areas of focus: credit decisions and predictive analysis. “Recent AI and ML models are allowing us to use voice or video files for emotional detection. So, when you get into a personal discussion for credit decisions, can you get into something like lie detection? It is easy to give a loan, but recovery is equally important. So, can we build early warning systems using publicly available information, like news, to do sentiment analysis?” He also spoke of analyzing someone with no, or very minimal, digital footprint. “China managed to crack the unbanked and the underbanked population and was able to rejuvenate the whole industry around it. That is the challenge for us to tackle.”

Organizations that manage to use data efficiently and innovatively can leverage this capability to drive growth, cut costs, and even prepare for unexpected challenges like the ongoing pandemic. It is crucial, therefore, that young and upcoming businesses steadily improve their data literacy.

“People say data is the new oil, but data can also be the new uranium – if not used properly, it will become radioactive. Data needs to be governed, cleaned, interpreted, and meta tagged – that requires hard work that most people underestimate,” said Abhay Johorey, Head of Digital Strategy at ICICI Bank. He added that businesses need to think big, but start small. “If you are imagining hordes of data scientists, that will not happen immediately. You will have to conduct a series of small experiments in which 8 of 10 will fail. But by distinguishing between failure of execution and failure of learning; what is working and what is not, over time, a business can see exponential improvement in whatever statistic they are following. You have to give time for data to accumulate and for patterns to emerge,” he said. According to Johorey, the future of BFSI businesses will be a culture of every decision being backed by data.

Concluding the conversation, Nilesh said, “The variety of opinions and perspectives about how organizations should manage enterprise data highlights the role that data is playing in shaping the future of the BFSI industry. The data and analytics applications discussed today made a few essential truths a bit clearer. Organizations, particularly those operating in the BFSI domain, can’t afford to not use data. Utilizing it effectively, however, requires a comprehensive and robust data strategy that starts small and works towards a bigger picture. Data governance is going to be critical, especially as the number of AI and machine learning use-cases continue to grow. By using data to drive their processes and decision-making, BFSI players have an unparalleled opportunity to augment human intelligence, thus creating incremental value and driving better outcomes.”

Disclaimer: The views expressed in this article are those of the author and do not necessarily reflect the views of ET Edge Insights, its management, or its members

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top