Top Data Science Applications in India’s Banking Industry

Top Data Science Applications in India’s Banking Industry

By Mohan Ramaswamy as Co-Founder & CEO of Rubix Data Sciences

The Indian banking sector’s competitive landscape will have a new player from this year. This will be the erstwhile PMC bank now acquired by a JV between BharatPe and Centrum Group. While M&As in the financial sector are commonplace, what stands out in the above case is that for the first time in India a FinTech company (BharatPe) has acquired a bank with the blessing of the RBI. This is a clear indication of how FinTechs are no longer viewed as just an extension of banking or merely digital infrastructure providers. This change in perception is driven by the explosion of the FinTech sector in India. According to VC firm BLinC Investment Management, India has the third-largest FinTech ecosystem globally with a very high adoption rate.

Therefore, it is no surprise that the Indian banking system is seeing rapid Digital Transformation. One of the areas of this transformation is the extensive use of data science technologies. The year 2022 is predicted to be a landmark year in the adoption and integration of data science into mainstream banking. Here is a look at the top areas where data science will help accelerate Indian banking:

Risk Management

Risk is an intrinsic component of lending. Therefore, it is essential to identify and quantify risk factors before taking lending, credit, or investment decisions. Data science driven risk analytics help organise and analyse unstructured data, which forms the bulk of a business’ risk-related information, and drastically reduce the probability of human error.

For instance, if a bank were to conduct the risk analysis of a potential commercial borrower before lending, data-backed smart tools can quickly analyse a vast quantity of the internal and external data about the borrower to provide insight into the business and its risk profile, as well as the track record of its directors or owners. Data-driven risk models can highlight the financial weaknesses of a business and provide a Credit Score and recommend credit limits. Based on the credit score generated by the risk assessment model, the lender can decide whether the business is creditworthy.

Even in cases of already disbursed loans, credit-risk monitoring tools based on data science enable surveillance and provide Early Warning Signals (EWS) about any deterioration in the business’ health. Such data-driven EWS tools provide dynamic credit scores that change automatically on the basis of the new data points that the tools have gathered. Banks can take quick action to reduce their exposure to a business in the event that the emerging data about it is negative and its credit score has dropped.

Lending Processes

With the spread of the COVID-19 pandemic, banks are having to conduct many of their processes online, and that includes the due diligence before lending. In such cases, banks can deploy data science to create robust risk algorithms that use data from myriad structured and unstructured sources, including social media. Data points collected for consumer lending consist of location, age, gender, income, type of employment, and so on.

For B2B lending, the algorithm uses firmographic data, identity, financial, compliance, legal, and financial data. The availability of a large range of data points helps banks better understand the behaviour of their borrowers, ensuring lower credit risk. The quality of lending improves significantly when data-driven models are deployed, as loans are approved on the back of objective data checks. Data-driven lending solutions help banks to identify and engage with the right customer profiles through the borrowers’ life cycle, thus improving their profitability. Digital banking, where all processes are done online, is also benefiting hugely from data science tools.

Minimizing Fraud

The flip side of digital transformation in banking is the increase in the quantum of frauds. As per RBI data, India saw over 229 banking frauds per day in FY 2021. There is a massive amount of fraud involving UPI transactions as well, most of it unreported. However, the good news is that data-science based Fraud Detection tools can analyse vast swathes of Know Your Customer (KYC) and payment transaction data to identify patterns of fraudulent transactions and flag suspicious activity. This helps banks in fraud prevention as well as in Anti-money Laundering (AML) activities.

Here is an example of a bank’s digital fraud detection tools in action: If an unusually high-value transaction takes place that is not in keeping with the account’s transaction history and the time and location of the transaction is unusual, it is red flagged. This transaction can now proceed only when the account holder confirms the details, usually with an OTP. In the case of new accounts, such digital tools can look deeper to see if multiple accounts have been opened within a short time frame using similar data; very often, this is done in order to facilitate fraud or money laundering. Flagging such instances results in more detailed KYC and AML checks of the account holders.

Customer Lifetime Value (CLV) Prediction

CLV is a metric to predict how much and for how long a customer will be of value to a business. Banks are increasingly using this metric to make projections about their business growth and profitability. CLV also helps banks to decide in which customer relationships they should invest. Data science tools help banks obtain a 360-degree view of each customer, as they can analyse the vast and varied data that the bank collects about each customer. This helps in more accurate CLV prediction.

Financial Inclusion Through FinTech

One of the main challenges that India and its banking sector face is financial inclusion. It also presents a tremendous opportunity for growth by improving access to finance for the traditionally marginalised sections of the population. In the past, the lack of access to this section of the population, the inadequacy of data to design suitable products for this demographic, and the resulting lack of confidence kept most banks, except PSU banks, away.

However, data and technology have changed all that by providing banks with the ability to leverage unstructured and alternate data to better understand the behaviour of various socio-economic segments and demographic groups.

India is among the few major economies of the world that have built Digital Public Goods (DPG). Popularly known as the ‘India Stack’, the series of volunteer-driven software platforms are central to the Indian Government’s digitisation programme and financial inclusion targets. The JAM trinity—Jan Dhan, Aadhaar, Mobile—has, in a short period of time, brought a large section of the population into the financial mainstream and put them on the Digital India bandwagon. The Direct Benefit Transfer (DBT) enabled by JAM is changing the face of finance in rural India.

India’s novel DPG architecture has laid the foundation for more inclusive financial integration with services in regional languages, tailormade insurance products, and services customised at the individual level. After the massive success of the Unified Payments Interface (UPI) that enables money to be transferred in under 6 seconds, many more exciting innovations in the banking and financial sector are on the horizon.

Conclusion

With the Indian Government’s massive push for financial inclusion along with digitisation of payments, data analytics has a huge role to play to boost revenue, improve customer experience, streamline costs, and predict risks for banks. There has been a tremendous explosion of data in the Indian financial services industry, and the adoption of data analytics is important to make banking more convenient and egalitarian while tailoring financial products to users’ needs. This will help meet financial inclusion and digitisation goals while accelerating growth and enhancing the profitability of Indian banks.

bankingdata analyticsData Science
Comments (0)
Add Comment