One sector that can benefit immensely from Deep Tech is the Banking, Financial Services and Insurance (BFSI) industry. The sector is increasingly turning to artificial intelligence, machine learning, and data analytics to ensure accuracy in decision making. In an interview with Dataquest, Subrata Das, Head – Analytics, U GRO Capital, shares his views on the future of Deep Tech in India.
DQ: Deep tech is a set of relatively new technologies. How has the COVID-19 pandemic impacted this technological shift in India?
Subrata Das: BFSI sector in India has seen rapid advancement in areas covering automation of processes, data processing and decision sciences. Today, it is unsurprising that loans are disbursed in minutes, systems are cloud based and ML/AI is shaping customer experience.
Covid-19 pandemic and the need for social distancing has made it imperative to change habits and adopt newer ways for the sake of sustenance. Businesses have lost customer footfalls, critical manpower and transport facilities had to take cost reduction measures and redesign the product offerings. From a lender’s perspective, the process to assess a borrower and extend the service – had to be reimagined over a matter of months. This has resulted in higher efficiency which will push the bar higher for good.
Specifically in the context of SME lending, the loan on boarding process is witnessing massive shifts. Given that need of social distancing will remain in the foreseeable future, organizations recognize the need to centralize many last mile physical processes. Several of these also require regulatory sponsorship – such as KYC, signing of legal agreements, collecting repayment instructions, payments and on-site verifications. A big step in SME lending today is the process of Personal Discussion, commonly referred to as PD, wherein an experienced Credit Manager visits the customer premises for a face to face discussion and physical inspection of the facilities. Digitizing many of these steps will require use of wide sources of alternate data, advanced algorithms and AI/ ML systems.
DQ: What are some of the trends we see in BFSI/Fintech as far as deep tech is concerned?
Subrata Das: Storage of diverse data and ML/ AI-based applications are being widely adopted and progressively refined to solve a wide array of use cases such as predictive sciences,recommended systems, reading documents through OCR, multi-lingual bots, Robotic process Automation. Blockchain tech has strong use cases and has been piloted for remittances or trade. Information security is a rising need in these times and also an active area of Deeptech research.BFSI and fintech have been beneficiaries of this, resulting in faster services and a growing coverage of banking services across markets.
DQ: Similarly, what are the challenges companies face in implementing Deep Tech solutions?
Subrata Das: The pace at which DeepTech will impact our businesses are related to a set of factors as well. Deeptech can sometimes have a gestation period and may require long term investments. In BFSI sector we are already witnessing the investments as well as several use cases which have proved tangible business benefits, it will be a journey to scale them up and apply universally.
Secondly, for the technology to be applicable in cross functional domains – several related ecosystems need to be developed simultaneously. For example, applying algorithm to predict customer default will be greatly bolstered by the advent of account aggregation. We have seen substantial Government sponsorship on digitization and technology rendering the ecosystem more conducive than ever.
Thirdly, adoption of any new tech also depends on human behavior and transaction habits. From consumer side, there are matters of trust such as providing account details over an online transaction. There are also habits like preference of cash v/s card. We have seen sweeping changes in digitization in previous years and this aspect is likely to witness massive shifts in the next generation of customers. At the same time, the industry also has majority workforce who are dealing with more than normal amount of change. Organisations and business models are taking large strides forward while balancing between progress and change management.
DQ: Kindly shed some light on the kind of Deep Tech that is being used in U GRO Capital?
Subrata Das: U GRO Capital puts data and technology at the core of the business model. We have consciously designed a “zero data loss” storage architecture, where every bit of customer data is stored for future analysis in a central repository. This is bolstered by the API partnerships where bank statements, tax reports and credit bureau reports are converted to machine readable data. We have invested in globally acclaimed statistics and machine learning software.
This enables us to process a loan application and potentially produce an in-principle decision in 60 minutes. This enables the use of proprietary scorecards customized to granular sub segments. We are equipped to leverage OCR and perform video-based KYC with features such as facial recognition.
We are a fast maturing analytics practice and making rapid strides in taking advance data science and ML/AI based applications to the market. We have developed an in-house ML deployment engine and are in final stages of implementing our first home-grown alternate data model for credit assessment. Work is underway to digitize processes that have always been physical – such as location assessment, alternate data-based fraud checks, facial recognition and object identification.
Today, we are benefitting from the advancements of ML/ AI, infrastructure and information security. As more DeepTech advancement touches the market, we would have the technical capability and organisational readiness to make the most of them for our businesses.
DQ: What according to you is the road ahead for Deep Tech?
Deep tech is gaining momentum and is likely to have a strong impact on our lives in days to come. One expects increasing productization of DeepTech and emergence of niche areas of applications in every industry.