Yes Bank to use data analytics to offer right mix of products

Team Udayavani, Dec 26, 2018, 9:17 AM IST

Bengaluru: With an increasing focus on retail customers, Yes Bank is now churning data to understand customer behaviour and offer a right mix of products based on a predictability model, a senior official said.

The bank has already diversified into retail segment and now harnessing data analytics to come up with models that help understand customer behaviour, Yes Bank, Business Analytics Head, Rajat Kanwar Gupta said.

“Yes bank started-off largely as a corporate-driven bank but with the version 2.0 coming in, we are very retail-centric. Right now our rank one objective, historically and continues to be, to always have sustainable liability generation. We are trying to focus upon building our CASA (current account savings account),” Gupta told PTI in an interview on the sidelines of two-day Yes Datathon on December 22-23.

As part of the programme, which was launched in September this year, Yes Bank reached out to the global developer, coder and data science community to work on its data stacks using curated, tokenised and anonymised data.

With focus on retail sector currently, keeping the corporate character of the bank intact, Gupta said data plays a big part to remain competitive in an environment where even fintech players have started to grab a big share in the financial services space.

“We want to compete with other banks or even with fintechs. It is actually going to be very disruptive. So, we have to think like that, we have to think that what our customers would want from us,” he said.

The bank has built a war chest of about Rs 120-150 crore for its data strategy up to 2022-23 and also plans to hire data scientists and experts who will be part of its TechTONic group of over 100 in-house experts.

It also has plans to induct part-time professionals.

Working on data analytics will also give the bank to read how the cybersecurity threats emanate and how it can be mitigated, Gupta said.

“As consumers, we are consuming a lot of data and in that process, we are generating a lot of data, so it has a snowball effect. We are also looking at data analytics to be playing a big role in security detection, fraud detection and threat prediction. So that itself is a big area for us to figure out and there is a lot of thought and idea going behind the scene,” he added.

YES Datathon saw the participation of over 6,000 data scientists, engineers and developers in diverse Machine Learning (ML) challenges.

The top models/products were based on predictive service delivery to enable the bank to provide proactive service, sales recommendation models, customer transaction relation graph, equivalent to a social network of transactions, fraud detection and analytics for card customer.

Models on customer sentiment analysis to predict customer engagement; tools to provide comprehensive products & services to customers and personal finance management tool without requiring access to customer SMS also remained at the top.

The top 20 models identified by the bank will be taken live within a month and the remaining will be moved to the bank’s product library to be developed and taken live within a period of three months.

“This will also be a first of its kind data product library which will be iteratively expanded to 100 plus products within 6 months,” Yes Bank said.