Now Reading
Clix Capital’s Katerina Folkman On How AI & Analytics Are Driving New Lending Models


Clix Capital’s Katerina Folkman On How AI & Analytics Are Driving New Lending Models


Candid picture of Katerina Folkman, Head of Analytics, Clix Capital

Analytics and technology are key drivers to have a competitive edge in every industry today. In the lending industry, to be a profitable lender, the company must have a deep understanding of real customer behaviour, needs and motives. Clix, a contemporary lending firm also leverages technologies such as analytics extensively to dig these insights from the data for which they use in-house models and algorithms. Clix Capital is erstwhile GE Capital, currently rapidly growing in Retail and Corporate Lending space.

Driving this unit is Katerina Folkman who has been based in India for over six years, and heads the analytics initiative at Clix Capital. Analytics India Magazine got in touch with Katerina to traverse down her analytics journey and understand how is she using the technology to expand to unchartered territories such as new to credit. She leads the in-house experimentation, development and implementation of analytical models, including artificial intelligence and machine learning.



Here are the interview excerpts and her take on some of the most important factors at Clix:

On driving analytics and data science initiatives

KF: The team is focusing on both – Growth & Cross-Sell and Risk Analytics and covering the customer journey end-to-end. My role is also to connect Clix with innovative partners, start-ups, incubators and data providers across industries, for truly unusual solutions.

On analytics adoption and use cases

KF: We put curiosity and willingness to take risk as our core value.  So adoption of latest analytical algorithms and continuous experimentation are part of our DNA. In-house analytics engines powers every new product launch, it drives cross-sell and top-ups, and monitor customer experience.

One great recent example is the scale-up of our two-wheeler business, where we developed and automated in-house underwriting decision engine DELPHI. This engine enables the digital customer application journey at motorcycle dealerships. Key elements are fast decisions, minimal manual underwriting and use of alternative data to service even new-to-credit applicants.

On expanding product portfolio using analytics

KF: Product, segment and geography expansion are key mandates for the analytics team. We have our vision defined as “To explore new algorithms, to seek out new customers and new data, and boldly go where no man has gone before”. The latest example is SuperWoman Loan product – we target new emerging segment of urban female entrepreneurs. This fast-growing segment has a lot of unmet borrowing demand. We developed proprietary techniques that allow us to understand and assess these customers and their business ideas even if they do not have prior borrowing history.

On Risk Analytics and Collection Analytics at Clix

KF: Risk analytics is crucial. As we expand and grow, we must still maintain a balanced and profitable portfolio. In addition to underwriting decision engines, we created Early Warning Signals models, fraud analytics and differentiated collections strategies. Collections analytics is especially important for us as we scale up and move to high-velocity low-ticket loans. We work with the Collections team to design data-driven strategies for a behavioural segment of defaulters. These treatment strategies are based on likelihood-to-collect, likelihood-to-contact and importantly, on a long-term estimation of customer profitability.

On Machine Learning algorithms and using it at Clix Capital

KF: Machine Learning models allow us to move away from traditional scorecards to more dynamic customer assessment. Our cross-sell recommender engine is a great example. It is powered by thousands of data points not only on customer transactions at Clix but also on customer behaviour with other lenders and other digital footprint data. So we can predict exactly when a customer needs another lending product from us, and act quickly to capture this demand. We experiment with deep learning models and hope to move to AI space soon too.

On open analytics tools vs in-built tools:

See Also

KF: Our team operates on Python. We also build our algorithms from scratch in house, and the best example is our DELPHI underwriting decision engine. We do use open source platforms like GitHub and TensorFlow. And of course, we are always open to new innovative partnerships too, as our tech platforms allow API connections for plug-and-play experiments. The best example is one of the psychometrics engines that we connected from an international partner.

On the analytics roadmap at Clix Capital

KF: We continuously focus on capturing alternative data from every customer touch point. It makes our decision-making truly differentiated from the competition and allows us to take decisions faster. It is a key for winning the best customers, for innovating, for entering new segments.

We want to augment our data lake with unusual data sources, for example, drone and satellite images. We see us setting up cross-industry partnerships to expand our data access and incubating some innovating start-ups. In addition, deeper use of ML and AI for internal Clix functions is in our plans too, such as to enhance our customer servicing and HR decision making.

On her agenda for coming years

KF: We believe that the financial services industry must move quickly to adopt analytics innovations and meet customer expectations. Customers expect the same depth and agility from us, as they get from best e-commerce, telecom and transportation providers such as Amazon and Uber. The new generation of digitally savvy customers needs lending companies to anticipate their needs and to give flexible offers. “Thick file” document requirements becoming the thing of the past too, as the digital data gives much better, true view of the person applying for the loan.

We see “virtual reality” tools as critical for customer experience, for example, video, voice and psychometric analytics. Servicing customers through virtual reality channels will help us scale up and reach to broader customer segments faster. This is an exciting time to work in analytics and push the new frontiers of innovation.



Register for our upcoming events:


Enjoyed this story? Join our Telegram group. And be part of an engaging community.

Provide your comments below

comments

What's Your Reaction?
Excited
0
Happy
0
In Love
0
Not Sure
0
Silly
0
Scroll To Top