The Story 

In the competitive customer engagement landscape, re-engaging dormant or sleeping customers is crucial to business growth. Union Bank of India (UBI) recognized the need to enhance its customer engagement strategies and increase the effectiveness of its marketing campaigns. By integrating a Customer Data Platform (CDP) with Machine Learning Operations (ML Ops), UBI successfully achieved remarkable results, including a 4x higher click-through rate (CTR) and a 26% uptake on their offers. This case study explores how UBI leveraged these technologies to reduce journey drop-off, cross-sell, and up-sell in real time. It also highlights how a credit risk modeling solution enabled a company to achieve a remarkable 69% open rate by sending real-time offers based on micro-segmentation. This approach enhanced customer engagement and successfully upsold credit products, all in real time.

The Problem

UBI faced several challenges in its customer engagement and marketing efforts, some of which are listed below (Marous, 2022):

1. Customer Retention

The bank struggled to retain customers, with high journey drop-off rates and limited personalization capabilities. They faced challenges in identifying potential cross-selling and up-selling opportunities. Retaining customers required a solution to harness their customer data effectively, enable real-time insights, and seamlessly deploy personalized campaigns. The bank needed a way to understand customer needs better and provide tailored experiences to keep them engaged.

2. Customer Reactivation

The bank faced the challenge of re-engaging dormant customers who had shown previous interest in their credit products but had yet to complete their application process. The existing approach lacked real-time assessment and approvals, resulting in missed opportunities and delays in engaging with potential customers. The company needed a solution that could identify creditworthy customers, segment them appropriately, and send personalized offers to re-engage and upsell credit products.

The Solution

To address the aforementioned challenges, Appice took the following actions as part of the solution:

1. Integrated Customer Data Platform (CDP)

Appice implemented an Integrated Customer Data Platform (CDP) to tackle UBI’s data challenges head-on. The CDP allowed UBI to collect, unify, and analyze customer data from various sources, including transactional data, online behavior, social media, and demographic information. This comprehensive view of customer profiles enabled the bank to create a single source of truth and generate actionable insights. By consolidating diverse data points, UBI gained a holistic understanding of its customers, paving the way for more personalized and effective marketing strategies.

2. Machine Learning Operations (ML Ops)

Leveraging Machine Learning Operations (ML Ops), UBI built predictive models that continuously analyzed customer data in real time. These models utilized advanced machine learning algorithms to identify customer patterns, preferences, and behaviors. Using ML Ops, UBI ensured that their predictive models were always up-to-date and accurate, adapting to changing customer dynamics. This real-time analysis allowed the bank to anticipate customer needs, optimize marketing efforts, and improve overall customer satisfaction.

3. Credit Risk Modelling

UBI implemented a credit risk modeling solution to address the challenges of real-time assessment and approvals. This solution leveraged advanced analytics and machine learning algorithms to analyze customer data, assess creditworthiness, and segment customers into micro-segments based on their credit profiles, behavior, and preferences. The credit risk modeling solution provided a detailed and dynamic understanding of each customer’s credit risk, enabling more precise and timely decision-making.

4. Real-Time Data Integration

Using real-time data integration, the credit risk modeling solution continuously analyzed customer information, including credit scores, income, payment history, and other relevant factors. This allowed UBI to make instant credit decisions and send personalized offers tailored to each customer’s needs and creditworthiness. By integrating and analyzing data in real-time, UBI enhanced its ability to respond quickly to customer needs and market changes, improving customer experience and operational efficiency.

The Impact

1. 4x Higher CTR: UBI created highly personalized and targeted marketing campaigns by utilizing real-time insights from the CDP and ML Ops models. These campaigns resonated with customers individually, resulting in a 4x higher click-through rate than their previous campaigns.

2. 26% Offer Uptake: The ability to identify potential cross-selling and up-selling opportunities in real-time allowed UBI to deliver personalized offers to customers at the right moment. This targeted approach led to a substantial 26% uptake on their offers, significantly surpassing their previous offer conversion rates.

3. Reduced Journey Drop-off: With the help of ML Ops, UBI continuously optimized customer journeys by identifying potential drop-off points and taking proactive measures to reduce them. This significantly reduced journey drop-off rates, enabling a smoother customer experience and improved campaign effectiveness.

4. 69% Open Rate: The company sent targeted offers to dormant customers by leveraging real-time assessment and approvals. These highly personalized and relevant offers led to a remarkable 69% open rate. Customers were intrigued by the tailored offers and were more likely to engage and explore credit products further.

5. Enhanced Engagement: The micro-segmentation approach provided a deeper understanding of customer preferences and needs. By offering credit products aligned with each customer’s unique profile, the company saw increased customer engagement and interest. Customers felt understood and valued, positively impacting their willingness to re-engage with the company.

6. Upselling Credit Products: The real-time assessment and approvals enabled the company to identify opportunities for upselling credit products to existing customers. By analyzing customer data in real-time and identifying potential credit needs, the company sent targeted offers to customers likely to benefit from additional credit products. This resulted in increased upselling success and revenue growth.

References

“Data & ML Operations – ProCogia.” Procogia, Inc, 29 Feb. 2024, procogia.com/data-science/data-operations-mloperations/. 

Marous, Jim. Banks Struggle to Provide Personalized Engagement: Here’s Why. The Financial Brand, 9 May 2022, thefinancialbrand.com/news/customer-experience-banking/why-customer-engagement-is-the-foundation-for-success-in-banking-143182/.

Shumarski, Alexander . “What Is a Customer Data Platform (CDP) | Progress Sitefinity.” Progress Blogs, 15 Aug. 2023, www.progress.com/blogs/how-to-choose-a-customer-data-platform-cdp.