Case Study: Regional Online Bank improves rewards personalization and loan default prediction with ValueMomentum

Regional Online Bank leverages machine learning using Python to personalize its rewards program service

Regional Online Bank, a financial services company in the USA, wanted to better understand account holder activity in its Rewards Program and how that activity related to loan behavior, including default risk. The bank turned to ValueMomentum and its Python-based machine learning capabilities, along with tools such as Anaconda Distribution, Python ML libraries, Flask, Tableau, and DataRobot, to gain actionable customer and loan insights.

ValueMomentum implemented reusable Python and SQL models to predict loan behavior, mine consumer insights, personalize the user experience, and centralize outcome data. The solution delivered web services, dashboards, and a deployment pipeline, reducing ALIP smoke testing from 4 hours to 20 minutes and helping the bank identify customers “gaming” rewards, improve rewards notifications, and flag high-probability default risk earlier.