Since 2021, Tien Phong Commercial Joint Stock Bank (TPBank) has been using  IBM Cloud Pak® for Data (CP4D) to modernize the bank’s data infrastructure and enhance performance of its transaction channels. This modernization was aimed at meeting the diverse demands of the bank’s 12 million corporate and individual customers in Vietnam. It would also pave the way for TPBank to develop and deploy machine learning models to scale AI initiatives across the business operations.

Two years later, TPBank initiated a program to acquire new credit card users from its current customer base through its digital banking channel. However, the take-up and conversion rate were lower than expected. To address this problem, the TPBank data team developed Credit Card Propensity models using IBM Watson® Studio, IBM Watson Machine Learning and IBM Watson Pipelines. These models helped identify potential leads, increase conversion efficiency and ensure credit quality.

The implementation process consisted of four main steps: data preparation, feature engineering, model development, and model deployment. The bank’s data team processed up to 25 GB data from transactional structured data sources to build propensity models following the cross-industry standard process for data mining (CRISP-DM) approach.

By using CP4D, the bank reduced the development and operation time of new models by 30% to 40%. In 2024, TPBank successfully developed and deployed over 15 data science and machine learning models to support its business.