DETERMINING MODEL TO IDENTIFY PROBABILITY OF CUSTOMER DROP-OFF: CASE STUDY IN BANK MULIA

• Main Author: Adhitia, Andrew
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/62829
• Institution: Institut Teknologi Bandung

Banks must comprehend the customer journey to ensure the company's long-term viability and the company's portfolio growth. Bank Mulia, an established bank in microfinance, needs to explore the Nurture Customer Journey to ensure its portfolio growth. They need to increase their retention ratio because there is a gap between customers who are performing well (~95 percent of the portfolio) and customers who drop-off with Bank Mulia (their retention rate is only ~60 percent). A business solution is required to assist the role of Community officers as bank Mulia intermediaries in reducing drop-off customers. The purpose of this research is to determine the model through customer segmentation and priorities by examining the causal relationship between customer information and the possibility of customers to continue financing with Bank Mulia, including potential strategies that are most likely to occur in the future by having this causality result. This research uses one of the Classification Models technique (Decision Tree - CHAID) because this technique generates a decision tree using chi-square statistics to identify the optimal split, and it can generate non-binary trees, where some nodes have more than two branches. The results of this study indicate that there are significant variables that are causally related to the possibility of customer drop-off, namely: times absence, times regular bi-weekly meeting, times deposit saving, Length of Relationship (LOR), total amount deposit, solidarity money usage, number of accounts, financing limit, and age of the customer. Thus, Bank Mulia knows its customer segments who have a High/Medium/Low possibility to continue their financing with Bank Mulia. They can prioritize to determine which Community Officer will approach the customer (Senior/ Junior Community Officer) and when they have to individually approach customers (at the end of the financing or several months in earlier).