OVERVIEW
A global bank requires analytical support for a customer retention project.
OBJECTIVES
Identify and model top risk factors that contribute to client loss.
ASSUMPTIONS
Some customers pose a higher risk of leaving than others.
DATA
Bank customer data
SKILLS
- Big Data
- Data Ethics
- Data Mining
- Predictive Analysis
- Time Series Analysis and Forecasting
TOOLS
- Excel
CHALLENGES
The bank does business all over the world and wants to best understand what factors impact customer retention to develop new policies and practices in the interest of retaining those customers.
ANALYSIS
Customers between the age of 43 and 66 leave the bank at a much higher rate than all other customers.
Of the members who leave the bank, 29% of them are not active members, as opposed to 12% of active members.
Women make up 46% of the bank’s customer base, but 59% of customers who leave are women.
German customers make up 26% of the total customer base, but account for 37% of customers who leave. While France has a higher percentage of customers overall who leave, compared to the percentage of overall customers from France, it is proportionally smaller.
A decision tree model illustrates customer risk for leaving the bank.
RECOMMENDATIONS
Through qualitative surveys or interviews, determine why women, members of a certain age, and members from specific countries leave the bank at a higher rate than other customers. Adjust customer service and other policies to better serve those customers. Determine why some customers are active and others are not – encourage non-active members to become more active through outreach or events (such as online webinars) to ensure they stay with the bank.