Customer Churn Analysis and Insights
This project performs customer churn analysis to identify key drivers and potentially predict customer attrition.
Overview
This project focuses on analyzing customer churn at DigiBank, a fictional digital bank experiencing a significant rise in customer attrition. The analysis is aimed at uncovering key insights and trends behind customer churn, using data visualization and exploratory data analysis techniques.
It also serves as a simulation of a real-world data analyst task, complete with Tableau dashboard reporting, and is based on a dataset sourced from Kaggle. The goal is to generate actionable insights to guide the marketing team in crafting effective customer retention strategies.
Challenges
- Understanding and cleaning raw data with multiple inconsistencies and missing values
- Identifying the key drivers of churn from numerous customer attributes
- Communicating complex data findings through clear visual storytelling
- Generating business-relevant recommendations using only EDA (no predictive modeling involved)
Solution
A comprehensive analytical approach was taken to uncover the root causes of customer churn:
- Data Cleaning: Handled missing values, standardized categorical columns, and verified data quality to ensure accurate analysis.
- Exploratory Data Analysis (EDA): Used visualizations to examine churn by age, gender, customer activity, credit card ownership, complaints, and satisfaction scores.
- Business Questions: Structured analysis around seven key business questions to provide focused insights tailored to stakeholder needs.
- Dashboard Creation: Built an interactive Tableau dashboard highlighting churn metrics, complaints, and satisfaction scores for easy consumption by non-technical stakeholders.
Results
The project successfully produced insights that can be directly used by DigiBank’s marketing and customer service teams:
- Identified that churn was significantly higher among customers with lower satisfaction scores and those who submitted complaints.
- Found that active members with credit cards were less likely to churn, suggesting these as positive engagement indicators.
- Uncovered demographic trends, such as higher churn rates among specific age groups and genders, useful for targeted retention campaigns.
- Provided a clear roadmap of data-driven recommendations, including loyalty programs and personalized customer outreach strategies.