Scribd
Upload
14 views

Customer_Churn_Analysis_Report_with_Visuals

Uploaded by

tryitdoit35
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
14 views

Customer_Churn_Analysis_Report_with_Visuals

Uploaded by

tryitdoit35
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Customer Churn Analysis Report

This report outlines the process of selecting, building, and evaluating a machine learning
model to predict customer churn. The goal is to identify customers at risk of leaving and
provide actionable insights to help mitigate this risk.

Data Identification and Collection


The data set includes the following key components:
- Customer Demographics: Age, Gender, Marital Status, Income Level
- Transaction History: Transaction Dates, Amount Spent, Product Categories
- Customer Service Interactions: Type and Resolution Status of Interactions
- Online Activity: Last Login Date, Login Frequency, Service Usage
- Churn Status: Indicating whether each customer has churned (1) or not (0)

Exploratory Data Analysis


Exploratory data analysis was conducted to uncover patterns and insights that could inform
predictive modelling. Key findings include demographic trends, spending patterns, and the
impact of customer service interactions on churn.

Figure 1: Age Distribution of Customers


Figure 2: Churn Rate by Gender
Figure 3: Average Amount Spent by Churn Status
Figure 4: Customer Service Interactions by Resolution Status and Churn
Model Selection and Building
After considering several algorithms, a random forest classifier was selected for its balance
of accuracy and interpretability. The model was trained using stratified 5-fold cross-
validation, ensuring robustness and generalisation.

Model Evaluation
The model's performance was evaluated using precision, recall, F1 score, and ROC-AUC
metrics. Feature importance analysis identified key predictors of churn, such as transaction
amount and login frequency.

Recommendations and Business Implications


The model's outputs can be used to develop targeted retention strategies for at-risk
customers. Further improvements could include integrating additional data sources and
exploring more complex models like XGBoost or Neural Networks.

You might also like