This project is about to identify and solving customer churn rate that happening in O-NET Company. The dataset originally came from kaggle. In this project I were able to do :

• Cleaned raw data using various method and make sure the data is feasible to analyze.
• Conducted exploratory data analysis (EDA) to find out what causes customer churn.
• Analyzed Customer Churn by their factor such as: Subscriber Package, Age Subscription and Service Failure.

Insights

1. The Internet-only package has the highest customer churn rate based on Customer Package Category.
2. Customers in the subscription age of 0-1 year constitute the group with the highest number of customer churns.
3. Surprisingly, Service Failure does not influence customer churn, as the majority of customer churns come from non-problematic service failures.

Project Files

• Project Notebook
• Tableau Dashboard

Project Background

O-Net Group is a leading internet service provider company. The CEO has assigned the Head of Data Analysis to conduct an in depth analysis to identify the customer churn rate and formulate strategies to mitigate this high churn rate.

Data Scope & Objectives

1. Data Scope

This dataset is a list of over 62.000 user customer that using O-Net internet service. The dataset includes the customer movie subs & TV subs, Remaining contract, Customer age subscription, Customer churn, Service Failure, Customer Bill Average.

2. Objectives

1. To investigate the factors influencing customer churn
2. Establish correlations
3. Predict customer churn using logistic regression modeling
4. Provide actionable recommendations that can be utilized to shape our future strategies

Data Cleaning

1. We removing irrelevant variables that we don't use in analysis.
2. We removing duplicates data.
3. We handle missing value by impute value (0) in it.
4. We identifying outliers by using interquartile methods and then removing the outliers.

EDA

Churn Rate & Subscriber

1. There are more than half, which is 36,523 customers (58.4%), who have churned.

1. The highest percentage of churn is among customers with the "internet-only" package, at 92.2%.
2. On the other hand, the "Full Package" customers represent the package with the lowest churn percentage at 36.6%

Customer Subscription Age

Interpretation: Customers in the subscription age group of 0-1 year representing approximately 54.8% of the total overall customer churn..

Service Failure

1. The number of Customer Churn instances is significantly higher among customers without issues, totaling 84.5%,