📉 Analyzing Customer Churn

Problem Statement

Customer churn poses a significant challenge for subscription-based businesses, as it directly affects both revenue and long-term growth. Databel, a fictitious telecom company, has recently experienced an increase in customer churn.
The objective of this project is to use Excel to analyze customer churn data, identify key drivers, and uncover actionable insights that could help reduce churn in the future.

📒 Report of Analyzing Customer Churn

Objectives

1. Identify the main reasons for customer churn and determine which features are most strongly associated with it.
2. Examine churn rates across different age categories to detect patterns or at-risk groups.
3. Analyze Unlimited Data Plan customers by segmenting their historical data usage into:
   • Less than 5 GB
   • Between 5 GB and 10 GB
   • More than 10 GB
4. Explore churn patterns across U.S. states and assess the impact of:
   • International Plan ownership
   • Customer tenure
   • Payment type (Month-to-Month, One Year, Two Years)

Data Sources

The dataset was obtained from the DataCamp Case Study: Analyzing Customer Churn.

• Customer Table – contains information for each individual customer.
• Aggregate Table – contains counts of customers by plan combinations and other service options.

Data Loading

Two tables, Customer and Aggregate, were imported into Excel using Power Query.

Import of Aggregate table into Power Query.

Import of Customer table into Power Query.

Customer table loaded into Excel.

Aggregate table loaded into Excel.

Data Cleaning

• Checked for duplicate records in the Customer ID column using Data → Remove Duplicates.

Removing duplicates from Customer ID column.

No duplicates were found.

Excel confirmation that no duplicates exist.

Descriptive Statistics

• Total number of customers: 6687
• Churn rate: 26.86%

These metrics served as the baseline for further analysis.

Data Exploration & Insights

1. Main Reasons for Churn

• Nearly 50% of customer departures are due to competitive offers — mainly better device options and more favorable plan conditions provided by competitors.
• Competition-related churn remains the most significant driver of customer loss.

Breakdown of customer churn reasons.

2. Churn by Age Group

• The highest churn rate occurs among customers aged 30–65 years.
• Within this group, churn is relatively evenly distributed, indicating no strong age sub-clusters.

Churn analysis segmented by customer age group.

3. Unlimited Data Plan – Data Usage Segmentation

• Customers with average data usage (5–10 GB) under the Unlimited Plan churn more frequently than low (<5 GB) or high (>10 GB) usage customers.
• This suggests possible dissatisfaction with perceived value in the mid-usage segment.

Churn rates for Unlimited Plan customers by data usage level.

4. Churn by Geography, International Plan, and Payment Type

• International Plan subscribers churn significantly more often than others in certain states — particularly CA, IN, NH, KY, and LA — which show the largest churn rate differences.
• Payment Type Impact:
  • Month-to-Month plans have the highest churn rates, especially within the first four years of service.
  • One- and Two-Year contracts show no significant churn spikes, indicating higher retention for long-term commitments.

Churn analysis based on International Plan ownership and state distribution.

Comparison of churn rates across different subscription types.

Conclusion

This analysis highlights competition as the primary driver of churn, with notable secondary factors including customer age group, specific data usage patterns, and certain geographic/plan combinations.
Addressing competitive disadvantages, targeting mid-usage Unlimited Plan customers, and incentivizing longer-term contracts may help Databel reduce future churn.

---

Tools Used: Excel, Power Query, Pivot Tables, Conditional Formatting.