Python-based exploratory data analysis to understand customer behavior, data quality issues, usage patterns, and segmentation opportunities for a telecommunications company.
This project analyzes customer, plan, and usage data for ConnectaTel, a telecommunications company. The main objective was to prepare and validate the data, explore customer behavior, and create meaningful segments based on usage patterns and demographics.
The analysis focuses on identifying data quality issues, understanding how customers interact with communication services, and translating technical findings into business insights that can support plan optimization, customer targeting, and future churn analysis.
Telecommunications companies rely heavily on customer usage data to design better pricing plans, identify high-value customers, and reduce churn risk. However, before using this data for decision-making, it is essential to validate its quality, clean inconsistent values, and understand how different customer groups behave.
In this project, customer activity across calls and text messages was analyzed to identify usage patterns and segment customers into actionable groups. These segments can help the business better understand which customers may need premium offerings, flexible pricing, or targeted retention strategies.
What data quality issues exist in the customer and usage datasets?
How are customers distributed across plans, cities, and age groups?
What are the main usage patterns for calls and text messages?
Are high-usage customers valid business cases or data outliers?
How can customers be segmented to support commercial decision-making?
The analysis identified several important data quality issues, including missing customer cities, inactive customer indicators, invalid age values, and inconsistent city values. These issues were reviewed and treated carefully to avoid introducing bias into the analysis.
Customer usage patterns showed right-skewed distributions, meaning that most customers have moderate activity while a smaller group shows significantly higher usage. These high-usage customers should not be removed as outliers because their behavior appears commercially relevant and may represent valuable customer segments.
The segmentation analysis classified customers by usage level and age group, creating a foundation for more advanced customer analytics. These segments can support future initiatives such as plan optimization, churn analysis, targeted campaigns, and customer lifetime value analysis.
Data cleaning and validation with Python
Exploratory data analysis
Missing value treatment
Outlier analysis
Customer segmentation
Data visualization
Business-oriented interpretation
Communication of technical findings for stakeholders