Efficiency analysis and churn prediction

Telecom study that identifies cancellation drivers through data cleansing, cohort analysis, and visual storytelling to recommend retention and efficiency actions.

code View on GitHub arrow_back Back to projects

Tech stack

Pandas Cohorts Tableau

flag

Objective

Reduce churn in telecom by identifying critical variables and prioritizing retention and operational efficiency initiatives.

account_tree

Methodology

Data cleaning and preparation: standardizing contracts, payments, and services; deriving contract duration.
EDA and cohorts: comparing active vs. churned customers by contract type and payment method.
Metrics: churn rate, average monthly charges, tenure, and service segmentation.
Visualization and insights: comparative charts and correlations for cancellation drivers.

bar_chart

Key visualizations

• Churn by contract type (monthly vs. annual) and payment method.
• Retention curves and monthly churn rates.
• Comparison of monthly charges between churned and active customers.

emoji_events

Results

• Identified monthly contracts as the main driver of churn.
• Recommended shifting clients toward annual plans and proactive onboarding.
• Cohorts ready to track retention after implemented actions.