simpli-cheekah/telco-churn-analysis

Exploratory data analysis and business insights on the Telco Customer Churn dataset — AnalystLab Africa Data Analytics Internship, Week 5 # Telco Customer Churn Analysis Business analytics case study completed as part of the AnalystLab Africa Data Analytics Internship (Batch B, Week 5). ## Project Overview This project analyzes the Telco Customer Churn dataset to answer a core business question: why are customers leaving, and which factors contribute most to churn? The goal was to move beyond simple description of the data and produce insights and recommendations a business manager could act on. ## Dataset - Source: Telco Customer Churn dataset (Kaggle) - Size: 7,043 customer records, 21 columns (7,032 after cleaning) - Target variable: Churn (Yes/No) ## Tools Used - Python (Pandas, Matplotlib, Seaborn) — data cleaning, exploratory analysis, visualization - Google Colab — notebook environment - Microsoft Word — case study report - PowerPoint — business presentation ## Process 1. Data cleaning — identified and resolved 11 hidden missing values in TotalCharges 2. Exploratory data analysis — dataset structure, summary statistics, churn baseline 3. Driver analysis — churn rate broken down by contract type, tenure, internet service, and payment method 4. Correlation analysis — quantified relationships between numeric variables and churn 5. Business insights and recommendations ## Key Findings | Driver | Finding | |---|---| | Contract Type | Month-to-month customers churn at 42.7%, vs. 2.8% for two-year contracts (~15x gap) | | Tenure | Churn falls from 47.7% in year one to 9.5% by year four-to-six | | Monthly Charges | Churners pay a noticeably higher median monthly charge | | Internet Service | Fiber optic customers churn at 41.9%, more than double the DSL rate | | Payment Method | Electronic check users churn at 45.3%, ~3x any automatic payment method | ## Recommendations 1. Launch a first-year retention program targeting the highest-risk window 2. Incentivize upgrades from month-to-month to longer contracts 3. Investigate fiber optic satisfaction (price, reliability, or support) 4. Encourag …