How sahbimlesh007/customer-churn---retention-analytics-system works
This is an end-to-end telecom churn analytics and prediction project. Its documented workflow starts with customer data, develops SQL and Python analyses and churn models, and presents findings through Power BI and a Streamlit prediction interface. The graph preserves documented segmentation, revenue-risk analysis, and business insights. The notebook and SQL internals were not sampled, so their detailed calls and data transformations are not inferred; README statements support the broad workflow.
Main entry points: app.py Streamlit App · create_BIviews.sql BI Views · customer_churn_eda.ipynb Exploratory Analysis · logistic_churn_model.pkl Churn Model
Components 14
Data Foundation
- Telco Dataset
dataset/Telco-Customer-Churn.csv - MySQL Database
sql/create_db.sql - Customer Tables
sql/create_table.sql - Cleaned Data
sql/create_cleaned_table.sql - BI Views
sql/create_BIviews.sql
Churn Analytics
- Exploratory Analysis
notebooks/customer_churn_eda.ipynb - Customer Segments
notebooks/customer_churn_eda.ipynb - Churn Model
notebooks/logistic_churn_model.pkl - Model Artifacts
notebooks/model_features.pkl
Decision Products
- Power BI Dashboard
powerbi/IBM Telco Customer Churn Analytics.pbix - Streamlit App
streamlit/app.py - Risk Prediction
streamlit/app.py - Business Insights
streamlit/app.py
Other components
- Analyst
Connections 16
- Telco Dataset → MySQL Database: loads into
README.md - MySQL Database → Customer Tables: defines tables
README.md - Customer Tables → Cleaned Data: prepares data
README.md - Cleaned Data → BI Views: supports analysis
README.md - BI Views → Exploratory Analysis: feeds analysis
README.md - Exploratory Analysis → Customer Segments: identifies segments
README.md - Exploratory Analysis → Churn Model: informs modeling
README.md - Churn Model → Model Artifacts: saved as
README.md - BI Views → Power BI Dashboard: informs dashboard
README.md - Analyst → Streamlit App: enters customer data
streamlit/app.py - Streamlit App → Model Artifacts: loads model inputs
streamlit/app.py - Streamlit App → Churn Model: predicts churn
streamlit/app.py - Streamlit App → Risk Prediction: shows risk score
streamlit/app.py - Streamlit App → Business Insights: shows insights
streamlit/app.py - Power BI Dashboard → Streamlit App: precedes deployment
README.md - Customer Segments → Power BI Dashboard: supports segmentation
README.md