Predictive Analytics Model Blueprint

Design a predictive analytics model for [USE CASE: churn prediction / demand forecasting / lead scoring / fraud detection / customer segmentation].

Data available: [DESCRIBE DATASETS]
Tool: [PYTHON / R / EXCEL / BI TOOL]

Provide:
1. **Problem Framing**: What we're predicting and what a good prediction enables
2. **Feature Engineering**: 15 candidate features with rationale
3. **Data Preparation**: Cleaning, transformation, and splitting strategy
4. **Model Selection**: 3 candidate algorithms with pros/cons for this problem
5. **Evaluation Metrics**: Which metrics matter (accuracy, precision, recall, AUC, RMSE)
6. **Implementation Code**: Python code for the recommended model
7. **Validation Approach**: Cross-validation and holdout testing
8. **Feature Importance**: How to identify which features drive predictions
9. **Deployment**: How to operationalize the model (API, scheduled batch, real-time)
10. **Monitoring**: How to detect model drift and when to retrain

Design a predictive analytics model for [USE CASE: churn prediction / demand forecasting / lead scoring / fraud detection / customer segmentation].

Data available: [DESCRIBE DATASETS]
Tool: [PYTHON / R / EXCEL / BI TOOL]

Provide:
1. **Problem Framing**: What we're predicting and what a good prediction enables
2. **Feature Engineering**: 15 candidate features with rationale
3. **Data Preparation**: Cleaning, transformation, and splitting strategy
4. **Model Selection**: 3 candidate algorithms with pros/cons for this problem
5. **Evaluation Metrics**: Which metrics matter (accuracy, precision, recall, AUC, RMSE)
6. **Implementation Code**: Python code for the recommended model
7. **Validation Approach**: Cross-validation and holdout testing
8. **Feature Importance**: How to identify which features drive predictions
9. **Deployment**: How to operationalize the model (API, scheduled batch, real-time)
10. **Monitoring**: How to detect model drift and when to retrain