Churn Prediction Model Explainer

Explain churn predictions from {{model_name}}:

Model overview:
- Type: {{model_type}} (logistic, tree, ensemble?)
- AUC: {{auc_score}}, Precision: {{precision}}, Recall: {{recall}}
- Training data: {{training_period}}
- Prediction window: {{prediction_window}}

Top risk factors:
{{feature_importance}}

For business:
1. Feature interpretation (why do these predict churn?)
2. Risk segments (which customers are most at-risk?)
3. Intervention design (what action reduces churn?)
4. ROI estimate (cost of retention vs CLV of saved customer)
5. Monitoring (how do we track model drift?)
6. Ethics (are we excluding protected classes from retention offers?)

Provide a customer communication template for the retention offer.