Data Science
Final Project Part 7: Business Recommendations
Senior Data Analyst
May 27, 2026
5 min read
Insights
1. Customers with >90 days since last order are 5x more likely to churn.
2. Low purchase frequency is a major predictor—consider a loyalty program.
3. High support ticket count correlates with churn—fix product issues.
Recommendations
*Day 148: Deployment to Production.*