The project followed an iterative approach covering business alignment, data preparation, forecasting, output design, and handover. Working closely with stakeholders, the team first clarified the churn definition and defined a reliable contract population for analysis. Several rounds of quality checks were then used to reduce the risk of misleading results before building an early-warning view of churn 12 months before lease end. The resulting outputs were prepared for practical use cases such as proactive retention actions, demand forecasting, and dealer learning.
