Vodafone

We are aiming to optimize campaign conversion rates and generate incremental revenue in product upselling through predictive analytics. We developed a hybrid Machine Learning model (Benchmark) that combines regression and classification to predict the optimal upgrade for each customer, creating a personalized upsell strategy.

To further enhance results, we implemented an elasticity-driven approach (Enhanced), introducing five new uplift points for customers with high propensity. The Benchmark model outperformed the business rules, achieving a 40.3% increase in ARPU and a 50% rise in conversion rate. The Enhanced model delivered even stronger results, adding an additional 8.4% boost to ARPU.