Uplift modeling is an application of causal machine learning and offers an assortment of analytical tools to identify likely responders to a particular treatment such as a medical prescription, a political maneuver, or an advertising stimulus. Although several targeted campaigns co-occur (e.g., through different marketing channels), recent literature has primarily examined the effectiveness of a single treatment. To address the practically more pertinent question of which treatment among several options to choose, we develop a prototype that identifies the most effective treatment for each unit of observation and further generalizes to both binary and continuous outcomes to support classification and regression problems. Using real-world data from e-mail merchandising and e-couponing campaigns, we verify our prototype’s financial advantage compared to previous efforts toward the single treatment case.