Firms are increasingly interested in developing targeted interventions for customers with the best response. This requires identifying differences in customer sensitivity, typically through the conditional average treatment effect (CATE) estimation. In theory, to optimize long-term business performance, firms should design targeting policies based on CATE models constructed using long-term outcomes. However, we show theoretically and empirically that this method can fail to improve long-term results, particularly when the desired outcome is the cumulative result of recurring customer actions, like repeated purchases, due to the accumulation of unexplained individual differences over time. To address this challenge, we propose using a surrogate index that leverages short-term outcomes for long-term CATE estimation and policy learning. Moreover, for the creation of this index, we propose the separate imputation strategy, designed to reduce the additional variance caused by the inseparable nature of customer churn and purchase intensity, prevalent in marketing contexts. This involves constructing two distinct surrogate models, one for the observed last purchase time and the other for the observed purchase intensity. Our simulation and real-world application show that (i) using short-term signals instead of the actual long-term outcome significantly improves long-run targeting performance, and (ii) the separate imputation technique outperforms existing imputation approaches
