Author Abstract
We estimate a sales response model to evaluate the short- and long-term value of pharmaceutical sales representatives' detailing visits to physicians of different types. By understanding the dynamic effect of sales calls across heterogeneous doctors, we provide guidance on the design of optimal call patterns for route sales. Our analyses reveal that the long-term persistence effect of detailing is more pronounced for specialist physicians; the contemporaneous marginal effect is higher for generalists. Free samples have little effect on any type of physician. We also introduce a key methodological innovation to the marketing and economics literature. We show that moment conditions—typically used in traditional dynamic panel data methods—are vulnerable to serial correlation in the error structure. However, traditional tests to detect serial correlation have weak power and can be misleading, resulting in misuse of moment conditions and incorrect inference. We present an appropriate set of moment conditions to properly address serially correlated errors in analyzing dynamic panel data.
Paper Information
- Full Working Paper Text
- Working Paper Publication Date: June 2017
- HBS Working Paper Number: HBS Working Paper #17-108
- Faculty Unit(s): Marketing