An assessment of when, where and under what conditions in-store sampling is most effective

• Main Authors: CHANDUKALA, Sandeep R., DOTSON, Jeffrey P., LIU, Qing
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2017
• Subjects: Marketing mix modeling; In-store sampling; Retail strategy; Marketing
• Online Access: https://ink.library.smu.edu.sg/lkcsb_research/5270; https://ink.library.smu.edu.sg/context/lkcsb_research/article/6269/viewcontent/1_s20_S0022435917300544_main.pdf
• Institution: Singapore Management University

In-store product sampling is a commonly used promotional technique designed to give prospective consumers an opportunity to experience a product prior to purchase. While prior research has documented a positive relationship between short-term sales and perceptual measures of the customer shopping experience, little is known about the long-term impact of sampling or factors that moderate its success. In this paper, we develop an empirical approach that allows us to study the short-term and long-term effects of in-store sampling on both own and competitive products. We apply our approach to six store-level scanner data sets across four different product categories and show that in-store sampling has both an immediate (short-term) and sustained (long-term) impact on sales. We also show that the impact of sampling on sales is moderated by the characteristics of the store conducting the event, and that repeated sampling for a single product leads to a multiplicative increase in its long-term sales performance. We find that, unlike many types of in-store promotion, sampling results in a category expansion effect as opposed to a pure substitution effect. We contrast the immediate and long-term sales patterns for in-store sampling to those of product displays and discuss managerially relevant differences. Finally, we demonstrate incremental profit implications and store selection scenarios for different incremental costs of conducting the in-store events using constrained optimizations.