An empirical analysis of learning and channel migration

One of the major differences between online and offline purchases is that a shopper's experience with products is limited in an online setting. This translates into greater perceived risk regarding product quality for some categories such as produce, meat etc. where quality tends to vary and an online store agent selects these items on behalf of consumers. In this study, our goal is to understand the characteristics of online consumption behavior for risky products and what it means for channel migration. Understanding the implications of channel migration is becoming more important every day as brick-and-mortar stores use the Internet as a major distribution channel.; This study focuses on produce migration to an online grocery store, where managers use the Internet as a distribution channel alternative to the already existing brick-and-mortar stores. The results and the framework, however, can be generalized to other product categories with similar characteristics. With that in mind, in particular, we investigate if consumers (i) have lower quality expectations about online produce (ii) have higher perceived quality risk (iii) eventually learn about the true online produce quality, eliminate the perceived risk, and change the likelihood of purchasing online produce, (iv) change their total produce consumption. We address the above issues by using a framework that accounts for consumer experience and derives learning from its behavioral antecedents such as expected quality, perceived risk and risk aversion. The learning model explicitly accounts for information asymmetries between the two channels.; In our study, we find that offline produce is considered to be higher in quality and, all else being equal, consumers are more likely to shop online. We also find that inherent quality variation has a bigger impact on produce purchase decisions than the risk introduced by grocery store agents selecting the produce. The results from our counterfactual policy simulation suggest that getting consumers to substitute to the online store (e.g. free delivery incentives) can hurt overall category sales because of perceived risk associated with the product in an online channel. This risk causes consumers to avoid buying from that category, which causes the category demand (taken across both offline and online channels) to fall. Therefore, it is advisable for retailers to alleviate the perceived risk before sending customers to this alternative distribution channel, the Internet. Alleviating this risk prior to attempting to migrate consumers seems to have the biggest impact on product sales. Also, we find that giving away free samples could play a central role in eliminating perceived risk because most of the risk is resolved after the first experience with the product. Resolving a primary deterrent such as quality risk addresses the fundamental problem of why consumers are reluctant to switch to an online channel, thereby yielding an increase in online traffic.