Customer evolution in sales channel migration

In recent years, research on multichannel shopping behavior has increasingly expanded (see Neslin et al. 2006 and Rangaswamy and Van Bruggen 2005 for a summary of the extant literature). Several managerial, (DoubleClick, 2004; Wall Street Journal, 2004; Myers, Van Metre & Pickersgill, 2004) as well as academic (Kumar & Venkatesan, 2005; Rangaswamy & Van Bruggen, 2005; Thomas & Sullivan, 2005) studies agree in considering multichannel customers as a great opportunity for firms. Customers have rapidly expanded their channel experiences and preferences beyond traditional channels (such as stores) and they expect the company with which they do business to have a presence on all these channels (Blattberg, Kim & Neslin, 2008). Therefore, there is increasing interest in understanding customers channel choices dynamics in a multichannel environment. A deeper understanding of the channel “migration” process (Blattberg, Kim & Neslin, 2008) can help managers to design marketing programs that evolve with their customers over time and evaluate the profitability of different customers in terms of channel choice behavior. Although recently the dynamic of customers' channel choices have been explored (Ansari, Mela & Neslin, 2008; Thomas & Sullivan, 2005; Knox, 2005; Venketesan & Kumar, 2007), with the exception of Knox (2005) a limited effort has been made to investigate and formally model the learning process per se, i.e., how customers’ decision process changes over time as they learn their preferences and become familiar with the firm’s marketing activities. A deeper understanding of this evolution and the process by which customers become loyal to certain channels is an important issue. We distinguish between two stages in the evolution of customers channel choices over time. An initial or “trial” stage when the customer is acquiring experience with the company channel offer and a second or “steady state” phase representing the decision process the customer evolves to in the long term. We aim to understand if the duration of these phases is homogeneous among customers. Furthermore, we aim to decompose the sensitiveness of the customers to marketing communications, distinguishing between the initial and the second stage. We develop and estimate a model of customer channel migration in order to study this phenomenon. We believe our contribution is twofold. First we propose a modeling approach which takes into account the existence of these stages in the channel choice migration process. Second, we show that the existence of this phenomenon has an impact on the efficacy of direct marketing communication and crate heterogeneity among customers in terms of the duration of this process.