Retail Distribution Strategy with Outlet Stores
研究一家已有主店的零售商是否开设折扣店以及如何选址，发现顾客分散度、产品匹配概率和旅行敏感度等因素共同影响最优策略。
Traditionally, outlet stores were situated away from main stores in order to provide older, less desirable products at discounted prices. More recently, some firms have featured an outlet‐within‐a‐store concept and offer consumers the experience of outlet shopping at the same location. With interest in examining the effect of location differentiation and uncertain outlet store's product fit on a firm's distribution strategy, we model a firm with an existing main store and study its motivation on whether to open an outlet store, and where to locate it. The analysis leads to a number of key observations. First, if the firm decides to open an outlet store, it may be co‐located with the main store or at a remote location, depending on the combined role of customer dispersion, outlet store's product fit, customer travel sensitivity, and fixed costs. Second, there exists a substitution effect between location and quality (and price) differentiation. We show that customer dispersion always encourages the firm to offer outlet stores so as to achieve better market coverage. While a higher probability of product fit at the outlet store makes its more attractive (e.g., it makes customers at one location more willing to travel to the outlet store at another location), interestingly, it has a non‐monotone effect on its location. Also, in contrast to intuition, higher customer travel sensitivity may even lead to locating the outlet store away from the city center depending on the level of customer dispersion. We also study three extensions. First, the results of our base model are shown to be robust if product quality at the outlet store can positively influence the outlet product's fit probability. Second, heterogeneous consumer travel sensitivity makes it easier for the firm to segment consumers and hence can increase profits even when travel sensitivity is high. Finally, we consider a continuum of possible outlet store locations, and demonstrate the robustness of the base model results.