Knowledge that Transforms

We study the pass-through of wholesale price changes onto regular retail prices using an unusually detailed data set obtained from a major retailer. We model pass-through as a two-stage decision process that reflects both whether as well as how much to change the regular retail price. We show that pass-through is strongly asymmetric with respect to wholesale price increases versus decreases. Wholesale price increases are passed through to regular retail prices 70% of the time while wholesale price decreases are passed through only 9% of the time. Pass-through is also asymmetric with respect to the magnitude of the wholesale price change, with the magnitude affecting the response to wholesale price increases but not decreases. Finally, we show that covariates such as private label versus national brand, 99-cent price endings, and the time since the last wholesale price change have a much stronger impact on the first stage of the decision process (i.e., whether to change the regular retail price) than on the second stage (i.e., how much to change the regular retail price). Data, as supplemental material, are available at http://dx.doi.org/10.1287/mksc.2015.0947 .

The proliferation of free trials for high-tech services calls for a careful study of their effectiveness, and the drivers thereof. On one hand, free trials can generate new paying subscribers by allowing consumers to become acquainted with the service free of charge. On the other hand, a disappointing trial experience might alienate potential customers, when they decide not to adopt the system and are lost for good. This dilemma is particularly worrisome in early periods, when service quality has not been “tried and tested” in the field, and breakdowns occur. We accommodate these phenomena in a model of consumers’ free-trial and regular adoption decisions. Among other effects, it incorporates usage- and word-of-mouth-based learning about quality in a setting where quality itself is evolving. Consumers are forward-looking in that they account for changes in quality and anticipate uncertainty reduction due to trial usage. We estimate our model and run simulations on the basis of a rich and unique data set that incorporates customers’ trial subscription, adoption, and usage behavior for an interactive digital television service. The results underscore that free trials constitute a double-edged sword, and that timing and consumers’ usage intensity during the trial are key to the effectiveness of these promotions. Implications for managers are also discussed. Data, as supplemental material, are available at http://dx.doi.org/10.1287/mksc.2015.0973 .

In large markets comprising hundreds of products, comprehensive visualization of competitive market structures can be cumbersome and complex. Yet, as we show empirically, reduction of the analysis to smaller representative product sets can obscure important information. Herein we use big search data from a product- and price-comparison site to derive consideration sets of consumers that reflect competition between products. We integrate these data into a new modeling and two-dimensional mapping approach that enables the user to visualize asymmetric competition in large markets (>1,000 products) and to identify distinct submarkets. An empirical application to the LED-TV market, comprising 1,124 products and 56 brands, leads to valid and useful insights and shows that our method outperforms traditional models such as multidimensional scaling. Likewise, we demonstrate that big search data from product- and price-comparison sites provide higher external validity than search data from Google and Amazon. Data, as supplemental material, are available at http://dx.doi.org/10.1287/mksc.2015.0950 .

An accurate prediction of what a customer will purchase next is of paramount importance to successful online retailing. In practice, customer purchase history data is readily available to make such predictions, sometimes complemented with customer characteristics. Given the large product assortments maintained by online retailers, scalability of the prediction method is just as important as its accuracy. We study two classes of models that use such data to predict what a customer will buy next, i.e., a novel approach that uses latent Dirichlet allocation (LDA), and mixtures of Dirichlet-Multinomials (MDM). A key benefit of a model-based approach is the potential to accommodate observed customer heterogeneity through the inclusion of predictor variables. We show that LDA can be extended in this direction while retaining its scalability. We apply the models to purchase data from an online retailer and contrast their predictive performance with that of a collaborative filter and a discrete choice model. Both LDA and MDM outperform the other methods. Moreover, LDA attains performance similar to that of MDM while being far more scalable, rendering it a promising approach to purchase prediction in large product assortments. Data, as supplemental material, are available at http://dx.doi.org/10.1287/mksc.2016.0985 .

We develop a game-theoretic model to examine the entry of copycats and its implications by incorporating two salient features; these features are two product attributes, i.e., physical resemblance and product quality, and two consumer utilities, i.e., consumption utility and status utility. Our equilibrium analysis suggests that copycats with a high physical resemblance but low product quality are more likely to successfully enter the market by defying the deterrence of the incumbent. Furthermore, we show that higher quality can prevent the copycat from successfully entering the market. Finally, we show that the entry of copycats does not always improve consumer surplus and social welfare. In particular, when the quality of the copycat is sufficiently low, the loss in status utility from consumers of the incumbent product overshadows the small gain in consumption utility from buyers of the copycat, leading to an overall decrease in consumer surplus and social welfare.

In this paper, we propose an automated and scalable garment recommender system using real-time in-store videos that can improve the experiences of garment shoppers and increase product sales. The video-based automated recommender (VAR) system is based on observations that garment shoppers tend to try on garments and evaluate themselves in front of store mirrors. Combining state-of-the-art computer vision techniques with marketing models of consumer preferences, the system automatically identifies shoppers’ preferences based on their reactions and uses that information to make meaningful personalized recommendations. First, the system uses a camera to capture a shopper’s behavior in front of the mirror to make inferences about her preferences based on her facial expressions and the part of the garment she is examining at each time point. Second, the system identifies shoppers with preferences similar to the focal customer from a database of shoppers whose preferences, purchasing, and/or consideration decisions are known. Finally, recommendations are made to the focal customer based on the preferences, purchasing, and/or consideration decisions of these like-minded shoppers. Each of the three steps can be implemented with several variations, and a retailing chain can choose the specific configuration that best serves its purpose. In this paper, we present an empirical test that compares one specific type of VAR system implementation against two alternative, nonautomated personal recommender systems: self-explicated conjoint (SEC) and self-evaluation after try-on (SET). The results show that VAR consistently outperforms SEC and SET. A second empirical study demonstrates the feasibility of VAR in real-time applications. Participants in the second study enjoyed the VAR experience, and almost all of them tried on the recommended garments. VAR should prove to be a valuable tool for both garment retailers and shoppers. Data, as supplemental material, are available at http://dx.doi.org/10.1287/mksc.2016.0984 .

I introduce the special section highlighting the background of the competition and the work of the finalists in the 2013–2014 Gary L. Lilien ISMS-MSI Practice Prize Competition, showcasing the best applications of rigor and relevance by marketing scientists in working with practical problems. The winning paper is by a team who developed an innovative pricing tool to enable an electric utility operating in Germany to acquire customers on online price comparison sites while optimizing various metrics of interest. The other two finalists comprise a team that developed an integrated marketing model to address multiple business objectives of the Georgia Aquarium and a team that developed a marketing science model of evaluation and purchase intentions incorporating customer emotions to test advertisement effectiveness for Kmart Australia.