James M. Lattin

Choice and the Internet: From Clickstream to Research Stream

The authors discuss research progress and future opportunities for modeling consumer choice on the Internet using clickstream data (the electronic records of Internet usage recorded by company web servers and syndicated data services). The authors compare the nature of Internet choice (as captured by clickstream data) with supermarket choice (as captured by UPC scanner panel data), highlighting the differences relevant to choice modelers. Though the application of choice models to clickstream data is relatively new, the authors review existing early work and provide a two-by-two categorization of the applications studied to date (delineating search versus purchase on the one hand and within-site versus across-site choices on the other). The paper offers directions for further research in these areas and discusses additional opportunities afforded by clickstream information, including personalization, data mining, automation, and customer valuation. Notwithstanding the numerous challenges associated with clickstream data research, the authors conclude that the detailed nature of the information tracked about Internet usage and e-commerce transactions presents an enormous opportunity for empirical modelers to enhance the understanding and prediction of choice behavior.

Individual Differences in Response to Consumer Promotions

Abstract This research hypothesizes important individual differences in response to promotions and tests for them using a cross-sectional multinomial logit choice model. Our hypotheses suggest interactions between individual brand preference and the effects of current promotion and past promotional purchase. To test for these interactions, we introduce a new method of measuring brand preference from past purchase data. The new method seeks to incorporate competitive purchase conditions as modifiers to observed brand purchase behavior in estimating consumer brand preference. We account for heterogeneity in the cross-sectional model by using two measures of loyalty: one to capture differences across individuals and one to capture the time-varying changes within an individual. Our empirical results, based on scanner panel data for instant caffeinated coffee, support our hypotheses and our model specification. We conclude that accounting for consumer heterogeneity both in response to promotion and in brand and size loyalty improves both fit and predictive ability.

A minimum-cost network-flow solution to the case V thurstone scaling problem

This paper presents an approach for determining unidimensional scale estimates that are relatively insensitive to limited inconsistencies in paired comparisons data. The solution procedure, shown to be a minimum-cost network-flow problem, is presented in conjunction with a sensitivity diagnostic that assesses the influence of a single pairwise comparison on traditional Thurstone (ordinary least squares) scale estimates. When the diagnostic indicates some source of distortion in the data, the network technique appears to be more successful than Thurstone scaling in preserving the interval scale properties of the estimates.

Coalition Loyalty Program Not Working? Perhaps You’re Doing It Wrong

In this paper we explore the determinants of profitability for coalition loyalty programs. We consider a setting in which each of two firms competing in one market may form a coalition loyalty program with one of two firms in a different market. Firms in the same program jointly set the reward to consumers who buy from both coalition partners, but they set their own prices independently. We find that these programs are profitable for all firms, even when no value is created by the mere existence of rewards (i.e., when firms and consumers value

Development and Testing of a Model of Consideration Set Composition

1 worth of rewards equally). The intuition is that joint loyalty programs allow each participating firm to leverage its partner’s market power and charge higher prices. This result, however, depends crucially on several design elements of the program. First, rewards must be structured so that consumers earn more when they shop broadly across firms in the coalition than when they shop at only a single firm. Second, the reward program manager must be able to take into account the prices of individual firms when setting the value of rewards. Third, firms joining a coalition must be able to negotiate the share of program costs they will carry; firms must be charged according to their value added to the coalition (e.g., firms with greater market power will bear a lower share of program costs) and not taxed as a proportion of their revenues. Our theoretical findings provide insight into the forces underlying coalition loyalty programs in competitive settings and are suggestive of the impact of practical design decisions on program profitability.