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.     

Prospects and problems in modeling group decisions

This paper summarizes some of the major issues related to group decision modeling. We briefly review the existing work on group choice models in marketing and consumer research. We draw some generalizations about which models work well when and use those generalizations to provide guidelines for future research.     

Behavioral Reasons for New Product Failure: Does Overconfidence Induce Overforecasts?

Efforts to organize and integrate research findings on new product performance determinants have lagged since the last significant overview paper appeared over a decade ago. Importantly, this literature has not considered entire categories of factors that are known to affect managerial decisions and behavior, namely those that pertain to decision‐makers' cognitive limitations and incentive structures. This research empirically investigates one specific cognitive distortion heretofore neglected in studies of new product commercialization—overconfidence, commonly defined in the literature as excessive belief in one's own abilities to generate superior performance. To lay the groundwork for subsequent exploration, the paper first introduces a behavioral model that both organizes well‐understood new product performance determinants and illuminates others heretofore not studied, namely incentive alignment and cognitive limitations and biases. The model summarizes extant research and allows development of research hypotheses related to overconfidence. The hypotheses and empirical investigation motivated by the model address two questions about the impact of overconfidence on new product commercialization activities. First, the study explores whether overconfidence is associated with overforecasting new product demand. Second, it evaluates two complementary mechanisms that may account for overconfidence‐induced overforecasts. The empirical findings are based on data generated in the course of management simulation workshops conducted among graduate students at three leading business schools in India. Three hundred thirty participants played individually four rounds of a computer‐based simulation game that involved decisions pertaining to new product development (including product formulation) and commercialization strategies. The decisions were captured and analyzed using statistical techniques. The results reveal that decision‐makers' overconfidence is associated with a higher likelihood of overforecasting new product sales. The observed effect is fully mediated by flawed tactical decisions that dampen demand, namely elevated product pricing. Sensitivity analyses show that these results are robust to a number of alternative explanations. However, the study finds no evidence implicating overconfident individuals as poor “innovators”—overconfident and nonoverconfident decision‐makers experienced comparable market demand for their new products. The paper concludes with a discussion of the results and provides specific recommendations for practice.     

Preference aggregation and repeat buying in households

Marketing researchers have long used brand switching analyses and Markov transition matrices to gain insights into managerial problems. Almost without exception, this work makes (inappropriate) inferences about individual consumers by analyzing household-level data. This paper presents a procedure based on the distribution of run lengths in household level panel data that allows more insights into the choice behavior of the individuals in the household. We test these procedures in a large simulation study by attempting to recover the underlying (known) structure of the process generating a string of panel data. Finally, we use the procedure to classify the purchase behavior, with respect to powdered soft drinks, of a set of households in a panel. Our results show that marketing scientists have the potential to learn and test more hypotheses about the individuals in a household by examining the distribution of run lengths.We gratefully acknowledge Professor Bari Harlam of the University of Rhode Island for providing the panel data.     

Choice in Interactive Environments

In the early 21st century, firms are thinking seriously and practically about an interactive marketing paradigm—one that integrates mass scale with individual responsiveness. The focus of this paper is on how this interactive environment is changing the customer decision-making process. With the increased amount of information available, the existence of sophisticated decision aids such as intelligent agents, and more latitude in how to interact beyond the basic desktop and laptop computers (e.g., personal digital assistants, cellular phones, tablet computers), customers have more choices than ever about how, when, and how much to interact with companies and each other. In this paper, we attempt to cover a few of the major areas of research on how customers make decisions in these environments.     

Towards a new way to measure power: Applying conjoint analysis to group decisions

This paper presents some preliminary development towards a methodology for measuring power in a group purchase decision. We define power as the capacity of an individual to change the probability of an alternative being chosen by expressing an opinion about that alternative. We present a linear model that relates the group members' preferences to the probability that a given alternative will be chosen. The model parameters reflect specific measures of power and can be estimated by conjoint analysis. We present the results of a pilot field study, conducted on a sample of purchasing managers, which favorably assesses certain psychometric properties of the resulting measures for dyads.