Wagner A. Kamakura

Satisfaction, Repurchase Intent, and Repurchase Behavior: Investigating the Moderating Effect of Customer Characteristics

Despite the claim that satisfaction ratings are linked to repurchase behavior, few attempts can be found that relate satisfaction ratings to actual repurchase behavior. This article fills this void by presenting a conceptual model for relating satisfaction ratings and repurchase behavior. The model is based on the premise that ratings observed in a typical customer satisfaction survey are error-prone measures of the customers true satisfaction, and they may vary systematically on the basis of consumer characteristics. The authors apply the model to a large-scale study of 100,040 automotive customers. Results show that consumers with different characteristics have different thresholds such that, at the same level of rated satisfaction, repurchase rates are systematically different among different customer groups. The authors also find that the nature and extent of response bias in satisfaction ratings varies by customer characteristics. In one group, the response bias is so high that rated satisfaction is completely uncorrelated to repurchase behavior (r = 0). Furthermore, the authors find that, though nonlinear, the functional form relating rated satisfaction to repurchase intent is different from the one relating it to repurchase behavior. Although the functional form exhibits decreasing returns in the case of repurchase intent, it exhibits monotonically increasing returns in the case of repurchase behavior.

Measuring Brand Value with Scanner Data

Using actual consumer choice data from a single-source scanner panel, we construct two measures of brand value which capture different aspects of brand equity. Brand Value measures perceived quality, the value assigned by consumers to the brand, after discounting for current price and recent advertising exposures. Brand Intangible Value isolates the component of brand value which cannot be directly attributed to the physical product, thus measuring the value created by such factors as brand name associations and perceptual distortions. We illustrate these measures in a study of the powder laundry detergent category and briefly relate the results to strategic variables (order of entry and cumulative advertising expenditures).

Defection Detection: Measuring and Understanding the Predictive Accuracy of Customer Churn Models

This article provides a descriptive analysis of how methodological factors contribute to the accuracy of customer churn predictive models. The study is based on a tournament in which both academics and practitioners downloaded data from a publicly available Web site, estimated a model, and made predictions on two validation databases. The results suggest several important findings. First, methods do matter. The differences observed in predictive accuracy across submissions could change the profitability of a churn management campaign by hundreds of thousands of dollars. Second, models have staying power. They suffer very little decrease in performance if they are used to predict churn for a database compiled three months after the calibration data. Third, researchers use a variety of modeling “approaches,” characterized by variables such as estimation technique, variable selection procedure, number of variables included, and time allocated to steps in the model-building process. The authors find important differences in performance among these approaches and discuss implications for both researchers and practitioners.

Cross-selling through database marketing: a mixed data factor analyzer for data augmentation and prediction

An important aspect of the new orientation on customer relationship marketing is the use of customer transaction databases for the cross-selling of new services and products. In this study, we propose a mixed data factor analyzer that combines information from a survey with data from the customer database on service usage and transaction volume, to make probabilistic predictions of ownership of services with the service provider and with competitors. This data-augmentation tool is more flexible in dealing with the type of data that are usually present in transaction databases. We test the proposed model using survey and transaction data from a large commercial bank. We assume four different types of distributions for the data: Bernoulli for binary service usage items, rank-order binomial for satisfaction rankings, Poisson for service usage frequency, and normal for transaction volumes. We estimate the model using simulated likelihood (SML). The graphical representation of the weights produced by the model provides managers with the opportunity to quickly identify cross-selling opportunities. We exemplify this and show the predictive validity of the model on a hold-out sample of customers, where survey data on service usage with competitors is lacking. We use Gini concentration coefficients to summarize power curves of prediction, which reveals that our model outperforms a competing latent trait model on the majority of service predictions.

Value Segmentation: A Model for the Measurement of Values and Value Systems

This article develops a model for the measurement of human values that, rather than obtain aggregate measurements, identifies distinct value systems within a population and classifies individuals according to them. These value systems are inferred from the stated priority rankings, obtained from each individual via the Rokeach value survey. Copyright 1991 by the University of Chicago.

Optimal Bundling and Pricing Under a Monopoly: Contrasting Complements and Substitutes from Independently Valued Products

We develop an analytical model of contingent valuations and address two questions of import to a monopolist: (i) should a given pair of complements or substitutes be sold separately (pure components), together (pure bundling), or both (mixed bundling), and at what prices? (ii) How do optimal bundling and pricing strategies for complements and substitutes differ from those for independently valued products? We find that the combination of marginal cost levels and the degree of complementarity or substitutability determines which of the three bundling strategies is optimal. Complements and substitutes should typically be priced higher than independently valued products.

Unobserved Heterogeneity as an Alternative Explanation for 'Reversal' Effects in Behavioral Research

Behavioral researchers use analysis of variance (ANOVA) tests of differences between treatment means or chi-square tests of differences between proportions to provide support for empirical hypotheses about consumer behavior. These tests are typically conducted on data from “between-subjects” experiments in which participants were randomly assigned to conditions. We show that, despite using internally valid experimental designs such as this, aggregation biases can arise in which the theoretically critical pattern holds in the aggregate even though it holds for no (or few) individuals. First, we show that crossover interactions – often taken as strong evidence of moderating variables – can arise from the aggregation of two or more segments that do not exhibit such interactions when considered separately. Second, we show that certain context effects that have been reported for choice problems can result from the aggregation of two (or more) segments that do not exhibit these effects when considered separately. Given these threats to the conclusions drawn from experimental results, we describe the conditions under which observed heterogeneity can be ruled out as an alternative explanation based on one or more of the following: a priori considerations, derived properties, diagnostic statistics, and the results of latent class modelling. When these tests cannot rule out explanations based on unobserved heterogeneity, this is a serious problem for theorists who assume implicitly that the same theoretical principle works equally for everyone, but for random error. The empirical data patterns revealed by our diagnostics can expose the weakness in the theory but not fix it. It remains for the researcher to do further work to understand the underlying constructs that drive heterogeneity effects and to revise theory accordingly.

Measuring Market Efficiency and Welfare Loss

This study presents a general methodology capable of addressing a number of fundamental questions in consumer policy. Are consumers paying more than the minimum price for a given bundle of attributes? If so, what brands cost more than the consumer needs to pay? What would be the degree of improvement in the consumers well being if some intervention sets the price of such inefficient brands at the efficient level? We apply the methodology to data on automobiles and several other goods and analyze the determinants of efficiency.

Discrete and Continuous Representations of Unobserved Heterogeneity in Choice Modeling

We attempt to provide insights into how heterogeneity has been and can be addressed in choice modeling. In doing so, we deal with three topics: Models of heterogeneity, Methods of estimation and Substantive issues. In describing models we focus on discrete versus continuous representations of heterogeneity. With respect to estimation we contrast Markov Chain Monte Carlo methods and (simulated) likelihood methods. The substantive issues discussed deal with empirical tests of heterogeneity assumptions, the formation of empirical generalisations, the confounding of heterogeneity with state dependence and consideration sets, and normative segmentation.

Geographic Patterns in Customer Service and Satisfaction: An Empirical Investigation

When firms’ customers are located in geographically dispersed areas, it can be difficult to manage service quality because its relative importance is likely to vary spatially. This article shows how addressing such spatial aspects of satisfaction data can improve managements ability to implement programs aimed at enhancing service quality. Specifically, managers can identify areas of high service responsiveness, that is, areas in which overall satisfaction is low but customers are highly responsive to improvements in service quality. The authors estimate the spatial patterns using geographically weighted regression, a technique that accounts for spatial dependence in the variables. They apply this methodology to a large national sample of automobile customers served by a network of dealerships across the United States. The authors also investigate the extent to which factors related to the physical and psychological landscape explain the importance that people in different regions place on dealership service and vehicle quality.