Dick R. Wittink

A model of consumer perceptions and store loyalty intentions for a supermarket retailer

Abstract Slow growth and intense competition in retail markets in recent years increases the need for retailers to use strategies focused on retaining and attracting the right customers. However, a strategy that is effective in acquiring new customers may not be the most effective in retaining current customers. In order to understand the effectiveness of activities designed to retain customers, we study the store loyalty intentions of current customers for a multi-store grocery retailer. Using Partial Least Squares, on data averaged across at least 100 customers per store for each of about 160 stores, we find that service quality is by far the most critical determinant of merchandise quality perception. Perceivged value for money depends on perceived relative price and sales promotion perceptions, and to a lesser extent on service quality and merchandise quality perceptions. Store loyalty intentions, measured by intent to continue shopping, intent to increase purchases and intent to recommend the store, depend on service quality and merchandise quality perception. By separating the stores according to average consumer perceptions of competitor attractiveness, we further find that perceived value does play an important role in the determination of store loyalty intention if there is a high degree of competitor attractiveness. When this attractiveness is low, our results fail to show a relevance for perceived value for money.

Commercial use of conjoint analysis in Europe: Results and critical reflections

Abstract We report the incidence of conjoint analysis applications by European market research suppliers. Based on responses to a survey, we document about 1,000 commercial projects over a five-year period and show breakdowns by product category, study purpose, and other characteristics such as study design, data collection, and data analysis. The results are compared with information collected for an earlier time period in the United States. We also discuss recent conjoint research results, identify issues that warrant further study, and suggest how the current practice of conjoint analysis can be improved.

Is 75% of the Sales Promotion Bump Due to Brand Switching? No, Only 33% Is

Several researchers have decomposed sales promotion elasticities based on household scanner-panel data. A key result is that the majority of the sales promotion elasticity, approximately 74% on average, is attributed to secondary demand effects (brand switching) and the remainder is attributed to primary demand effects (timing acceleration and quantity increases). The authors demonstrate that this result does not imply that if a brand gains 100 units in sales during a promotion, the other brands in the category lose 74 units. The authors offer a complementary decomposition measure based on unit sales. The measure shows the ratio of the current cross-brand unit sales loss to the current own-brand unit sales gain during promotion; the authors report empirical results for this measure. They also derive analytical expressions that transform the elasticity decomposition into a decomposition of unit sales effects. These expressions show the nature of the difference between the two decompositions. To gain insight into the magnitude of the difference, the authors apply these expressions to previously reported elasticity decomposition results and find that approximately 33% of the unit sales increase is attributable to losses incurred by other brands in the same category.

Diagnosing competitive reactions using (aggregated) scanner data

Abstract We study competitive response functions with scanner data on price and promotional activities. Causality tests are used prior to parameter estimation. The results indicate that price and feature have statistically significant causal effects more frequently than other promotional variables and they have disproportionately greater frequencies for quick reactions relative to other instruments. Multiple competitive reactions are common, although the same marketing instrument is more likely to be used to react than a different one. Reactions occur with decreasing probability over time suggesting distinctions between retailer- and manufacturer-dominated reactions. The results can also be used to identify other strategic differences, such as the classification of competitors as leaders and followers.

The effect of differences in the number of attribute levels on conjoint results

It is well known that the range of attribute variation used in a conjoint design influences the inferred attribute importance. However, even if the range is held constant, the addition of intermediate levels can increase this importance. In this paper we show why the problem occurs for rankorder preferences. The results from an experimental study confirm the existence of a systematic influence due to the number of (intermediate) levels. Surprisingly, the problem is equally strong when rating scale preferences are collected. Several possible solutions are suggested.

Varying parameter models to accommodate dynamic promotion effects

Abstract The purpose of this paper is to examine the dynamic effects of sales promotions. We create dynamic brand sales models (for weekly store-level scanner data) by relating store intercepts and a brands own price elasticity to a measure of the cumulated previous price discounts – amount and time – for that brand as well as for other brands. The brands own non-price promotional response parameters are related to the time since the most recent promotion for that brand as well as for other brands. We demonstrate that these dynamic econometric models provide greater managerial relevance than static models can.

Patient preferences for treatment of rheumatoid arthritis.

Objective: To elicit treatment preferences of patients with rheumatoid arthritis (RA) for disease modifying antirheumatic drugs (DMARDs) with varying risk profiles. Methods: Patient values for 16 DMARD characteristics were ascertained using published data about side effects, effectiveness, and cost. Patient preferences were determined by Adaptive Conjoint Analysis, an interactive computer program that predicts preferences by asking patients to make trade-offs between specific treatment characteristics. Simulations were run to derive preferences for four drugs: methotrexate, gold, leflunomide, and etanercept, under different risk-benefit scenarios. Infliximab was not included because it is given with methotrexate, and we did not include preferences for combination therapy. Based on each patient’s expressed preferences, and the characteristics of the treatments available at the time of the study, the option that best fitted each patient’s perspective was identified. Results: 120 patients (mean age 70 years) were interviewed. For the base case scenario (which assumed the maximum benefits reported in the literature, a low probability of adverse effects, and low equal monthly “co-pays” (out of pocket costs)), 95% of the respondents preferred etanercept over the other treatment options. When all four options were described as being equally effective, 88% continued to prefer etanercept owing to its safer short term adverse effect profile. Increasing etanercept’s co-pay to

Verbal versus Realistic Pictorial Representations in Conjoint Analysis with Design Attributes

30.00 decreased the percentage of patients preferring this option to 80%. Conclusions: In this study, older patients with RA, when asked to consider trade-offs between specific risk and benefits, preferred etanercept over other treatment options. Preference for etanercept is explained by older patients’ risk aversion for drug toxicity.

Forecasting with Conjoint Analysis

Abstract The current generation of high-powered graphics software offers an effective means for presenting product designs. Armed with the right tools for generating photorealistic representations of alternative designs, product development teams can obtain useful consumer input about product design attributes. However, generating computer-based models carries greater costs than producing verbal representations (written, key-word descriptions). 1 If a verbal representation can effectively communicate the relevant design and styling attributes, can product developers justify the costs associated with generating a computer-based model? Marco Vriens, Gerard H. Loosschilder, Edward Rosbergen, and Dick R. Wittink highlight a fundamental question in the choice between verbal and pictorial representations 2 : Does the type of representation used affect the nature and the quality of the results that product developers obtain? Specifically, does the type of representation used in a study affect the information that the study provides about market segmentation and the relative importance of different design attributes? And does the choice of representation type affect a study’s reliability and predictive accuracy? To address these questions, the authors conducted a study with a European subsidiary of a Japanese manufacturer of car stereo equipment. The study involves the selection of product designs from those made available by the manufacturer. Respondents were asked to evaluate both verbal representations and photorealistic pictorial representations of proposed car stereo designs. Half the respondents evaluated the verbal representations first, while the other half rated the pictorial representations first. In this study, the pictorial representations produced higher relative importance ratings for two of the three design attributes, as well as somewhat greater heterogeneity (that is, segmentation) among respondents. However, the verbal representations produced greater predictive accuracy, especially for respondents who rated the verbal descriptions after they had evaluated the pictorial representations. These results suggest that the pictorial representations improved the respondents’ understanding of the design attributes, while the verbal representations seem to facilitate judgment.

Understanding Patient Preferences for the Treatment of Lupus Nephritis With Adaptive Conjoint Analysis

Conjoint analysis is a survey-based method managers often use to obtain consumer input to guide their new-product decisions. The commercial popularity of the method suggests that conjoint results improve the quality of those decisions. We discuss the basic elements of conjoint analysis, describe conditions under which the method should work well, and identify difficulties with forecasting marketplace behavior. We introduce one forecasting principle that establishes the forecast accuracy of new-product performance in the marketplace. However, practical complexities make it very difficult for researchers to obtain incontrovertible evidence about the external validity of conjoint results. Since published studies typically rely on holdout tasks to compare the predictive validities of alternative conjoint procedures, we describe the characteristics of such tasks, and discuss the linkages to conjoint data and marketplace choices. We then introduce five other principles that can guide conjoint studies to enhance forecast accuracy.