Gilbert Swinnen

Using association rules for product assortment decisions: a case study

It has been claimed that the discovery of association rules is well-suited for applications of market basket analysis to reveal regularities in the purchase behaviour of customers. Moreover, recent work indicates that the discovery of interesting rules can in fact only be addressed within a microeconomic framework. This study integrates the discovery of frequent itemsets with a (microeconomic) model for product selection (PROFSET). The model enables the integration of both quantitative and qualitative (domain knowledge) criteria. Sales transaction data from a fullyautomated convenience store is used to demonstrate the effectiveness of the model against a heuristic for product selection based on product-specific profitability. We show that with the use of frequent itemsets we are able to identify the cross-sales potential of product items and use this information for better product selection. Furthermore, we demonstrate that the impact of product assortment decisions on overall assortment profitability can easily be evaluated by means of sensitivity analysis.

A data mining framework for optimal product selection in retail supermarket data: the generalized PROFSET model

In recent years, data mining researchers have developed efficient association rule algorithms for retail market basket analysis. Still, retailers often complain about how to adopt association rules to optimize concrete retail marketing-mix decisions. It is in this context that, in a previous paper, the authors have introduced a product selection model called PROFSET. This model selects the most interesting products from a product assortment based on their cross-selling potential given some retailer defined constraints. However this model suffered from an important deficiency: it could not deal effectively with supermarket data, and no provisions were taken to include retail category management principles. Therefore, in this paper, the authors present an important generalization of the existing model in order to make it suitable for supermarket data as well, and to enable retailers to add category restrictions to the model. Experiments on real world data obtained from a Belgian supermarket chain produce very promising results and demonstrate the effectiveness of the generalized PROFSET model.

Building an Association Rules Framework to Improve Product Assortment Decisions

It has been claimed that the discovery of association rules is well suited for applications of market basket analysis to reveal regularities in the purchase behaviour of customers. However today, one disadvantage of associations discovery is that there is no provision for taking into account the business value of an association. Therefore, recent work indicates that the discovery of interesting rules can in fact best be addressed within a microeconomic framework. This study integrates the discovery of frequent itemsets with a (microeconomic) model for product selection (PROFSET). The model enables the integration of both quantitative and qualitative (domain knowledge) criteria. Sales transaction data from a fully automated convenience store are used to demonstrate the effectiveness of the model against a heuristic for product selection based on product-specific profitability. We show that with the use of frequent itemsets we are able to identify the cross-sales potential of product items and use this information for better product selection. Furthermore, we demonstrate that the impact of product assortment decisions on overall assortment profitability can easily be evaluated by means of sensitivity analysis.

Comparing complete and partial classification for identifying customers at risk

This paper evaluates complete versus partial classification for the problem of identifying customers at risk. We define customers at risk as customers reporting overall satisfaction, but these customers also possess characteristics that are strongly associated with dissatisfied customers. This definition enables two viable methodological approaches for identifying such customers, i.e. complete and partial classification. Complete classification entails the induction of a classification model to discriminate between overall dissatisfied and overall satisfied instances, where customers at risk are defined as overall satisfied customers who are classified as overall dissatisfied. Partial classification entails the induction of the most prevalent characteristics of overall dissatisfied customers in order to discover overall satisfied customers who match these characteristics. In our empirical work, we evaluate complete and partial classification techniques and compare their performance on both quantitative and qualitative criteria. The intent of the paper is not on proving the superiority of partial classification, but rather to provide an alternative and valuable approach that offers new and different insights. In fact, taking predictive accuracy as the performance criterion, results for this study show the superiority of the complete classification approach. On the other hand, partial classification offers additional insights that complete classification techniques do not offer, i.e. it offers a rule-based description of criteria that lead to dissatisfaction for locally dense regions in the multidimensional instance space.

Identifying latently dissatisfied customers and measures for dissatisfaction management

Customer satisfaction continues to be an important topic in the financial services industry. However, there is an increasing awareness that customer satisfaction as such is not enough. Distinguishes between overall satisfied customers and latently dissatisfied customers; the latter being those customers who, although reporting satisfaction in a survey, have other characteristics (i.e. satisfaction with specific service items and/or socio‐demographic characteristics) that resemble dissatisfied customers. The identification of these latently dissatisfied customers may function as an early warning signal. Indeed, their probability to defect is relatively high and can be compared to that of dissatisfied customers. Proposes a data mining technique called “characteristic rules” to identify latently dissatisfied customers of a Belgian bank. Appropriate marketing actions (dissatisfaction management) may help to avoid these customers leaving. Therefore, the objective of this study is to provide scholars and business managers with theoretical, methodological and managerial insights into identifying latently dissatisfied customers.

The improvement of response modeling: combining rule-induction and case-based reasoning

Abstract Direct mail is a typical example for response modeling to be used. In order to decide which people will receive the mailing, the potential customers are divided into two groups or classes (buyers and non-buyers) and a response model is created. Since the improvement of response modeling is the purpose of this paper, we suggest a combined approach of rule-induction and case-based reasoning. The initial classification of buyers and non-buyers is done by means of the C5-algorithm. To improve the ranking of the classified cases, we introduce in this research rule-predicted typicality . The combination of these two approaches is tested on synergy by elaborating a direct mail example.

Fashion Store Personality: Scale Development and Relation to Self-Congruity Theory

Abstract Over five decades ago, Martineau (1958, p. 47) introduced the notion of store personality (SP), which he defined as “the way in which the store is defined in the shopper’s mind, partly by its functional qualities and partly by an aura of psychological attributes”. The strategic role of these symbolic, humanlike attributes that can be attributed to stores, has been empirically demonstrated, with respect to customer satisfaction and perceived retail differentiation (Chun & Davies, 2006), as well as with respect to store patronage and loyalty behavior (Sirgy & Samli, 1985; Zentes, Morschett, & Schramm-Klein, 2008). A potential and reasonable explanation of the power of associations with humanlike personality, can be found in the self-congruity theory. This theoretical framework argues that if retailers can position their stores in such a way that the store’s personality is congruent with that of target shoppers, they are likely to succeed in attracting and retaining these consumers (Bellenger, Steinberg, & Stanton, 1976; Zentes et al., 2008), which would in turn enhance profitability (Sirgy, Grewal, & Mangleburg, 2000). However, self-image congruity is still in its infancy in retailing research (Chebat, El Hedhli, & Sirgy, 2009; O’Cass & Grace, 2008; Sirgy et al., 2000), in comparison to the extensive corroborations of the theory across many product categories (Sirgy, 1982). The present study aims to fill this gap in the literature, focusing on fashion retailing in particular. As image and identity concepts are arguably more salient in fashion retailing than in any other sector (Cheng, Hines, & Grime, 2008; Zentes et al., 2008), store personality perceptions can be expected to be particularly relevant in this area of the retail industry. Compared to weekly grocery shopping, for instance, shopping for clothes is an opportunity for self-expression par excellence (Buttle, 1992; for an overview of the role of clothing, see Burns, 2010). We start by introducing “concept-scale interaction” effects in Section 1. This phenomenon implies that rather than applying general brand/store personality scales such as the ones that were developed by Aaker (1997) or d’Astous and Lévesque (2003), it is advisable to adopt a contextual approach by developing a SP scale tailor-made for fashion retailing. In Section 2, we subsequently define the construct “fashion store personality” according to Rossiter’s (2002) C-OAR-SE procedure and Section 3 briefly discusses self-congruity theory and its applications in a retail context. Subsequently, a measurement instrument is developed for Fashion Store Personality (FSP) in Section 4. The scale construction procedure consists of three steps. First, repertory grid analysis is carried out in fifty-one individual interviews, in order to generate adjectives that people naturally use to describe the personality of fashion stores. Female participants dominate throughout this study as the interest in fashion is characteristic among women within this age range (Evans, 1993) and clothing for men is often bought by their female partner (Banister & Hogg, 2004). After a preliminary purification of the elicited item pool, a consumer survey (n=481) is carried out. By means of Principal Component Analysis with Varimax rotation five underlying FSP dimensions are identified in the resulting dataset: “chaos”, “innovativeness”, “sophistication”, “agreeable-ness”, and “conspicuousness”. The psychometric properties of this measurement instrument are checked and the scale’s reliability, stability and validity meet the common standards. Subsequently, based on this operationalization, the role of self-congruity in the context of fashion retailing is explored in Section 5. First, the extent to which consumers shop in fashion stores with a personality that they perceive to be similar to their own, is studied (Section 5.1). An inspection of the correlations between the consumer’s self-image and her perceptions of FSP of her most patronized fashion store supports the idea of SC theory in fashion retailing. Furthermore, a multidimensional scaling (MDS) visually identifies, in a multidimensional space, which types of consumer personality do or do not cluster with certain types of fashion store personality. Overall, the findings indicate that (1) agreeable consumers seem to patronize agreeable fashion stores, (2) open-minded and extraverted consumers correspond to innovative fashion stores, (3) sophisticated and little chaotic stores match best with conscientious consumers, and (4) neuroticism in a consumer is found to be associated with conspicuousness and chaos in a fashion store. Finally, the relative importance of the five FSP dimensions in explaining fashion store choice is assessed using a stepwise multiple discriminant analysis (Section 5.2). Overall, the results indicate that it is worthwhile to take FSP into consideration when explaining consumers store choices. In particular, chaos and sophistication are the two FSP dimensions that consumers mind most when choosing a fashion store to patronize. As self-image congruence seems to have an impact on consumers’ store choice, the findings of the present study highlight the role of FSP in retailing positioning strategies. In order for retailers to exhibit a personality that matches the one of their target group, a wide variety of tools can be used (Brengman & Willems, 2009, on determinants of fashion store personality). Further research could provide additional support for this scale’s superiority over general brand/personality scales. Moreover, a cross-cultural validation as well as a study of the scale’s use among male consumers would be valuable.

Am I cheap? Testing the role of store personality and self-congruity in discount retailing

This study examines whether consumer perceptions of store personality (SP) differ according to the stores format (hard-discount versus soft-discount and value retailing). A consumer survey (n = 306) is conducted in which respondents are asked to rate these retail formats on five SP dimensions and in terms of self-congruity (SC). The findings of both repeated measures ANOVA and hypothesis testing of proportion difference indicate that the three formats differ significantly in terms of all five SP dimensions. Moreover, consumers perceive a greater match between their self-concept and the value retailers personality than the discounters. A PLS model is estimated linking SP and SC to store loyalty and word-of-mouth intentions (WOM). The SP measurement model is only partially confirmed by the data. The effect of SC is found to dominate in explaining loyalty and WOM. These results provide valuable insights for optimising retail positioning strategies.

Comparison of some AI and statistical classification methods for a marketing case

Recent progress in data processing technology has made the accumulation and systematic organization of large volumes of data a routine activity. As a result of these developments, there is an increasing need for data-based or data-driven methods of model development. This paper describes data-driven classification methods and shows that the automatic development and refinement of decision support models is now possible when the machine is given a large (or sometimes even a small) amount of observations that express instances of a certain task domain. The classifier obtained may be used to build a decision support system, to refine or update an existing system and to understand or improve a decision-making process. The described AI classification methods are compared with statistical classification methods for a marketing application. They can act as a basis for data-driven decision support systems that have two basic components: an automated knowledge module and an advice module or, in different terms, an automated knowledge acquisition/retrieval module and a knowledge processing module. When these modules are integrated or linked, a decision support system can be created which enables an organization to make better-quality decisions, with reduced variance, probably using fewer people.

The changing consumer in Belgium

Abstract Centrally located at the cross-roads of the main European Cultures that constitute its roots, Belgium can in many respects be regarded as a miniature Europe. Indeed, most general economic and consumption tendencies that have a grip over Europe, also reign the Belgian society. Yet, Belgiums macro marketing mix exhibits a number of particular characteristics. The remarkably developed retail apparatus and extensive product assortments are typical of a small but rich consumption society. At the same time, promotional expenditures remain modest, and price competition in major sectors is fierce. In view of the future, the foregoing aspects characterize Belgium as a potentially attractive test market for internationally oriented firms. Conversely, it appears to be an interesting commercial arena only for marketers with high cost efficiency, or with superior targeting skills allowing them to cater to the needs of small market segments with considerable purchasing power.