This research presents a new approach to derive recommendations for segment-specific, targeted marketing campaigns on the product category level. The proposed methodological framework serves as a decision support tool for customer relationship managers or direct marketers to select attractive product categories for their target marketing efforts, such as segment-specific rewards in loyalty programs, cross-merchandising activities, targeted direct mailings, customized supplements in catalogues, or customized promotions. The proposed methodology requires cus- tomers' multi-category purchase histories as input data and proceeds in a stepwise manner. It combines various data compression techniques and integrates an opti- mization approach which suggests candidate product categories for segment-specific targeted marketing such that cross-category spillover effects for non-promoted categories are maximized. To demonstrate the empirical performance of our pro- posed procedure, we examine the transactions from a real-world loyalty program of a major grocery retailer. A simple scenario-based analysis using promotion responsiveness reported in previous empirical studies and prior experience by domain experts suggests that targeted promotions might boost profitability between 15 % and 128 % relative to an undifferentiated standard campaign.

abstract: Private labels command a growing share of food retailers' shelf space. In this dissertation, I explain this phenomenon as resulting from "umbrella branding," or the ability of a single brand to reach across categories. Conceptually, I define umbrella branding as a behavioral attribute that describes a shopper's tendency to ascribe a performance bond to a brand, or to associate certain performance characteristics to a private label brand, across multiple categories. In the second chapter, I describe the performance bond theory in detail, and then test this theory using scanner data in the chapter that follows. Because secondary data has limitations for testing behavioral theories, however, I test the performance bond theory of umbrella branding using a laboratory experiment in the fourth chapter. In this chapter, I find that households tend to transfer their perception of private label performance across categories, or that a manifestation of umbrella branding behavior can indeed explain private labels' success. In the fifth chapter, I extend this theory to compare umbrella branding in international markets, and find that performance transference takes its roots in consumers' cultural backgrounds. Taken together, my results suggest that umbrella branding is an important behavioral mechanism, and one that can be further exploited by retailers across any consumer good category with strong credence attributes. / Dissertation/Thesis / Doctoral Dissertation Agribusiness 2014

The broad objective of this study is to examine how assortment factors and category performance are related within the context of specialty retailing. This study formulates two clusters of research questions. The first cluster of research questions focuses on product assortment in general, for example assortment variety and composition. The second cluster of research questions concentrates on a specific area of product assortment, that is, private label products. An organic retailer in Australia collaborates by providing its assortment records and sales reports. The Australian organic retailing industry is an ideal candidate for this study for 1) it is specialty retailing, and 2) the supply situation allows organic retailers considerable flexibility to experiment with different assortment compositions. This study analyses store level cross sub category data and, to supplement this, it conducts a qualitative study and collects field data. Included in the cross sub category analyses are approximately 140 to 180 organic sub categories. The results show that assortment variety has a positive influence on sub category sales. The strength of this positive relationship varies across different sub category types, for example food or non-food. For the private label analyses, the results show that, within the focal store, private label SKUs are more likely to be present in sub categories with larger sales and with supermarket competition present. This study also finds that a deeper manufacturer brand assortment hurts private label performance. This study contributes to a body of cross category empirical generalisations about the complex decisions retailers face by examining the effects of assortment decisions within the context of specialty retailing. It provides some clear empirical evidence for how assortment factors and sub category performance are related through an empirical investigation in a bricks and mortar retail environment. In addition, it tests the generalisability of extant private label research beyond the much discussed conventional supermarket industry and convenience consumer goods contexts. Keywords: assortment, private label, store brand, specialty retailing, cross category, sub category, empirical investigation, organic retailing, Australia

The broad objective of this study is to examine how assortment factors and category performance are related within the context of specialty retailing. This study formulates two clusters of research questions. The first cluster of research questions focuses on product assortment in general, for example assortment variety and composition. The second cluster of research questions concentrates on a specific area of product assortment, that is, private label products. An organic retailer in Australia collaborates by providing its assortment records and sales reports. The Australian organic retailing industry is an ideal candidate for this study for 1) it is specialty retailing, and 2) the supply situation allows organic retailers considerable flexibility to experiment with different assortment compositions. This study analyses store level cross sub category data and, to supplement this, it conducts a qualitative study and collects field data. Included in the cross sub category analyses are approximately 140 to 180 organic sub categories. The results show that assortment variety has a positive influence on sub category sales. The strength of this positive relationship varies across different sub category types, for example food or non-food. For the private label analyses, the results show that, within the focal store, private label SKUs are more likely to be present in sub categories with larger sales and with supermarket competition present. This study also finds that a deeper manufacturer brand assortment hurts private label performance. This study contributes to a body of cross category empirical generalisations about the complex decisions retailers face by examining the effects of assortment decisions within the context of specialty retailing. It provides some clear empirical evidence for how assortment factors and sub category performance are related through an empirical investigation in a bricks and mortar retail environment. In addition, it tests the generalisability of extant private label research beyond the much discussed conventional supermarket industry and convenience consumer goods contexts. Keywords: assortment, private label, store brand, specialty retailing, cross category, sub category, empirical investigation, organic retailing, Australia

In this thesis, we develop a SKU level market basket model and apply the model to investigate cross-category reference price effects. This research extends previous work on the category-level multivariate logit model (Russell and Petersen 2000). Our model is a generalization of the multivariate logit model which allows for both complementarity and substitution effects at the brand level. The modeling effort in this thesis allows us to use conditional probability distributions of individual items to construct the final joint-distribution of all possible basket selections. The resulting model is very flexible and accommodates a large variety of market structure patterns. The model structure implies that the changes in brand-level marketing variables directly affect category incidence (by altering category attractiveness) and indirectly determine market basket composition. Because the model can be written in a closed form manner, we can easily study the pattern of brand price competition by computing a matrix of cross-price elasticities. We use scanner panel data for the yogurt category to demonstrate the structural flexibility of the model. The results from this application reveal asymmetric competition consistent with price-tier competition literature. We use this model to investigate how consumers' responses to reference prices within a category spillover into their choices across multiple categories. The notion is that a consumer's subjective judgment of the fairness of the price levels in one category influences the choice decisions of related items in other categories. We begin with building within-category SKU-level model based on previous findings from single category reference price models (i.e., internal versus external reference prices, asymmetric response due to loss aversion, and heterogeneity in response across consumers). We then develop four alternative model specifications for cross-category spillover effects and test competing theories about those effects. Using scanner panel data for detergent and softener categories, we discover valuable implications for reference price effects. First, SKU-level reference price effects exist and improve forecasting ability. Second, those reference price effects influence category attractiveness, but do no spillover across categories. Finally, category-level reference dependent evaluation may exist but not be important in forecasting.

This dissertation examines cross-category effects in consumer purchases from the big data and analytics perspectives. It uses data from Nielsen Consumer Panel and Scanner databases for its investigations. With big data analytics it becomes possible to examine the cross effects of many product categories on each other. The number of categories whose cross effects are studied is called category scale or just scale in this dissertation. The larger the category scale the higher the number of categories whose cross effects are studied. This dissertation extends research on models of cross effects by (1) examining the performance of MVP model across category scale; (2) customizing artificial neural network (ANN) techniques for large-scale cross effect analysis; (3) examining the performance of ANN across scale; and (4) developing a conceptual model of spending habits as a source of cross effect heterogeneity. The results provide researchers and managers new knowledge about using the two techniques in large category scale settings The computational capabilities required by MVP models grow exponentially with scale and thus are more significantly limited by computational capabilities than are ANN models. In our experiments, for scales 4, 8, 16 and 32, using Nielsen data, MVP models could not be estimated using baskets with 16 and more categories. We attempted to and could calibrate ANN models, on the other hand, for both scales 16 and 32. Surprisingly, the predictive results of ANN models exhibit an inverted U relationship with scale. As an ancillary result we provide a method for determining the existence and extent of non-linear own and cross category effects on likelihood of purchase of a category using ANN models. Besides our empirical studies, we draw on the mental budgeting model and impulsive spending literature, to provide a conceptualization of consumer spending habits as a source of heterogeneity in cross effect context. Finally, after a discussion of conclusions and limitations, the dissertation concludes with a discussion of open questions for future research.

Die vorliegende Arbeit besteht aus vier Aufsätzen, die sich mit der Modellierung von Markenwahlverhalten und Markentreue beschäftigen. Der erste Aufsatz gibt eine Einführung in das Nested Logit Modell und weist auf die Existenz von zwei unterschiedlichen Spezifikationen hin. Das utility maximization nested logit (UMNL) und das non-normalized nested logit (NNNL) Modell besitzen unterschiedliche Eigenschaften, die die Schätzergebnisse beeinflussen. Mit einer Simulationsstudie werden die Konsequenzen der Verwendung verschiedener Softwarepakete demonstriert. Außerdem wird gezeigt, dass nur die UMNL Spezifikation bei Auferlegung einer Parameterrestriktion mit der Zufallsnutzentheorie konform ist. Der zweite Aufsatz untersucht anhand von realen Haushaltspaneldaten den Erfolg einer Familienmarkenstrategie. Die Signaling Theorie liefert einen Rahmen für die dem Markenwahlverhalten zugrunde liegenden psychologischen Prozesse zur Entstehung und Erklärung von produktkategorieübergreifender Markentreue. In einer empirischen Studie wird untersucht, inwieweit in einer Kategorie markentreue Kunden dieser Marke auch in anderen Produktkategorien treu sind. Es wird ein Markentreue-Hebel-Index entwickelt. Im dritten Aufsatz stehen die psychologischen Determinanten von kategorieübergreifenden Zusammenhängen im Markenwahlverhalten im Mittelpunkt. In einer empirischen Studie wird die Risikoaversion als entscheidender Bestimmungsfaktor von kategorieübergreifender Markentreue untersucht. Die konsumentenspezifische Risikoaversion wird dabei über Innovativeness und Status Quo Bias erfasst. Im vierten Aufsatz wird das Hybride Wahlmodell einem breiten Marketingpublikum vorgestellt. Klassische Wahlmodelle gehen davon aus, dass das beobachtbare Verhalten das Resultat eines nicht spezifizierten Evaluationsprozesses des Individuums ist. Der kausalanalytische Ansatz hingegen erlaubt die Spezifikation nicht direkt messbarer Faktoren als latente Variablen und kann somit Wahlmodelle sinnvoll ergänzen. / This thesis is composed of four essays that pick up topics in brand choice and brand loyalty modeling. The first essay gives an introduction to the nested logit model and points attention to the existence of two different specifications. The utility maximization nested logit (UMNL) model and the non-normalized nested logit (NNNL) model have different properties which impact the estimation results. In a simulation study, the consequences of the usage of different software packages for model estimation on the estimation results is demonstrated. It is also shown that only the UMNL specification with an imposed parameter restriction is consistent with the underlying random utility theory. The second essay investigates the success of an umbrella branding strategy using household panel data. Signaling theory provides a framework for the underlying psychological processes in consumers'' brand choice behavior and can contribute in the formation and explanation of loyalty to the brand in multiple categories. An empirical study determines whether there is a tendency for loyal consumers from one product category to be loyal to the same brand in other product categories as well. Therefore, a cross-category brand loyalty leverage index is developed. In the third essay, consumer-specific psychological determinants of cross-category relations between brand loyal choice decisions are discussed. In an empirical study, the concept of risk aversion is considered as the key determinant of cross-category brand loyalty. Consumers'' risk aversion is derived from their innovativeness and status quo bias. In the fourth essay, the hybrid choice model is introduced to the broad marketing audience. Traditional choice models assume that observable behavior results from an unspecified evaluation process of the observed individual. The causal-analytic approach offers the possibility to specify not directly measurable factors as latent variables, and can thus reasonably supplement choice models.

In der akademischen Fachliteratur existieren kaum Forschungsergebnisse zu One-to-One-Marketing, die auf Anwendungen im Einzelhandel ausgerichtet sind. Zu den Hauptgründen zählen, dass Ansätze nicht auf die Größe typischer Einzelhandelsanwendungen skalieren und dass die Datenverfügbarkeit auf Händler und Marketing-Systemanbieter beschränkt ist. Die vorliegende Dissertation entwickelt neue deskriptive, prädiktive und präskriptive Modelle für automatisiertes Target Marketing, die auf Representation Learning und Deep Learning basieren, und untersucht deren Wirksamkeit in Praxisanwendungen. Im ersten Schritt zeigt die Arbeit, dass Representation Learning in der Lage ist, skalierbar Marktstrukturen zu analysieren. Der vorgeschlagene Ansatz zur Visualisierung von Marktstrukturen ist vollständig automatisiert und existierenden Methoden überlegen. Die Arbeit entwickelt anschließend ein skalierbares, nichtparametrisches Modell, das Produktwahl auf Konsumentenebene für alle Produkte im Sortiment großer Einzelhändler vorhersagt. Das Deep Neural Network übertrifft die Vorhersagekraft existierender Benchmarks und auf Basis des Modells abgeleitete Coupons erzielen signifikant höhere Umsatzsteigerungen. Die Dissertation untersucht abschließend eine Coupon-Engine, die auf den entwickelten Modellen basiert. Der Vergleich personalisierter Werbeaktionen mit Massenmarketing belegt, dass One-to-One Marketing Einlösungsraten, Umsätze und Gewinne steigern kann. Eine Analyse der Kundenreaktionen auf personalisierte Coupons im Rahmen eines Kundenbindungsprogrammes zeigt, dass personalisiertes Marketing Systemnutzung erhöht. Dies illustriert, wie Target Marketing und Kundenbindungsprogramme effizient kombiniert werden können. Die vorliegende Dissertation ist somit sowohl für Forscher als auch für Praktiker relevant. Neben leistungsfähigeren Modellansätzen bietet diese Arbeit relevante Implikationen für effizientes Promotion-Management und One-to-One-Marketing im Einzelhandel. / Research on one-to-one marketing with a focus on retailing is scarce in academic literature. The two main reasons are that the target marketing approaches proposed by researchers do not scale to the size of typical retail applications and that data regarding one-to-one marketing remain locked within retailers and marketing solution providers. This dissertation develops new descriptive, predictive, and prescriptive marketing models for automated target marketing that are based on representation learning and deep learning and studies the models’ impact in real-life applications. First, this thesis shows that representation learning is capable of analyzing market structures at scale. The proposed approach to visualizing market structures is fully automated and superior to existing mapping methods that are based on the same input data. The thesis then proposes a scalable, nonparametric model that predicts product choice for the entire assortment of a large retailer. The deep neural network outperforms benchmark methods for predicting customer purchases. Coupon policies based on the proposed model lead to substantially higher revenue lifts than policies based on the benchmark models. The remainder of the thesis studies a real-time offer engine that is based on the proposed models. The comparison of personalized promotions to non-targeted promotions shows that one-to-one marketing increases redemption rates, revenues, and profits. A study of customer responses to personalized price promotions within the retailer’s loyalty program reveals that personalized marketing also increases loyalty program usage. This illustrates how targeted price promotions can be integrated smoothly into loyalty programs. In summary, this thesis is highly relevant for both researchers and practitioners. The new deep learning models facilitate more scalable and efficient one-to-one marketing. In addition, this research offers pertinent implications for promotion management and one-to-one marketing.