The bigger the better? : the role of assortment variety and presentation in redesigning the new physical bookstore

(1)

The BIGGER the BETTER?

The role of assortment variety and presentation in redesigning

the new physical bookstore

Lucienne Mathot

5815835

Business Studies

Universiteit van Amsterdam

Faculteit Economie en Bedrijfskunde

03-03-2009

Thesis begeleider: drs. F.W.J. Quix

Tweede beoordelaar: prof. dr. J.H.J.P Tettero

(2)

Abstract

Retailers use their assortment to differentiate themselves from competitors (Simonson, 1999). Physical retailers can no longer excel in offering the largest assortment as this is online. The assortment,

however, remains one of the most important aspects of attracting customers (Da Silva, 2008 and Morales, Kahn, McAlister and Broniarczyk, 2005). Changes in the depth and width of the assortment do not necessarily lead to lower perceptions of variety (Broniarczyk, Hoyer and McAlister, 1998; Campo and Gijsbrechts, 2005) thus providing an opportunity for retailers to cut products and reduce costs. In addition, the way the assortment is organized also influences the perception of the customer. In this thesis a research on a physical book store is conducted, putting extra difficulties on the subject of assortment because of the Dutch fixed book price agreement and because of the notion of experience goods and quality. This thesis looks at the question of how a book retailer can better fulfil her customer’s in-store assortment needs now their positioning of assortment has changed. It investigates the

importance customers add to assortment variety and the presentation of the assortment contrasted with actual assortment performance in-store to provide a framework for practioners how to adjust their assortment to their performance and the actual assortment wishes of the customer.

(3)

Preface

Although starting a second master and writing a second thesis could have been more of the same. By choosing the subject of business administration and retailing, the challenges have been quite different, the requirements more practical and the learning on the process has added new insights. Therefore, I would like to thank Frank Quix for the introduction in retailing, his advice concerning my thesis at BGN and his practical guidance along the way. When new in a company, the first month is all about getting to know the business and people you work with. Add to that the newness of book retailing, company culture and immediately coming up with a subject for your thesis. A challenge, which I couldn’t have completed without the help and knowledgeable advice provided by various people. I would like to thank BGN (particularly Matthijs van der Lely and Arjen Luykx) for providing me the opportunity to research their company, learn all about book retailing, giving me guidance and sharing their knowledge on this and other subjects. Specifically I would like to thank Jesse Kröger for guiding me in the process and advising me on my thesis and Maarten van den Bovenkamp for introducing me to all the work at BGN, his daily involvement and sharing his insights and knowledge. In addition I would also like to thank all other employees at BGN for their help, advice and friendly working atmosphere.

(4)

3.3 The winkelcompas ... 15 3. 4 Conclusion ... 15 4. Literature review ... 16 4.1 Product assortment ... 16 4.2 Assortment satisfaction ... 16 4.3 Assortment variety ... 18 4.4 Assortment presentation ... 19 4.5 Assortment organization ... 20 4.6 Shopping aids ... 20 4.7 Assortment performance ... 23

4.8 Consumer decision making ... 23

4. 9 Conclusion ... 24

5. Analysis ... 26

(5)

5.2 Survey research ... 27

5.3 Sampling ... 27

5.4 Importance-performance mapping ... 28

5.5 Strategic Resource Management model ... 29

5.6 Development of the questionnaire; the interviews at BGN ... 30

5.7 Segmentation ... 33

5.8 Preliminary results ... 34

6. Overall results ... 35

6.1 Exploring the results ... 35

6.2 Importance of assortment needs ... 38

6.3 Importance performance analysis ... 40

6.4 Segmentation ... 44

6.5 Interpretation of the results ... 49

7. Store analysis ... 51

7.1 Consumer behavior ... 51

7.2 Store analysis Leiden ... 56

7.3 IPA analysis assortment presentation Leiden ... 58

7.4 Assortment variety Leiden ... 60

8. Conclusion and discussion ... 67

8.2 Advice to BGN ... 70

8.3 Reflection and shortfalls of the research ... 75

8.4 Implications ... 77

Concluding remark ... 78

References ... 79

Appendix A Interviewlist BGN ... 82

Appendix B origin of questions ... 83

Appendix C Preliminary results ... 84

(6)

Appendix E Example data sheet Kooyker Leiden ... 90 Abbreviations ... 92

(7)

1. Introduction

Older and recent developments have put constrains on product assortment in the classical book retailer store. An inability to differentiate on a price and quality level, the notion of books as ‘experience goods’, and the fixed book price agreement are all characteristics that identify a typical Dutch bookstore. Besides these difficulties, recent developments have changed the way book retailers look at their assortments. While retailers use their assortment to differentiate themselves from competitors (Simonson, 1999) physical book retailers can no longer excel in offering the largest assortment as this is online, thus causing necessary changes in the positioning of a book store. The assortment, however, remains one of the most important aspects of attracting customers (Da Silva, 2008 and Morales, Kahn, McAlister and Broniarczyk, 2005). This not only incorporates breadth and depth of the assortment it also concerns the way it is or can be organized (Childers, Carr, peck and Carson, 2001) including the use of

recommendation agents and shopping aids (Haubl and Trifts in Chang and Burke, 2007: 340). As experience based consumption and emotional experiences in retailing are gaining momentum, does a book retailer has to look for new ways to fulfil their customer’s in-store assortment needs?

The organization under study, Boekhandel Groep Nederland (BGN), tries to offer excellent choice, customer service and knowledge to differentiate from competition, making service and assortment top priorities. Once a bookstore with an academic orientation and a wide and varied assortment, because of the notion of the long tail, the positioning of BGN is changing. Nevertheless, BGN can create

differentiation by using its assortment variety and presentation to become more attractive to consumers. Although the stocking of books and a varied offering in the stores still gets customers appreciation, in-store book retailers need to focus on merchandise that can offer experiential attributes and thus provide tailored information to the needs of customers. One way to adjust to the wants and needs of the customers is by offering the “right’ assortment, an optimal assortment variety can cause greater satisfaction with the product at hand. On the other hand assortment presentation; attractive book presentations and retailing shopper aids can help consumers to enhance their shopping experience and help to reduce decision deferral. This way assortment variety, assortment presentation and

customer behaviour can both help customers to fulfil their in-store assortment needs. But, do customers really know what they want? Are their assortment wishes based on their actual behaviour? To look at these issues, the actual sales of assortment categories are contrasted with their assortment desires to provide a solid basis for assortment recommendations.

(8)

1.1 Research objective

This thesis is meant to advice to BGN how to design an assortment in a new store or restyle an assortment of BGN based on their customer wishes and behaviour. By providing an advice to BGN concerning their assortment variety and assortment presentation this thesis will answer the main research question.

1.2 Research question

How can BGN better fulfil her customer’s in-store assortment needs? To answer the main research question the following questions must be answered

1. What assortment issues are of importance for customers and BGN?

2. What assortment variety for BGN best satisfies its customer assortment needs? 3. What assortment presentation best satisfies BGN customers assortment needs? 4. Do different groups of customers have different assortment needs?

5. How do different BGN assortment categories perform?

6. What is the difference in customer perceptions and customer behaviors concerning assortment needs?

7. How can BGN adapt its assortment to its customer wishes and behavior?

1.3 Theoretical and practical relevance

The Dutch book market is a distinctive retail market because it consists of experience goods and deals with a fixed price agreement, making traditional models of the assortment mix harder to apply. Together with the new developments of shopping aids on assortment presentation, a research in an offline bookstore can deepen the theoretical understanding about assortment variety and presentation in bookstores. The review of Hart and Rafiq(2006) shows that assortment research has typically focused on micro -and operational levels of assortment. They advice more research in the direction of macro level assortment issues. Issues concern store assortment consistency, image of individual departments, number of departments and space allocations to departments. This thesis helps to contribute to these more macro level issues of assortment research.

(9)

atmospheric circumstances and the entertainment value of shopping need to be considered (Hart and Rafiq, 2006). This research gives owners of bookstores an advice how to adjust their assortment to the current changes in this market. It gives store owners knowledge about the perceived customer value and behaviour towards a diverse assortment and the presentation of it and provides tools to redesign their stores and think about their current positioning.

1.4 Data collection

Hart and Rafiq (2006) comment on the fact that a new department or category can not only be

measured in sales and margins, but also how this is perceived by customers. Customer assortment needs are assessed by survey research. Survey research is a method of asking questions to obtain information about others or subjects (Colman, 1995). Survey research is used when researchers want to know the opinion of people about matters or wants to ask them directly about behavior (Babbie, 2004). In this thesis non-probability sampling is used to administer a questionnaire to all of BGN customers who receive the information newsletter. The population is the customer base of BGN, the sample of customers who receive the newsletter (N=70.000) is representative for the entire customer base. The research is cross-sectional as it is only measured at a single point in time. Internet is used as the medium to administer the questionnaire, this because of the advantages of low costs and fast timing. To develop the questionnaire an extensive literature review is used together with managerial advice to develop an importance performance analysis. In the second part of the study, the assortment performance measures are collected by visiting selexyz stores and designing specific areas of genres to squared meters.

1.5 Data analysis

To analyze the data of the questionnaire, SPSS is used. To assess the needs of assortment variety and assortment presentation, descriptives of the population, correlations, segmentation and importance performance analysis will be used to address these questions (Field, 2004). Importance performance analysis (IPA) is developed by Martilla and James (1977) as a marketing tool to measure customer satisfaction on different attributes on different concepts. IPA can be used to identify areas for improvement and actions for minimizing gaps between importance and performance. Especially the difference between the means of the concepts is interesting as this shows if importance is higher than

(10)

performance and thus if improvements should be made. Importance is represented on the y-axis and performance on the x-axis. Together they form a grid of low/high performance/importance showing where each attribute scores. The attributes are measured on a 6 point likert-scale (Martilla and James, 1977).

Assortment performance will be measured with the help of the strategic resource management model (SRM) of Lush and Serpkinci (1981; in Ring, Tigert and Serpkinci , 2002). The SRM or Trinity model is a management tool for measuring performance on the level of three success factors: Inventory, space and people. In this case the gross margin per selling feet and inventory is of importance. These variables can measure what the gross margin is per genre to contrast selling importance with customer ratings. Together the results of these two methods will be contrasted with one another.

1.6 Proceedings

Before conducting a literature review, first of all the important aspects of retailing in books and the focal company; BGN, are analyzed. Thereafter in chapter three, three methods are used to assess the

competitive advantage of BGN. In chapter four literature on assortment variety, assortment presentation and consumer decision making is presented . Together these three chapters shape the theoretical background and answer the first subquestion of which assortment issues are important for the customers of BGN. In chapter five a theoretical and more practical view of the methods used in this thesis is presented to lead up to the first results of the questionnaire. In chapter six and seven

correlations, regression, segmentation and IPA analysis is used to uncover what assortment variety and presentation best satisfies the customer’s needs together with a differentiation in different groups of customers (subquestion 2, 3 and 4) and the performance of the different book categories of BGN is assessed (subquestion 5). In chapter eight the difference in customer perceptions and behavior are analyzed which results in the conclusion and advice to BGN how to adapt their assortment to the customers whishes and behavior (subquestion 6 and 7).

(11)

2. Retailing in books

Before an extensive literature review is provided, the specific circumstances of a traditional ‘brick and mortar’ bookstore in the Netherlands are assessed. This research is conducted within the context of the retail environment, the books market and the Dutch fixed book price agreement, all having different constrains on the traditional literature about consumer behavior. This chapter describes these different constrains and discusses the implications for the existing literature.

2.1 Retail market

Retailing sets itself apart from different industries by the direct contact retailers have with their customers (Mulhern,1997). According to Mulhern (1997) retailers in the past focused too much on the product instead of their relationship with the customer and the opportunity to measure the relation between merchandising and buying behavior of the customer. In the coming years this focus on customer relationships and communication is bound to intensify heavily. Retailing combines selling products with the aspect of service encounters with customers. Mulhern (1997) describes the price/quality dimension as one of the ways a store must position itself. However, in the book market variations in price are limited because of the fixed book price agreement and the quality of books is open to discussion. In retailing, quality generally refers to the product or package being consumed. A book however is consumed in a different way, by experience. Although books do come in diverse quality (hard or soft cover) the quality of the book itself is something that is assessed different by each consumer. Quality in books is thus highly subjective and can only be measured by reading the book; thus the notion of experience goods.

2.2 Books market

The book market in the Netherlands has an annual turnover of more than 946 million (2008), some bookstores even have more than 20.000 titles in their stores, not to mention the huge amount of titles available on the internet. With around 1800 stores (Hoofdbedrijfschap detailhandel, 2007), more than 60% concentrated in chains, this market is considered a dense market. On average a bookstore has an annual turnover of €795,600 in 2007 (Hoofdbedrijfschap detailhandel, 2007). The book market consists of online and offline sales channels, although online sales have grown rapidly, in 2007 offline bookstores

(12)

still were the main sales portal with a 55% market share (CBS, 2007). However, the numbers of bookstores do diminish and the Dutch population is spending less time on reading than previous years (CBS, 2007). Royle, Cooper and Stockdale(1997) comment on the fact that books are different from other consumer goods, this because of the lack of brand identification of books. Consumers do not identify with the brand or product behind the book. Books are intangible goods because of their “experience goods” status; this provides an inability to measure repeat purchase. The dense market and the inability to predict sale on a basis of repeat purchase provides an extra difficulty for the book retailer.

2.3 Fixed price agreement

The book industry in the Netherlands differs from various other industries mainly because of its fixed book price agreement. This law of the fixed book price means publishers get to set a fixed book price for all the retailers for the first two years of the edition of a book (Commisariaat voor de media, 2008). As a consequence the price for a (new) book is the same in all bookstores and online portals. The Fixed Book Price Agreement tends to promote books by assuring a high supply of diverse books, availability

everywhere and for everyone (Commisariaat voor de media, 2008 & CBP, 2002). Critics report high margins for book publishers and retailers resulting in high book prices and a distortion of the free market (van der Ploeg, 2004). However adversaries see the opportunities for a wide variation of titles, an

accessible start-up market and a larger density of bookshops (van der Ploeg, 2004). The fixed price agreement causes distinctiveness for this market, whereas marketing theories of category management, purchase decision and product attributes are based on price diversion, in the case of the fixed book price these theories are not applicable.

2.4 BGN

With an annual turnover of €186.265.000 in 2007 and more than 11 million visits a year, BGN is one of the biggest book market exploiters in the Netherlands (Jaarverslag BGN, 2007). Selexyz stores offer an assortment mix of general books (47%), scientific books (42%) and non-books (11%). BGN was founded in 1992 when publisher Wolters-Kluwer decided to spin-off its division of bookstores. Since then BGN grew strongly and acquired several leading bookstores in different location in the Netherlands. BGN operates on three different levels; consumers, business and students. To offer their services to these three groups BGN has 16 large bookstores, 26 campus bookstores and an internet portal (www.selexyz.nl) all known by their name of selexyz. BGN differentiates itself from other book retailers by its wide assortment, its

(13)

high level of service, their ambiances and their locations (Jaarverslag BGN, 2007). Their mission is to be the best bookstore in their location positioning, to offer excellence in choice, customer service and knowledge of the product. Advising, giving information and tempting their customers are the ideals (www.BGN.nl).

2.5 Conclusion

An inability to differentiate on a price/quality level, not being able to predict sales by repeat purchases, the notion of experience goods, a dense market, and the fixed price agreement are all issues that

identify a typical Dutch bookstore. Besides these difficulties BGN tries to offer excellent choice, customer service and knowledge to differentiate from competition, making service and assortment top priorities.

(14)

3. Competitive advantage

According to Barney (1991) the resource-based-view of competitive advantage describes the link

between performance and the internal characteristics of a firm. A competitive advantage can be reached when having a value creating strategy that is unique to the firm. This competitive advantage can be sustainable when duplication of the unique benefits is impossible for another organization. An

organization has got to have resources that are heterogenic and immobile and to be sustainable should be valuable, rare, inimitable and non-substitutable (Barney, 1991). It is beyond the scope of this research to asses this, however for BGN, following a differentiation strategy, service, knowledge and assortment are their main drivers for a sustainable advantage. However, recent developments like the long tail have put more pressure on the positioning of BGN in terms of their assortment. In this light, a deeper look at BGN’s competitive positioning is given and three models are presented.

3.1 The myth of excellence

Crawford and Mathews (2001) contest the idea that a company should be the best at everything. In their book the Myth of excellence they advice a company to be great at one aspect of a commercial

transaction, be better at another and be the industry average at the last three of five aspects. The five aspects consist of price, service, product, experience and access. BGN states that they want to excel in assortment, service level and knowledge, ambiance and location; respectively product, service,

experience and access. Because domination on all levels, according to Crawford and Mathews (2001) is impossible, a choice has to be made. Access is one of the aspects which is currently hard to change for BGN, as location is scarce. Their service level is stable at this point whereas positioning is changing. Therefore this thesis looks at the option to excel in product, the assortment of BGN and be better at the experience in their stores.

3.2 The pentagon model

To build a competitive advantage the pentagon model of Ring and Tigert (2001) can also be used. According to Ring and Tigert (1995) companies that follow a differentiation strategy should focus on the pentagon model (instead of the triangle model for cost leadership by grace of its system, logistics and/or partners). Ring and Tigert (2001) state that competitive advantage through differentiation can be

(15)

location and design of a store. The product consists of the assortment mix, the intensity and the

assortment angle. The value represents price and value equation. People consist of services, climate and knowledge. The positions of the pentagon represent the way the customers looks at the store or image of the store and is visible for customers and competitors (Ring & Tigert, 1995). To be able to stay competitive a retailer must keep track of its own position and rejuvenate if appropriate. In contrast to the model of Crawford and Mathews (2001) Ring and Tigert (2001 and 1995) advice to focus on all of these aspects and do not advice an order. However, in this thesis assortment and the experience are central, therefore mainly focusing upon the product variable.

3.3 The winkelcompas

The practical implications of the pentagon model can be reached by using the ‘winkelcompas’ of Quix and Hemmer (2006) which makes a differentiation on the pentagon model based on retailing and offers more guidelines to managers. Where, what, value, who and why are the buttons one can push to reach a competitive advantage. To reach a competitive advantage one has to excel at 1 or 2 points, be average at 1 or 2 points and the last 2 or 3 points can be lower than average. When it comes to BGN, the question of Where is important; however their positions are already chosen and quick expansion is not an option. The selling of what is of importance here: what should BGN sell, the intensity of the goods and also the style counts. BGN might be able to offer extra ‘What’ by a new assortment presentation and improve its positioning. The question of “value” is the price/quality perception, which is at the moment hard to use because of the Fixed Price Agreement and experience goods. The question of “Who” is service, another important aspect of BGN. The question of ‘Why’ is related to positioning and promotion. Considering BGN means that the two buttons here are ‘what’ and ‘who’. In this thesis we will focus on the changing positioning and thus aspects of ‘what’ referring to assortment variety and assortment style.

3. 4 Conclusion

The value creating strategy of BGN is focused on the service level, location and assortment. Considering BGN’s positioning and the models of Crawford and Mathews (2001) and Quix and Hemmer (2006) BGN should focus on one or two aspects to excel or be better at. In this thesis the main focus will be about the assortment; the product or the ‘what’ of a retailer. Assortment refers to both the depth and breadth, or assortment variety and the assortment style, the way the assortment is organized in helping a

(16)

4. Literature review

In this chapter an in-depth review on existing literature of the assortment of a retailer is conducted. The importance of an assortment, the product assortment and assortment variety are the first concepts in this chapter. Thereafter assortment presentation and shopping aids are looked at. At last theory about consumer decision making and attributes is discussed leading to the conclusion of the literature review.

4.1 Product assortment

Product assortment is a varying numbers of products within a category consumers can choose from (Chernev, 2003). According to Simonson (1999) a retailer can create differentiation by using its product assortment as a tool to become more attractive to consumers. By offering a more extended or a different assortment some stores might be better at fulfilling its customer needs and expectations (Simonson, 1999). The retailer decides upon this assortment and thus already makes a choice for customers. Overall product assortment has two goals: fulfilling the wants and needs of customers and the shaping of these needs and wants leading to purchases (Simonson, 1999). The fulfilling of the customer needs is dynamic, as consumers wants and needs shift, a retailers adjusts its assortment (Hart & Rafiq, 2006). The product assortment guides consumers in three ways: “ease of purchase justification, ease of information processing, and activation of decision rules” (Simonson, 1999: 364). These different functions of the product assortment will return in this review when discussing assortment variety and assortment presentation.

4.2 Assortment satisfaction

Assortment or merchandise mix has traditionally been one of the most important aspects in attracting customers and customer satisfaction (Da Silva, 2008). Store brand, store image and store preference are all strongly influenced by the specific assortment the book retailer provides in its store (Da Silva, 2008). The assortment in a book store focuses on the quantity, depth and selection of books, instead of the quality and fashion of the assortment in other retailer formats. One of the important aspects here is the notion of selection. The traditional book retailer decides which books to sell and thus makes a selection for the customer, hoping to seduce the customer into buying books. As books are experience goods the voice of the customer in this decision is somewhat smaller than in a traditional retail environment. A diverse assortment, both in depth and quantity and having a book in store when the customer wants it is

(17)

a mean concern in the bookstore and contributes to the satisfaction of the customer with the assortment and the store. Da Silva (2008) notes that assortment satisfaction in a bookstore typically concentrates on how varying the assortment is (the covering of different topics in the bookstore) and how well the books are stored (how easy it is to find something that you want). The first issue relates to assortment variety whereas the second issue relates to what this author calls assortment presentation. Recent developments in assortment satisfaction

Recent developments have caused constrains on product assortment in the classical book retailer store. According to Simonson (1999) new challenges in the environment make duplication a lot easier while attracting from the same pool of customers. Differentiation through product assortment becomes harder because of a decrease of the meaning of location and the growth of e-tailer bookstores with a ‘long tail’; the space online retailers have available for their product is unlimited. There is a greater choice of books online which causes consumers to look for forgotten books and breaking with the mainstream choices (Anderson, 2004). Anderson (2004: 173) describes in his theory about “the long tail” that while bestsellers were ruling in offline stores and only twenty percent of books had a chance of getting on the selves, the endless shelf space online makes every article available and sales are now about the “lost” chances instead of the “big hits”. Anderson(2004) reports that 50% of the sales of Amazon.com are generated by books outside the top 130.000 titles. Making every title available, cut costs and a good search engine provides lots of opportunities for online book retailers. On the contrary, offline book retailers must compete with this large online library of books.

Because of their limited space, offline bookstores already lost the battle on variety, depth and width of the assortment, raising the question if remaining to compete on variety with other offline bookstores is a viable strategy. On the contrary, according to Da Silva (2008) the assortment; the quantity of a product, the depth and the selection of products, is important for customer satisfaction. The stocking of books and a varied offering in the stores gets customers appreciation (Da Silva, 2008). So a strong merchandise mix still remains a central point in physical bookstores. Hart and Rafiq(2006) comment on the recent development in the book market. Online booksellers should offer a wide assortment which can be easily reduced by an electronic shopping assistant. In-store book retailers need to focus on merchandise that can offer experiential attributes and thus provide tailored information to the needs of customers. Merchandise presentation, atmospheric circumstances and the entertainment value of shopping need to be considered. These two issues of assortment variety and presentation are discussed here.

(18)

4.3 Assortment variety

The retail environment where people buy their products is related to the perception of variety a consumer has of that site (Morales, Kahn, McAlister and Broniarczyk, 2005). One way to adjust to the wants and needs of the customers is by offering the “right’ assortment through perceived assortment variety. According to Herpen and Pieters (2002), assortment variety is a predictor of store choice, sales and satisfaction. To differentiate as a retailer, quality and variety is key. Ailawadi and Keller(2004) report that the depth and breadth of an assortment and the service offered by a retailer contributes to the feeling of customers about the brand -and store image. An important note here is that perceived variety refers to how consumers think about the variety. The real depth and width offered at a store and the perceived variety are not the same. Unclear remains how customers asses this perceived variety, it is not a simple ‘the bigger the better’.

Assortment variety issues

A wide assortment variety can cause greater satisfaction with the product at hand and can be a

legitimate source for a higher price. On the contrary a very large assortment can diminish this effect and be considered overwhelming causing fewer purchases. Lyengar and Lepper(2000) state that consumer satisfaction is positively influenced by a large assortment and a large variety because of the freedom of choice. Chernev (2003) contests the idea that a larger assortment is better for customers because a match between products and consumer choices has a larger chance. Chernev (2003) finds that larger assortments can confuse customers and delay purchase decisions. Larger assortments can foster

customer preferences by offering a better match opportunity and giving customers the chance to change their minds and give them freedom and flexibility (Chernev, 2003). However, large assortments ask more effort of their customers, confuses customers and make comparisons across attributes almost impossible (Chernev, 2003). There is a tense line between offering too much variety, which leads to less purchases and the use of filtering (using less) which increases satisfaction but lowers the perception of assortment variety (Morales et al., 2005).

The effects of changing assortment variety are not yet clear. Herpen and Pieters(2002) note that some products and or attributes have a larger effect on perceptions of assortment variety than others. They also report that a research of Boatwright and Nunes (2001 in Herpen and Pieters, 2002: 340) showed an increase in sales volume when removing some products. Although this research was constructed in food-retailing, it shows evidence that bigger is not necessarily better. Broniarczyk, Hoyer and McAlister(1998)

(19)

state that a reduction on assortment of twenty-five percent is possible without affecting consumer variety perceptions. The reduction of items from a product category does not have to affect assortment variety when placing importance upon favorite items and not lowering category space (same width, less depth). On the other hand, Borle et al.(2005) conducted a study of the reduction of products and its effects on sales and found that reducing assortments might lower sales and might cause changes in the entire customer base. Authors seem to differ on this point; reductions might lead to short term winnings, the results in the long run and on store image however remains unclear. For book retailers, deleting “products” would actually mean a reduction in titles and also a reduction in space. Since an existing store is hard to shrink the question here will rather be about the division of space in an existing store and the needed space in a new store.

Campo and Gijsbrechts (2005) state that category management effects are not yet fully understood because a lack of realistic research methods. Long term responses to changes in the assortment might vary from short term responses. Research shows conflicting findings and different answers for different product categories, settings and consumers. Campo and Gijsbrechts (2005: 385) gathered some

academic insights and advice for changes in the size and composition of the assortment to “ensure availability of favorite items, maintain depth along key attributes: brand, size, variety and look for unique, category-enhancing additions.” Campo and Gijsbrechts (2005) add that the manner in which an assortment is presented (organized) to the customer influences the variety perceived by that customer. For a book retailer offering favorite items (bestsellers), depth among key attributes (variety), unique additions (tips) and assortment presentation enhances the overall perceived assortment variety.

4.4 Assortment presentation

Not only variety matters, also the way people perceive an assortment, the style in which it is presented, is of importance. New developments in the bookseller’s environment have made the concept of

assortment less clear cut. When it comes to an assortment online, consumers attach great value to the way the assortment is presented. This not only incorporates breadth and depth of the assortment like in traditional stores, it also concerns the way it is or can be organized (Childers, Carr, peck and Carson, 2001) and the use of recommendation agents and comparison matrixes (Haubl and Trifts in Chang and Burke, 2007: 340). A match between how consumers organize products and how stores do, diminishes search costs and causes greater satisfaction. Books are experience goods and that way hard to compare

(20)

based on clear attributes besides look and feel of the product. On the internet this problem is currently compensated by reader’s reviews, offering comments and making suggestions. In the store this is typically done by providing a top-10 of the best sold books in a week or other time frame. New developments like RFID and consumer generated content might add this extra value also in in-store purchase decisions. Retailers might want to adjust their assortment presentation based on best-practice online and provide opportunities for readers’ reviews, offering comments and making suggestions.

4.5 Assortment organization

A match between the way people organize products and the way the store organizes them makes it easier for consumers to process these items. Evaluations are based on this own categorization of the customers (Morales et al., 2005). A match between the categorization consumers use internally and the categorization in stores leads to higher perceptions of assortment variety (Morales et al., 2005). On the contrary a match between the internal categorization of customers and the filter used in stores causes a perception of less variety. When consumers frequently see the same store lay out, their own

categorization intertwines with the stores categorization and therefore provides a solid match (Morales et al., 2005). If people are unfamiliar with a product assortment they tend to look at the store lay out based on their shopping goal. If retailers want their consumers to assess the store as having a large variety they should make sure consumers actually get to see this variety by demanding the customer to take a bigger look around (Morales et al., 2005). In his study, Chernev (2003) finds evidence that when customers who already have an idea what they are going to buy, are in a large assortment situation, this strengthens their choice. When offering a large assortment in a bookstore and people are just

“browsing” (like 74% does in a bookstore according to BGN managers), the large variety overwhelms them and makes decision making harder (Chernev, 2003), making the quest for shopping aids more important.

4.6 Shopping aids

One way to use on-line best practice is the use of in-store shopping aids. Shopping aids are for example recommendations, extended information on displays, reviews and an information kiosk (Backström & Johansson, 2006). Retailing shopper aids have two effects: enhancing the shopping experience and helping to reduce decision deferral. Decision deferral results from information overload, an unsatisfying assortment, difficulties in examining alternatives and out-of-stock (Chang and Burke, 2007). However

(21)

just providing these opportunities in each store does not immediately solve problems. In their study, Chang and Burke (2007) find evidence that new technologies must be tailored to the specific wishes of the different customer and product categories. Customer-based-technologies lower the search costs by handling lots of information, screening and selecting, thus improving the buying decision. However quality may be poor, especially in books were experience is a big part of the process together with subjectivity and devices might be too difficult (Burke, 2002). Will customers really want to use shopping aids while looking for books?

Complementary, Chang and Burke (2007) found that purchase goal, prior knowledge about the product and preference of heterogeneity influences the use of different shopping aids in PDA’s. Evaluative information was wished for when product knowledge and need for heterogeneity was low. Extended options were useful when customers had expert knowledge and a specific need. In Both cases decision satisfaction and the purchase decision were positively affected. Chang and Burke (2007) advice, among others, to investigate the effect of shopping aids on merchandise quality perceptions. This research was performed using PDA’s as a focal point, the main difference in books is that knowledge about a product category (books) is an experience/subjective one rather than just gathering attributes. However in their study, Aggrawal and Vaidyanathan (2003) found that although rule based recommendation and

collaborative filtering (based on self-provided attributes versus based on other buyers) is difficult for experience goods but both types were perceived to have the same effectiveness and were welcomed by customers.

The second effect of shopping aids is to enhance the shopping experience of customers. Backström and Johansson (2006) report that experience based consumption and emotional experiences in the retail store environments are gaining momentum in research. Shoppers shop for hedonic (emotional) as well as utilitarian (practical) reasons (Sullivan & Heimeijer, 2008), shopping aids can foster both type of customers, however, the way in which this is reached might be different. Backström and Johansson (2006) conduct a case study in which they find that customers have a lower need for entertaining experience then retailers think they do. Retailers especially attach value to the role of education and knowledge in creating consumers in-store experiences, ways to enhance this is by the use of digital displays, internet access, computers and information kiosks. Consumers on the other hand typically base their in-store experiences on the personnel and service level. Although this difference exists, the authors

(22)

do advice for more attention towards the topic of in-store displays when the hygiene, personnel and service levels are satisfied.

When it comes to enhancing the shopping experience, Burke (2002) found that shoppers shopping for books want fun shopping experiences and privacy. It’s not the technology or aid that adds value, it’s the way this can help consumers. Infrequent shoppers have higher informational needs while frequent shoppers have more transactional needs. Fun shoppers like shopping aids better. Complementary, Kaltcheva and Weitz (2006) found that recreational motivated shoppers like high arousal (shopping experience and fun) and this in turn heightens store visits and sales. On the other hand, task motivation people don’t like high arousal and this has a negative effect on shopping behaviour. Thus, those

shoppers looking for experience based consumption seem to advocate for the use of specialized shopping aids. Do customers with a different purchase goal (recreational versus goal-oriented) perceive the use of shopping aids any different?

Different shopping aids

What shopping aids can be used in store? In the study of Burke (2002), more than 50% of the

respondents indicated they would like to use hand-held scanners, order out-of-stock items themselves through a touch screen tablet, see electronic shelf tags and use kiosks and electronic registers. On the other hand not all shopping aids are wished for; cell phone appliances, sms texting for waiting in line, providing personalized advertising and a map of the store where not that welcome. However, they do form opportunities for shopping aids. Besides these, the use of recommendation agents and comparison matrixes online can be extended to the shop floors. Kourouthanassis, Giaglis and Vrechopoulos (2007) extend the aids by saying that sensing and recognition technologies help the development of these aids and enhance the user experience. Location-based-information, movable screens and wireless guides might help fulfil the efficiency assortment needs of customers because of a heightening in information processing and entertainment value. In addition to sensor technologies, RFID uses radio waves to track and identify objects leading to the use of smart shelves, kiosks with locations and personalized marketing promotions. Besides these aids mentioned in the literature interviews within BGN (a leader in the use of RFID) are conducted to presents a more thorough list of customer shopping aids specialized at

(23)

Figure 4.1 shopping aids Type of aid Options

Shopping aids Handwritten reviews, groundplan, routing, register tips, Top10, Top100, Theme’s, ads, local ads, posters, displays, shopping window, communication system.

Technological shopping aids

Information kiosk: search, profile, internet, tips, reviews GSM: waiting line, groundplan, profile, tips, reviews Scanner: ground plan, search, reviews, tips

Television: tips, reviews

4.7 Assortment performance

Behaviour and wishes of customers are not always the same. While customers might claim for ‘more is better’, their behaviour might show that they only buy the better and not the more. Meaning, customers might know what they want however their behaviour might be different. In a store environment choice is of importance; however we might still just buy our regular item. An assortment might be presented attractive by adding choice and different items, the customer might still just buy the bestseller. On the one hand assortment variety and assortment presentation provide a basis for the customer to consider the store and its assortment. On the other hand the actual buying behaviour of the customer might be different and show the consequences of an assortment. The importance an assortment has in a store can be measured by looking at the sales this genre generates in comparison to other genres. Thus the

performance of an assortment is assessed by the customer buying behaviour, or sales.

4.8 Consumer decision making

Consumers traditionally make buy decisions based on the comparison of products or attributes. People want to reduce effort and reduce uncertainty and mistakes; they want to lessen their cognitive effort (Haubl and Trifts, 2000). Consumers are able to assess the attributes of a product, define their aspects and give them utilities (Simonson, 1999) which in turn guide the purchase decision. Attributes which are easy to compare win of harder to access attributes, easy attributes in this way get overweight because of the simplicity of comparing (Simonson, 1999). For books this includes the comparison in different genres, prices (mid price, ramsj), author, lay out of the book and the use of top titles, reviews and so on.

(24)

Consumer decision making is thus influenced by the way an assortment is presented and can be influenced with shopping aids that help to compare these attributes or products. For example a theme table about an author can focus on this aspect while, a table with pockets can focus on mid/low price and a table of the house of books can focus on lay out.

Complementary, assortment variety can be based on the notion of comparing products (Hoch, Bradlow and Wansink, 1999) or attributes (Herpen and Pieters, 2002) in shaping consumer perceptions of assortment variety. In their research Herpen and Pieters (2002) conclude that attribute based

perceptions of variety correlate less with assortment size than product based perceptions. Thus, changes in assortment size and composition could be looked at from an attribute perspective (Herpen and Pieters, 2002). However, Da Silva (2008) assumes that the physical quality of a book is not considered when comparing attributes and Royle, Cooper and Stockdale (1997) comment on the fact that books are different from other consumer goods because a lack of brand identification of books. In this thesis assortment variety is looked at from a product based perspective which in turn causes a higher correlation with assortment size.

Peoples choices are not stable, can be influenced, are not always clear and can be changed (Simonson, 1999). The context in which the assortment is considered is of importance to the decision making process. Price, promotions and advertising guide theses decisions. However, when it comes to designing a new bookstore or remodeling an old one; price and quality levels are hard to measure. One the one hand the fixed book price agreement causes troubles for competing and influencing with prices, on the other hand books are experience goods making choices in quality very difficult. Ass Simonson

(1999)states; adding options, diminishing marginal value and adding quality levels to lure the consumer into buying a certain product works well. However, in a bookstore this is hard to accomplish putting more constrain on helping customers to fulfill their in-store assortment needs to help them with their decision making process.

4. 9 Conclusion

A retailer can create customer satisfaction by using its assortment variety and presentation to become more attractive to consumers. However, differentiation through the product assortment has becomes harder because of a decrease of the meaning of location and the growth of e-tailer bookstores. The

(25)

stocking of books and a varied offering in the stores gets customers appreciation, however, does an in-store book retailer need to focus on merchandise that can offer experiential attributes and thus provide tailored information to the needs of customers? One way to adjust to the wants and needs of the customers is by offering the “right’ assortment, a wide assortment variety can cause greater satisfaction with the assortment at hand and can be a legitimate source for a higher price. On the contrary a very large assortment can diminish this effect and be considered overwhelming causing fewer purchases. Besides, it seems that category management effects are not yet fully understood. The way people perceive an assortment, the style in which it is presented is very important. Assortment organization could provide a match between the way people organize products and the way the store organizes them. On the other hand retailing shopper aids can help consumers enhance their shopping experience and help to reduce decision deferral. This way assortment variety and assortment presentation can both help customers to fulfill their in-store assortment needs.

(26)

5. Analysis

In this thesis quantitative analysis will be used to be able to segment between different customer groups. Babbie (2004) makes a distinction between pure and applied research. In this thesis applied research will be the main focus, thus using research to apply to a real life case, in this case BGN and their assortment issues. This chapter will first focus on quantitative versus qualitative research and describes the survey research. Thereafter the measures to asses assortment variety, assortment presentation and assortment performance: importance-performance analysis and the strategic resource management model are discussed. At last segmentation variables, the development of the questionnaire and the first results of the survey are presented.

5.1 Quantitative versus qualitative research

According to Babbie (2004) the difference in quantitative and qualitative research is the difference between numbers and nonnumeric data. The advantages of quantitative data are the abilities to compare, aggregate and summarize data and patterns can be uncovered with the help of statistical analysis (Babbie, 2004). The downside is a loss of meaning of the data. Instead of the distinction between qualitative and quantitative research, Hanson and Grimmer (2005) make a distinction between hard positivist viewpoints at the one hand of the spectrum and the constructionist view on the other hand of the spectrum. The positivist standpoint assumes an objective truth is “out there” and can (partially) be measured. The softer version of positivism pays attention to the role of the researcher. There is a truth but this is hard to measure and one can only describe the complexity of the situation. Being able only to describe part of reality, a certain amount remains insecure (Hanson and Grimmer, 2005).

According to Hanson and Grimmer (2005) almost all marketing research is positivist research,

quantitative taking the harder stand against the softer qualitative positivist view. Lee (1986) refers to this subjective kind of research by phenomenology. Lee (1986) describes the differences between the two in terms of objective versus subjective and a reliance on statistics and data versus language and

descriptions. In quantitative research the role of the researcher is that of an outsider who shows a certain amount of detachment to the subject under study (Lee, 1986). Long, White, Friedman and Brazeal (2000) simplify this distinction into objective versus subjective or whether to “crunch numbers” or not. This researcher takes an objective positivist standpoint. This research is conducted from the

(27)

viewpoint that there is a part of reality that is measurable and numbers can help to uncover relationships between phenomena.

5.2 Survey research

Customer assortment needs are assessed by survey research. Survey research is used when researchers want to know the opinion of people about matters or want to ask them directly about behavior (Babbie, 2004). There are several forms of a survey; telephone, mail, email and personal interviewing. Each survey method has its own limitations and constrains with concern to time, money and response rates (Colman, 1995). Internet is used as the medium to administer the questionnaire, this because of the advantages of low costs and fast timing. A disadvantage might be that not all people in the customer base use internet or are described to the newsletter.

A survey is held after a sample of the population is chosen, this sample must be representative of the total population to be able to make predictions which can be generalized (Colman, 1995).Survey research makes it possible to describe the characteristics of large populations, make large samples feasible and provide general measurement (Babbie, 2004). On the other hand, this means that the research is not suitable to deal with complex social life, is somewhat inflexible and generally weak on validity and strong on reliability. Respondents must be willing to answer and answer correctly (Colman, 1995; Babbie, 2004). The research is cross-sectional as it is only measured at a single point in time. Cross-sectional research can best be used when the construct measured is externally oriented, diverse scales and measurements are used and they are rooted in theory (Rindfleish, Malter, Ganesan and Moorman, 2008) as is the case in this thesis.

5.3 Sampling

In this thesis non probability sampling is used to administer a questionnaire to all of BGN customers who receive the information newsletter (N=70.000). The population is the customer base of BGN, the sample of customers who receive the newsletter is representative for the entire customer base. The customer base of BGN consists of higher educated (HBO+) consumers. Most of them are in their thirties and forties. They typically shop at selexyz stores because they like to read, give books as a present or use it for work and studies. Seventy percent visits the store once per month or more and on average spends 23.20 euro. The problem of internet based surveys is a relative high non-response rate causing sampling

(28)

errors. However, lower response rates do not necessarily mean that the results are less representative (Krosnick, 1999). To make sure valid results are provided, assumptions will only be made on a level at which enough respondents per variable are gathered.

5.4 Importance-performance mapping

Importance performance analysis (IPA) is developed by Martilla and James (1977) as a marketing tool to measure customer satisfaction on different attributes on different concepts. Although mostly used for service quality, it is also used for satisfaction assessment (Olujide and Majebi, 2007). IPA can be used to identify areas for improvement and actions to minimize gaps between importance and performance. The attributes used in the importance-performance section can best be selected through managerial

judgement, focus group interviews, unstructured interviews or previous research findings and are generally measured on a 5 or 7 point Likert-scale. In this thesis managerial judgement is assessed by semi-structured interviews (5) with the CEO, commercial manager, marketing manager, overall store manager and project leader of the VVO project (measuring store performance). Together with previous research findings these provide the input for the questionnaire. Together importance and performance form a grid whereby importance is represented on the y-axis and performance on the x-axis.

Figure 5.1 IPA grid

The positioning of the axes can be adjusted to the levels scores. When ranked on a 5 point likert scale and answers are normally distributed 2,5 can be used as the intersection point. When all attributes score well the axis might be shifted, for example towards the mean to prioritize better.

I m p o r t a n c e Performance Low High High Low Concentrate

here Keep up good work

(29)

A lot of critique on this measurement method has risen over the years (Olujide and Mejabi, 2007). The low levels of the predictive power, the statistical measures (means) used and the simplicity of the model all have received some criticism, as well as choosing attributes, analyzing the model and positioning the axes of the model. Using the median instead of mean scores, using regression and correlation analysis and not using IPA at all were some of the options to improve the method. Bacon (2003) tested several of these improvements to IPA and found that using mean scores is a better predictor than correlation or regression scores. Especially the difference between the means of the concepts is interesting as this shows if importance is higher than performance and thus if improvements should be made.

Although IPA has a predictive power of only 40% it does show richness of data because of both

importance and performance ratings and it is a useful and guiding tool for practioners. When using IPA Bacon (2003) advises to segment the data based on cluster analysis. Eskilden and Kristensen (2006) commented on the fact that the IPA assumes importance and performance to be two independent constructs whereas empirical studies have proven that changes in performance also cause changes in the importance. This puts a constrain on the advise based on the map because making sure performance is heightened might cause a shift not only from under to up but also from the right to the left. The results of the study thus have to be evaluated from this point of view.

5.5 Strategic Resource Management model

Assortment performance will be measured with the help of the Strategic Resource Management model (SRM) of Lush and Serpkinci (1981; in Ring, Tigert and Serpkinci , 2002). The SRM or Trinity model is a management tool for measuring performance on the level of three success factors: Inventory, space and people. The difference with other performance based models lies in the fact that this model especially assesses the efficiency instead of just productivity. This model can be used for an internal benchmark of different stores in a chain to assess how well each unit performs on these different parameters. Also different departments can be contrasted with one another or a comparison between competitors can be made (Ring et al., 2002).

To assess how efficient a store or department performs the first variable to look at is the gross margin return on inventory (GMROI). This variable measures the efficiency of the inventory by multiplying gross margin(%) by the sales to inventory (S/I) ratio (Van der Kind and Quix, 2008). A second important

(30)

variable in this study is the intensity of goods. This is measured by dividing the inventory by the sellable squared floor meters (VVO) (Van der Kind and Quix, 2008) and shows how many goods per square meter are stored. The third variable in this study, the gross margin return on space (GMROS), measures the floor efficiency by dividing the net gross profit by squared meters (Van der Kind and Quix, 2008). In the trinity model the self-service ratio and gross margin return on labor (GMROL) are also of importance. However personnel (FTE) are the basis of these variables and because selexyz cannot assign personnel to each different department these measures are not calculated.

In this thesis inventory efficiency is of importance as it shows if the books in the store are not too old and the assortment is new and attractive. Floor efficiency is of importance to asses which genres are

performing well and should be assigned more space and which ones underperforms and should be shrunken. Finally the intensity of the goods explains if consumers have a need for more space or more goods in the store. Typically stationary articles need lots of goods for a wider choice while consumers prefer more space and thus fewer goods in woman garments (Van der Kind and Quix, 2004). To be able to use this trinity model the squared meters per category are needed. BGN already has inventory and turnover information on each store, however an assessment of the squared meters per category is missing. CAD it used to define the sellable meters in store for the sixteen general bookstores of selexyz. Thereafter these drawings are used when visiting the stores to claim space to each category present in that store. Afterwards CAD is used to compute the total of squared meters per genre. This is then

contrasted with the sales and inventory per genre already known. This results in the assortment needs of customers based on their own behavior. These results will then be contrasted with the results of the survey to be able to make a solid advice to BGN concerning the layout of an assortment in a selexyz store.

5.6 Development of the questionnaire; the interviews at BGN

The questionnaire was designed based on three inputs; earlier questions of BGN research to be able to compare data (ROI), questions based on the interviews (management) and questions based on literature (literature). In appendix B (origin of questions) an explanation is given on the origin of each question. For a better understanding pictures are taken during store visits to show to the respondents.

(31)

Interviews with five employees of BGN were held to act as an input for the development of the

questionnaire, importance performance analysis, segmentation and overall background information on customers and the book market. Semi-structured interviews of approximately one hour each were held with the “Directeur” (Matthijs van der Lely), “operationeel directeur” (Arjen Luyxk), senior marketing manager (Jesse Kröger), “directeur winkels” (Jan-Sander de Bie) and projectleader of the VVO project, measuring store performance (Maarten van den Bovenkamp). The format for the interviews can be found in appendix A (Interviewlist BGN). A short summary of the results is given hereunder.

Special issues

First of all, when discussing the assortment of selexyz a lot of different subjects come to mind. Overall store lay out and the service levels in the stores are considered to be intertwined with the assortment needs. Especially service is seen as one of the most helpful aids a customer can use in the store. The overall store lay out is important in luring customers into the store, helping customers find their way by clear routing and providing a clean and attractive atmosphere for the customers. Media, friends and family influence the decisions made in the bookstores, which are not controllable.

The assortment

A clear picture about what differentiates the assortment of a selexyz store from competitors can be given by stating that selexyz stores offer a broad and in-depth assortment with a wide variety. Originally an academic bookstore, this strength is combined with an offering of diverse themes, bestsellers and the special picks of selexyz. A broad and in-depth assortment is offered by combining a lot of different genres of departments of books together with a lot of titles in that genre. Besides this varied offering, selexyz attracts customers with their own assortment, not only bestsellers but also the more surprising titles and a fitting assortment of non-books like agenda’s and calendars.

Assortment needs

About 74% of customers in retail do not know upfront what they will buy. However uncertain is if this percentage is the same for selexyz stores. If three third of the customers whishes is unknown even to themselves this provides an opportunity for selexyz to influence these whishes and seduce customers. According to the interviews, customer often do have a reading whish, however they do not exactly know what this whish will result in. What customers assortment needs are could be dependent upon gender, education level, shopping goal, distance, frequency of visits and frequency of purchases (yearly and per

(32)

visit). The shopping goal can also be defined in many ways, a business goal, private goal or study related goal. Looking for scientific books, entertainment books or study books. Books related to a profession, as a gift, for relaxation or personal development. When it comes to the different genres selexyz offers (>30) the focal ones are literature, history, travel, children books and cooking. Customers will probably

recognize hobby, science and philosophy also, however this is unknown. The interviewers expect that customers might have difficulty in recognizing the differences between health and medicines,

management and psychology, esotery, foreign books and theology.

Assortment presentation

The shopping window of the stores acts as a mirror of the stores inside. A clear routing, ground plan and information boards help the customer to navigate its way through the store. The store changes twice a month and special theme tables and displays are altered. The books are presented on displays, or on shelves and can be presented with their fronts or standing ‘ruggelings’, the right mix of these three forms of presentation is of main importance here. When it comes to shopping aids selexyz uses top 10’s, top 100’s, themes, information kiosks, posters, ‘bellystrokes’ with written reviews (buikbandjes), a commercial center, register best buys (kassakoopje) and RFID kiosks (in 3 stores). Options which might be used in the future or are available in the market elsewhere are applications with RFID. An RFID kiosks showing tips and offerings to a specific customer, showing reviews of books by others or personnel, showing the customers own books and making recommendations based on historic purchases data. Also mobile applications can be used providing information about product location and stockings. Hand held scanners can be used to show reviews of books. Other options are narrowcasting, an “omroepsysteem” and providing a match with internet in the stores.

Shopping aids

According to the interviewees extra sales because of tips, belly strokes and stickers seem to prove that customers indeed like shopping aids or do use it a lot. However, when it comes to new shopping aids and technological solutions with RFID and mobile phones, ease of use, clarity, communication and attraction of the device might be important. Customers under the age of 35 use the kiosks more often so age might be important, however also shopping goal, education and being technically alert influences the use of shopping aids. The most important aspect remains that a shopping aid must add value for it to be used.

(33)

5.7 Segmentation

Not all customers will share the same meaning when it comes to their assortment needs. People differ and buying situations differ, making it hard to come up with one universal model. To be able to

differentiate between different groups of customers, ex-post and ex-ante segmentation techniques are used. Segmentation can only be used when the following conditions are apparent (van der Kind and Quix, 2008):

 Heterogeneity of needs (different assortment needs)  Measurability of the segment (sampling)

 Approachability of the segment (are we able to reach them)  Sufficient volume (enough respondents)

In this case these conditions are met as the customers of selexyz have different assortment needs, we can actually measure these needs by asking consumers about them, we can reach those consumers through different channels, and in this case an online survey and sufficient volume can be reached because of the large customer base of the selexyz stores.

Ex-post segmentation

On the basis of the literature review and the interviews a view differences in needs are expected: Purchase intention: researchers state that people with a clear goal value assortment variety different than people without a clear goal. In this study we will research whether well defined goals in advance causes differences in perceptions of assortment variety and assortment presentation.

A difference in purchase goals is also expected. Customers buying books for study or business might look at assortment variety and assortment presentation in a different way. Hopefully these customers can be segmented and decisions based on this goal can be made.

Ex-ante segmentation

After gathering the results of the questionnaire different techniques can be used to segment. For example, on the basis of demographic variables, geographic variables, psychographic variables and domain specific variables (Van der Kind and Quix, 2008). In this thesis domain specific segmentation is used to be able to segment on the basis of product category related variables. The attitude towards books and general assortment needs are used as a basis for segmentation. To conduct domain specific segmentation, cluster analysis is used to define different customer groups. Cluster analysis is a tool which aggregates different ‘clusters’ on the basis of same responses. Cluster analysis can present

(34)

different groups of customers. To decide upon how many different clusters are meaningful managerial consent is used.

5.8 Preliminary results

5245 customers responded to the survey request. In total there was a response rate of 7,5% after one mailing. The first results of the questionnaire are presented in appendix C. Important to note is that the results of the questionnaire have been adjusted by incorporating factor scores to the distribution of males versus females and the right age distribution. On average customers are satisfied with their selexyz store and rate them with a 8.05 on a 1-10 point scale.

5.9 Conclusion

An in-depth and wide (scientific) assortment is what defines the selexyz stores, in which browsing of customers and the routing in store are important aspects according to the managers of BGN . Partially based on these interviews, 5245 customers successfully completed the quantitative online survey. The typical selexyz customers is higher educated, spends 10 to 30 minutes in de store and spends more than 250 euro per year. They visit the store monthly to fulfill their readingpleasure and to browse through the assortment of books while travelling more than 25 minutes to reach their store. Importance

performance mapping, the strategic resource model and special segmentation techniques are used to further asses the assortment needs of customers in the next chapter.

The bigger the better? : the role of assortment variety and presentation in redesigning the new physical bookstore