Extending multichannel customer segments : does mobile channel use matter?

(1)

MASTER THESIS

Extending Multichannel Customer Segments

Does Mobile Channel Use Matter?

University of Amsterdam

Faculty of Economics and Business

MSc. in Business Administration

Track: Marketing

Under supervision of: dr. Umut Konus

By:

Student: Ellen Nijboer

Student Number: 10549668

Date: 20

th

_{of January 2015}

_!

!

(2)

University of Amsterdam

_II!

Statement of Originality

This document is written by Student Ellen Nijboer who declares to take full responsibility for

the contents of this document.

I declare that the text and the work presented in this document is original and that no sources

other than those mentioned in the text and its references have been used in creating it. The faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

_III!

Acknowledgement

With the completion of this thesis I have reached the final step in achieving my master’s degree in Marketing at the University of Amsterdam. Writing this thesis has been a very interesting experience. From the first moment of choice I directly knew that multichannel

marketing was the subject that really gained my interest. The specific focus of a multichannel

segmentation research in which I investigated the importance of mobile channels next to the online non-mobile PC/Laptop and in-store channels has even increased my love for this field of study, and helped me during my job interview to become the new junior digital marketer at LINDA. Magazine.

I would like to take this opportunity to thank my supervisor dr. Umut Konus for his multichannel expertise and positive criticism, which enthused and guided me through this important process of writing the master thesis. Furthermore I would like to thank my family and friends for having a lot of patience and faith in me.

I hope you will all enjoy reading this thesis and become interested in this highly important research field that has extremely practical implications.

Kind regards,

Ellen Nijboer

20th of January 2015

(4)

_IV!

Abstract

The multichannel-shopping domain has changed during the last five years. Mobile shopping channels are becoming more popular and are being used by more consumers during both the pre-purchase (information search) and purchase phase. It is expected that this is changing/ enriching the existing customer segments based on online non-mobile PC/Laptop and in-store shopping behaviour, but real evidence has not yet been found. This research therefore investigated: ’Whether and how the inclusion of mobile channels change

segments which are identified on the basis of customer multichannel appropriateness, use behaviour and their covariates’. This research extension is relevant as it identifies the

importance of mobile channels in the multichannel-shopping domain and could therefore help managers to tailor the right strategies (if needed) for different types of multichannel shoppers. Survey data has been collected from 196 customers’ multichannel appropriateness, actual use behaviour and covariates to execute a latent class cluster analysis (LCCA) that 1) identified distinct multichannel customer segments, 2) tested whether covariates profiled the identified segments and 3) explored whether within category (tickets,

clothing and consumer electronics) multichannel segmentation differences existed. Main findings indicated the appearance of a multichannel mobile, multichannel non-mobile and

research shopper segment in terms of customers multichannel appropriateness and a multichannel store, multichannel online and multichannel mobile segment with respect to

their actual channel use behaviour. It has therefore been concluded that mobile channel use matter in the identification of multichannel segments. Interesting, the multichannel mobile segment appeared to be the smallest actual use segment, and has been profiled by innovative 30+ consumers. Youngsters (18-29) appeared to profile the less mobile oriented segments (multichannel- online and store). Looking at the within category identified and profiled multichannel segments, differences did appear, which situated out of category specific characteristics. Aggregate findings controlled for the category specific outliers.

Keywords: Multichannel segmentation; Mobile; Online non-mobile PC/Laptop; Shopping

(5)

_V!

3. Conceptual Framework ... 16 3.1 The Framework ... 16 3.2 Segmentation Indicators... 17 3.3 Active Covariates ... 18 3.3.1 Expected benefits... 19 3.3.2 Psychological traits... 21 3.3.3 Demographical traits ... 24 4. Methodology ... 26 4.1 The Sample ... 26 4.2 Research Design... 27 4.2.1 Survey ... 27 4.3 The Procedure... 28 ! 4.3.1 Pilot study ... 28 4.3.2 Main study ... 28

4.4 Latent Class Cluster Analysis... 29

! 4.4.1 Measures indicator variables... 31

4.4.2 Measures descriptive variables ... 31

5. Results and Analysis ... 33

5.1 Preliminary and Exploratory Analysis... 33

5.1.2 Reliability ... 34

5.2 Segmentation Results ... 37

! 5.2.1 Multichannel appropriateness segments ... 37

5.2.2 Covariates of multichannel appropriateness segments... 40

5.2.3 Multichannel use segments ... 43

5.2.4 Covariates of multichannel use segments... 46

5.3 Comparing Segments... 48

5.4 Category Specific Multichannel Segments... 49

! 5.4.1 Covariates category specific multichannel segments... 51!

6. Discussion and Conclusions ... 53

6.1 Discussion ... 53

6.2 Overall Conclusion ... 59

6.3 Theoretical Implications... 59

6.4 Managerial Implications... 60

6.4 Limitations and Future Research ... 62

7. References... 64

8. Appendices ... 71

(6)

_VI!

Appendix 2. The Spend Cycle... 71

Appendix 3. The Questionnaire... 72

Appendix 4. The Scale Measures ... 76

Appendix 5. Descriptive Statistics ... 77

Appendix 6. Initial Factor Analysis ... 78

Appendix 7. Within Category Segmentation Analysis Results ... 80

Appendix 8. Sample Education Profile ... 82

!

List of Tables

Table 1. Review of multichannel segmentation literature ... 12

Table 2. Segmentation indicators ... 17

Table 3. Reliability analysis ... 36

Table 4. Log-Likelihood statistics for model selection ... 38

Table 5. Profile of multichannel appropriateness segments ... 39

Table 6. Covariates of multichannel appropriateness segments ... 41

Table 7. Log-Likelihood statistics for model selection ... 44

Table 8. Profile of multichannel use segments ... 45

Table 9. Covariates of multichannel use segments ... 47

Table 10. Segment profiles... 49

Table 11. Cross category results ... 50

Table 12. Within category covariates... 52

Table 13. Multichannel Segmentation profiles... 55

List of Figures

Figure 1. Multichannel shopping ... 5

Figure 2. The rise of M-commerce ... 8

Figure 3. Increase of mobile and tablet shipment ... 9

(7)

_1!

1. Introduction

Multichannel shopping, the use of two or more channels in the shopping process, including pre-purchase, purchase and (outside of this research) post-purchasing activities has increased during the last ten years (Neslin et al. 2005, Verde Group & Baker Retailing Center, 2012). Next to the highly researched offline- (in-store) and online shopping channels (PC/Laptop) (Dholakia et al. 2010, Children et al. 2000), mobile shopping channels (Smartphone’s/Tablets) have recently become a hot topic within this multichannel domain (TNS, 2014). Mobile channels portability, wireless network and build-in touch screen offer consumers the ability to search (pre-purchase) and purchase products in an easy and fast way at any time. It therefore, as indicated by Chong (2013) and Shankar et al. (2010), represents an ideal supplementary or substitutionary channel for online non-mobile (PC/Laptop) and offline in-store shopping channels.

Neslin et al. (2005) already indicated that understanding consumer behaviour within the multichannel domain represents a core challenge for firms. The continual and rapid growth of mobile channels for shopping makes it necessary to gain a further understanding

of consumer’s multichannel behaviour. Specifically, whether and how mobile channels are used for shopping next to the already researched online non-mobile PC/Laptop and in-store

channels during both the pre-purchase and purchase phase (Dholakia et al. 2010). The execution of a segmentation study based on consumer behaviour is therefore considered

highly important. Firstly, because it will give an indication to what extent these mobile channels actually matter for consumer multichannel-shopping behaviour. Secondly it may have crucial managerial implications as it could help marketers to design effective multichannel customer strategies (Neslin et al. 2005, Konus, Neslin and Verhoef, 2008). If, for example, varying multichannel segments show different mobile usage patterns next to the use of online non-mobile PC/Laptop and in-store channels, it suggests firms to tailor different strategies for these segments. If not, no varying strategies would be necessary. In case varying strategies are needed, marketers should understand the characteristics of these customer segments. Covariates (e.g. personal motivations/ traits) could help to understand

(8)

_2!

why, for example, ‘segment A’ focuses on mobile channels during the pre-purchase phase, whereas they consider online non-mobile PC/laptop channels for purchase (Ailawadi, Neslin and Gedenk, 2001). Knowing and understanding multichannel shopping segments for multiple industry categories (different categories could influence segment composition) could motivate managers to invest in an effective multichannel mix, incorporating specific segment-oriented strategies. This enhances customer value and increases firm’s profit potential (Bhatnagar and Ghose, 2004).

Multiple past segmentation studies have looked into consumers multichannel shopping behaviour and covariates to identify and profile segments (Konus, Verhoef and

Neslin, 2008, P. 399). For representative findings these studies incorporated different categories. Keen et al. (2004) for example identified segments based on consumer channel preference and profiled them with demographic covariates. Konus, Verhoef and Neslin (2008) furthermore used channel appropriateness to segment consumers, in which they considered multiple covariates to profile the multichannel segments respectively for seven distinct categories. Even though these segmentation studies have contributed to the understanding of multichannel shopping behaviour, fact is that most of these studies looked into online non-mobile PC/Laptop and offline in-store multichannel shopping behaviour. Little is known about the inclusion of today’s highly popular mobile channels. Questions whether and how consumers use mobile channels and how this may change customer segments identified on the basis of their multichannel behaviour have not yet been answered. A consequence of this phenomenon is that very few leading retailers/ managers have paid serious attention to their mobile presence yet and (if present) how they could meet their mobile multichannel customer needs. Researches that have been conducted within the mobile channel field have mainly looked into the technological acceptance side of m-commerce (Chong, 2013) or looked into the consumer personal characteristics or motivations whether to adopt mobile shopping or not (Chen & Chou, 2011, Yang & Kim,

2012, Okazaki & Mendez, 2013, Chong, 2013). Segmentation studies concerning consumer’s appropriateness and/or actual use of mobile channels during the pre-purchase

(9)

_3!

and purchase phase next to the online non-mobile PC/Laptop and in-store channels to identify multichannel segments remain sparse for all shopping categories. This in combination with profiling them by underlying expected benefits and traits (covariates) has not been researched before and therefore represents a gap in the existing multichannel segmentation literature.

The purpose of this research is therefore to fill the identified gap by investigating: ‘Whether and how the inclusion of mobile channels change segments which are identified on

the basis of customer multichannel appropriateness, use behaviour and their covariates’.

Accordingly, the main research objectives are:

1. To identify segments on the basis of customer multichannel appropriateness and actual use of mobile, online non-mobile PC/Laptop and in-store channels during the pre-purchase and purchase phase to identify if mobile channel use matter. 2. To explain and profile the identified segments based on the following covariates:

expected benefits, psychological and demographical traits.

3. To identify and explore how multichannel customer segmentation might differ considering three different categories (tickets, clothing and consumer electronics). Execution of these objectives will result in both theoretical and managerial contributions. Firstly, this research tries to extent consumer behaviour research within the multichannel domain by trying to identify and understand a mobile-oriented multichannel-shopping segment (Neslin et al. 2005). Secondly, if multichannel mobile segments are identified, underlying covariates might profile these shoppers. This will provide theoretical knowledge on mobile shoppers. And thirdly, the inclusion of mobile channels enriches the theoretical framework identified by Konus, Neslin and verhoef (2008) for motivating the existence of segments and covariates that might profile identified segments. Furthermore this research is of managerial relevance as it identifies the importance of mobile channels by considering consumer multichannel behaviour. If considered important it could motivate

managers to invest in an efficient m-shopping tool. Beside, if more- and less mobile oriented multichannel segments are identified and profiled, managers should be able to tailor specific

(10)

_4!

strategies for each segment in order to attract new, keep the existing, and grow their most loyal customer base.

In order to achieve these contributions, the research is structured as follows. First the multichannel (segmentation) literature will be reviewed. Secondly, a visualization of the conceptual framework will be given, followed by a detailed explanation of segmentation indicators and hypothesis on covariates. Thirdly, the sample and methodology used will be explained followed a detailed chapter on the segmentation results. And finally, the research ends with a detailed discussion and final conclusion on findings followed by theoretical implications, managerial implications, limitations and future research suggestions.

(11)

_5!

2. Literature Review

This chapter provides a detailed theoretical background for the execution of this multichannel segmentation study incorporating mobile channels. Firstly, it provides an understanding of the multichannel-shopping domain in general. Next it explains the appearance of mobile channels for shopping within the multichannel domain. And lastly the multichannel segmentation literature has been reviewed followed by an identification of the research gap and development of the main research question.

2.1. Multichannel Shopping

The single channel environment, the search and purchase of products and services within just one channel, is considered an unprofitable

business strategy (Hsiao, Yen and Li, 2011). Today the Internet, tablets, smart phones, call centres, home shopping networks, catalogues and physical stores are common channels by which consumers

search for information and shop. If two or more of these channels are considered during the process of

buying a product the phenomenon is called multichannel shopping (figure 1). (Neslin et al. 2005, Yang & Kim, 2012). Although the phenomenon multichannel shopping has been

around for years, it was the introduction of the Internet that firstly grasped the marketer and consumer’s attention towards the multichannel environment (Schoenbachler and Gordon, 2002). Reasons as identified by Schoenbachler and Gordon (2002) were the increased information access about products, services, inventories, prices and competitive offerings. No longer were consumers dependent on the prices and rules set by the retailers (Loewe and Boncheck, 1999) but competitiveness between (in that time) catalogues, stores and online non-mobile PC/Laptop channels made them able to search and buy from the ‘consumer’s perspective’ most valuable channels. Economic goals, switching costs, efficiency concerns, risk aversion and geo-demographic characteristics have been identified Figure 1. Multichannel shopping (Head, 2014)

(12)

_6!

as reasons why consumers choose multiple channels to satisfy their changed needs (Dholakia, 2010, Balasubramanian et al. 2005, Frambach, Roet & Krishnan, 2007, Johnson, 2008, Inman, Shankar and Ferraro, 2004).

The proliferation of channels due to the introduction of the Internet created challenges and opportunities for firms and academics. Challenges, in understanding consumers’ changed drivers and preferences for multichannel use and opportunities to effectively anticipate on these changes by adding new shopping channels to serve customers better. Not anticipating will result in loosing customers somewhere in the shopping process. This shopping process, from the economist’s perspective (Court et al. 2009), could be explained

as the pre-purchase (information gathering phase), purchase (transaction) and post purchase phase (evaluation). Academic researchers proposed additional shopping phases. Choudhury and Karahanna (2008) for example suggested four stages by splitting the pre-purchase phase into requirements determinants and vendor selection. Schoder and Zaharia (2008), Konus, Verhoef and Neslin (2008) and Kollman, Kuckertz and Kayser (2012) on the other hand differentiated between just two channel selection stages; namely the information search prior to a purchase and the actual purchase. The two phases also considered within the context of this research.

As identified by Konus, Verhoef and Neslin (2008) and Balasubramanian, Raghunathan and Mahajan (2005) the channel preferences during different phases could be different per consumer. Verhoef, Neslin and Vroomer (2007) for example identified a significant group of consumers that search for products on the Internet and then switch to the physical store (from same retailer or competitor) to purchase. Neslin et al. (2005) on the other hand identified that customers could use multiple channels during search phase, but just one during the purchase phase, referred to as a research shopper. A customer channel migration model by Ansari, Mela, and Neslin (2008) concluded that many consumers appear to prefer the use of a single offline channel for purchase and that their migration to newer

(13)

_7!

the first place mainly functioned as an information search tool (pre-purchase phase) (Dhokalia et al. 2010 p. 88).

Neslin et al. (2005) indicated in his research that the identification of reasons for the increased use of multiple channels during different phases is a crucial aspect of this process. A customer may, for example, prefer the Internet for search because it is easy to use. Schoenbachler and Gorden (2002) therefore investigated the drivers of consumer’s channel choice considering offline (in-store and catalogue) and online non-mobile PC/Laptop channels, which turned out to be perceived risk, past direct marketing experience, motivation to buy from a channel and the specific product category. Another research by Farag et al.

(2005) looked more specifically into consumer’s preferences in shopping online non-mobile or offline (in-store or catalogue). It appeared that consumers shopping behaviour, shopping attitudes, Internet behaviour and lifestyle/personality indicators had an enormous influence on their preference. A later conducted research by Dholakia et al. (2010) agreed on these factors, but extended them with variables such as social influences, ability for co-creation, ease of processing and self-efficiency.

As past research has contributed significantly to the understanding of consumers channel choice within the multichannel domain (considering offline and online non-mobile channels) a new interesting channel that needs additional investigation are mobile channels (Smartphone’s and Tablets). Internet trends research by Mary Meeker (2014) indicated that global Smartphone users as percentage of mobile phone users increased exponentially between 2009 and 2013. Same research indicated a fast growing tablet unit shipment of 52 per cent as the age of technology has now enables customers and retailers to do business and consume via mobile phones and tablets. Research by Shankar, Venkatesh, Hofacker and Naik (2010) already indicated that mobile devices are becoming ubiquitous. One of the challenges therefore is to understand ‘whether’ and ‘how’ the inclusion of mobile channels is going to change consumers multichannel-shopping behaviour during the shopping process.

(14)

_8!

2.2 Mobile Shopping

Mobile shopping (M-shopping), according to Kourouthanasis and Giaglis (2012) is Internet commerce (e-commerce) conducted over mobile or wireless networks while using mobile devices. There are many mobile devices (Mobile Phones, Smartphone’s, Tablets, E-readers), but the two highly preferred for shopping, as indicated by Viswanathan (2014), are Smartphone’s and Tablets due to their conveniences of a personal computer, along with the fact that they are handheld devices. According to Shankar and Balasubramanian (2009) and Sharma (2011) the real intelligence and functionality of these two mobile devices lie in its software. Due to it’s wireless connection, location-sensitivity, portability, touch screen and

easy navigation format, Smartphone’s and tablet devices enable consumers to gather information about products and services, compare offerings, enables consumers to directly purchase from their devices and get after-sales services wherever and whenever they want. Furthermore as indicated by Shankar, Venkatesh, Hofacker and Naik (2010) due to mobiles personal nature it could be seen as a cultural object as well as it has become part of consumers daily routines and practices. This has therefore resulted in a so called ‘mobile lifestyle’, in which consumers more and more use mobile channels for multiple activities1_in

which shopping has become more and more of interest.

There are, according to Saylo (2012) two forms of m-commerce, 1) m-internet, shopping via the open-ended source

(e.g. browser) and 2) m-apps, shopping within mobile applications (e.g. closed-ended). A mobile life survey by TNS (2012) indicated that mobile commerce is on the rise globally. A Morgan Stanley research

by Murphy and Meeker (2011) (Figure 2) already revealed that m-commerce platforms are

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

"

_!

_{For more information about the mobile domain look into Shankar et al (2010)}

!

(15)

_9!

becoming more and more of interest when comparing towards desktop Internet commerce. Same research conducted in 2014 indicated that mobile and tablet shipment increased exponentially from 2009

compared to all other devices (figure 3). Forester Research (2011) already expected a mobile shopping growth of 39 per cent between 2011 and 2016. As identified by

(Sharma, 2011) fact that

Smartphone’s and Tablets are more and more used in shopping is due to their ability to make the spend cycle, identified by Philip Kotler, easier (Appendix 2). Instead of spending time doing physical research, customers might ask their Smartphone’s and Tablets (in app applications) about products & services. As stated by Sharma (2011, P9): ‘By taking out the

steps between information search and transaction, mobile offers convenience, efficiency and accountability that only gets better with time and use’. Murphy and Meeker (2011) and

Sultan, Rohm and Gao (2009) indicated that tools such as real time location-based services (deals & offers in your area), transparent pricing (comparison online and local store prices), discounted offers (in-app tools) and immediate gratification (virtual goods; music iTunes) are both m-internet and m-apps tools that are changing the way in which consumers search for information and shop.

According to research by Okazaki and Mendez (2013) and Yang & Kim (2012) mainly due to the introduction of the built-in touch screen keyboards (introduced by iPhone in 2007), wireless mobile network and multi functional tools, consumers can use their smart phones and tablets more easily in information search and purchase processes wherever and whenever they want. It therefore has interesting multichannel shopping opportunities. Results

from Comscore-Milliannial media retail study (2011) has shown that over 52 per cent of consumers are using mobile devices across different categories such as electronics,

(16)

_10!

clothing, food, entertainment, travel and books. A research by Okazaki and Mendez (2013) furthermore indicated that travel search on a mobile device increased by 1200 per cent. Same millennial media research (2011) indicated that almost 40 per cent already purchased via mobile devices. Research by Shankar, Venkatesh, Hofacker and Naik (2010), on the other hand, argued that only 29 percent of retail shoppers in possession of a mobile phone actually used their mobile phone as part of the shopping experience. 72 per cent of those who used mobile phones used it mainly for pre-purchase purposes such as comparing prices and reading reviews. Reasons as indicated by Shankar et al. (2010) were that even though mobile channels offer high-tech new shopping tools, challenges situate in that some consider

the touch screen and mobile’s small screen a limitation of usage.

So to date different opinions about mobile shopping situate and the identification of a real pattern in mobile shopping during the pre-purchase and purchase phase has not yet been found. What Sultan, Rohm and Gao (2009) Shankar et al. (2010) and Yang and Kim (2012) did find were enabler and inhibitor factors and multi-dimensional mobile shopping motivations that indicate why consumers accept or reject mobile channels. Factors such as the TAM model (technology acceptance model), economic barriers, trust and risk factors and gratification motivation theories on how mobile channels are personally used for both utilitarian and hedonic purposes. Namely: idea (browsing product information), efficiency (quick buying and saving time), adventure (exploring) and gratification (personalized) shopping motivations.

Even though our limited understanding of consumers mobile channels use during the pre-purchase and purchase phase, fact is that the multichannel landscape is changing due to its appearance. It is therefore expected in Yang’s (2010) research that the numerous advantages of mobile shopping devices (e.g. ubiquity, mobility, convenience, personalization, flexibility and quick information) will have an impact on the use of other channels. Previously unmet needs could now be met. So are mobile shoppers no longer bound by geographical

constraints as compared to offline in-store and online non-mobile PC/Laptop channels, which could result in a more favourable attitude towards mobile shopping channels (Chong, 2013).

(17)

_11!

Given that mobile phone subscribers are overtaking the number of Internet users in some countries, it becomes interesting to research ‘whether’ we can find and identify customer segments, which are characterized by the use of mobile channels. ‘How’ this changes the consumer multichannel segments, and lastly ‘what’ are the characteristics/ covariates of these mobile shoppers compared to those that do not shop within the mobile channel domain. Questions that this study tries to answer with the execution of a segmentation study.

2.3 Multichannel Customer Segmentation

As has already been questioned by Neslin et al. (2005 P.97): ‘is a multichannel

approach a means to segment customers? That is, are there distinct segments of customers who use various channels and combinations of channels?’ A highly interesting question as it

helps managers in defining the target markets for their channels in use during the pre-purchase and pre-purchase phase and enables them to develop appropriate strategies for all identified segments. Berry (1999) indicated that customer segmentation is crucial for the success of commerce on channels. He stated that next to multichannel use information, is it highly vital to understand the need and benefits that customers desire in commerce, in order to profile identified segments. And lastly how customers are segmented with respect to their multichannel use appears to differ per category. Knowledge on this diagnostic information, as indicated by Bhatnagar and Ghose (2004) should help firms make better decisions about their multichannel marketing strategy.

Looking into past segmentation research, findings on consumer’s use, preferences and intentions to use certain channels during the pre-purchase and purchase phase in combination with their motivational drivers, benefits and personal characteristics (covariates) made it able to segment and profile customers with same multichannel usage behaviours (Falk, 1997; Alba et al. 1997; Ghosh, 1998, Dholakia et al. 2010; Children et al. 2001 and To, Liao and Lin, 2007 and Konus, Verhoef and Neslin, 2008). Highly interesting multichannel

(18)

_12!

to segment consumers in combination with covariates to profile the identified segments have been listed in table 1.

Table 1. Review of multichannel segmentation literature

Covariates Multichannel use Empirical/ theoretical Motivati ons Benefits Psycho graphics Demo graphics Multiple phases Considers mobile

Keen et al. (2004) Online, Store Empirical Rohm & Swaminathan

(2004)

Online Empirical ! ! !

Bhatnagar and Ghose (2004a)

Online Empirical ! ! !

Kumar and Venkatesan, 2005

Store, Mail, Online, Telephone

Empirical ! ! !

Thomas and Sullivan, 2005 Store, Catalogue Internet Theoretical ! ! ! ! Balasubramanian et al. (2007)

Online, Store Theoretical ! ! !

Konus, Verhoef and Neslin (2008)

Store, Internet, Catalogue

Empirical ! ! ! !

Schoder and Zaharia (2008)

Store, Online, Catalogue

Empirical ! !

Aldas-Manzan (2008) Mobile Empirical ! ! ! !

Hillman et al. (2012) Mobile Theoretical ! ! !

Yang & Kim (2012) Mobile Empirical ! ! !

San-Martin, Lopez-Catalan and Ramon-Jeronimo (2013)

Mobile Empirical ! ! !

This research Mobile, Online, Offline

Empirical ! ! ! ! !

(19)

_13!

Keen et al. (2004) was one of the first who executed a multichannel segmentation research on multichannel shopping behaviour and identified four customer segments based on their preference for e-tailing and retailing (retail/ catalogue). Namely, Generalist (e.g. who care about all channels), Formatters (e.g. who have channel preferences), Price-sensitive’s (e.g. who select channels according to price), and Experiencers (e.g. who tend to use the same channel continuously). In their research they considered two different categories and profiled the identified segments with psychological characteristics. Thomas and Sullivan (2005) furthermore identified and delivered on theory how to segment groups based on store, catalogue and Internet shoppers. They used a year of behavioural data (e.g. channel use)

and independent measures (e.g. demographics) which enabled them to segment and profile customers into various single and multichannel groups. Namely; store-only shoppers, catalogue only shoppers, internet only shoppers, dual channel shoppers and a very small segment of customers who shopped in all three (e.g. triple-channel customers). Konus, Verhoef and Neslin (2008) furthermore proposed three multichannel segments based on consumers appropriateness level of using Internet, catalogue and in-store channels for shopping. Namely; multichannel enthusiasts (e.g. positive attitudes towards all channels), store focused consumer (e.g. store oriented) and lastly uninvolved shoppers (e.g. little interest in any channel – low involved shoppers). An important finding with respect to their inclusion of different categories indicated that multi-channel usage behaviour appears to be small for the clothing category, whereas it appears to be big for electronics. Within their research, Konus, Verhoef and Neslin (2008) have shown that differences in consumer psychographics and expected benefits resulted in varying consumer perceptions towards what channels to use during different phases (pre-purchase and purchase) of the shopping process. A segmentation framework obtaining different multichannel shopping segments profiled on motivational benefits were the result.

More specifically, looking into researches that incorporated mobile channels within

their segmentation study remain limited. One highly interesting segmentation research that did look into the mobile channel use, but not specifically on mobile shopping was a research

(20)

_14!

by Junco and Mastrodicase (2007). They identified three segments, namely millennials (10 to 25 year old who quickly adapt mobile innovations), road warriors (who adopt mobile channels to gain more control over their fast-paced and often stressful) and concerned parents (use mobile channels to stay in touch through voice ant text). San-Martin, Lopez-Catalan and Remon-Jernimo’s (2013) research furthermore identified distinct mobile shopper profiles by exploring their TAM (technology acceptance model) drivers and impediments such as perceived ease of use, perceived usefulness and problems using the mobile phone in combination with demographic traits. Others by Yang and Kim (2012) and Hillman et al. (2012) looked into mobile shopping motivations/ behaviours in which Hillman specifically

focused on trust as a determinant of mobile shoppers. And lastly, Aldas Manzano (2008) looked into the individual personality factors of mobile shoppers such as innovativeness, compatibility and mobile affinity. Limitation of these studies is their specific focus on the identification and profiling of a mobile segment, whereas placing their mobile behaviours within the context of their use of offline in-store and online non-mobile PC/ Laptop channels would have been of high value.

The above stated findings have contributed concisely to the multichannel segmentation literature, and have extreme managerial implications. As has been stated by Thomas and Sullivan (2005 p.33): ’The better identification of your segments the better

communication strategies can be developed’. But a limitation is the fact that the identified

multichannel segmentation studies focussed mainly on offline in-store and online non-mobile PC/Laptop channels for the identification of multichannel segments. They did not incorporate mobile channels. The studies that did look into mobile channels, did not consider them within the context of multichannel use, but only look at it separately (single channel). This enabled them to identify a mobile segment, but more relevant would have been to identify segments in the context of consumer’s multichannel use during the pre-purchase and purchase phase considering mobile, online non-mobile PC/Laptop and offline in-store channels.

This research therefore explores whether we can identify multichannel-shopping segments by including mobile channels (Smartphone, Tablet), next to the (already

(21)

_15!

researched) online non-mobile PC/laptop and offline (in-store) channels while controlling for differences among categories. To identify segments within this ‘extended’ multichannel research it has been chosen to look into both consumers’ appropriateness (attitude) and their actual use (behaviour) of the three channels during both the pre-purchase and purchase phase. Furthermore to profile and explain the characteristics of the identified segments, active covariates have been used (explained in my conceptual framework). Therefore the aim of this research is; ‘Whether and how the inclusion of mobile channels change segments

which are identified on the basis of customer multichannel appropriateness, use behaviour and their covariates?’

(22)

_16!

3. Conceptual Framework

This chapter visualizes and explains the conceptual framework, describes the segmentation indicators used to identify multichannel customer segments and explains the covariates that are hypothesized to profile the segment identified.

Figure 4. Conceptual framework

3.1 The Framework

Figure 4 provides an overview of the conceptual framework for this study. The overall purpose is to identify distinct multichannel segments based on consumer’s appropriateness and use of mobile, online non-mobile PC/Laptop and in-store channels during both the pre-purchase and pre-purchase phase. As explained in the literature review, due to the extensive and increasing use of mobile channels it is expected to identify segments that are characterized by mobile channels. It is therefore expected to identify a multichannel mobile segment, one in which all channels, but especially mobile channels are used during both the pre-purchase and purchase phase. I may have a segment that searches (pre-purchase) mobile but does not purchase mobile. Furthermore I could identify a segment that only uses mobile channels, or lastly one that does not use any mobile channels at all. There are many

(23)

_17!

possibilities, which have been explored considering respondents channel appropriateness and actual use during the pre-purchase and purchase phase and hopefully new segment compositions could be identified. To increase the explanation power of the segments, active covariates, as shown in figure 4, have been included in the framework as well, which could give a final profile to the identified segments.

3.2. Segmentation Indicators

As it is stated by Hair et al. (2011) to identify different segments, one must divide the total sample into smaller subgroups. Thomas and Sullivan (2005) referred in their research to the use of basic descriptors ‘actual consumer observations’ to identify segments. In the context of this study, we therefore use the ‘indicators’ channel appropriateness and actual use to identify distinct multichannel segments. In specific, consumers attitude about the ’appropriateness’ of mobile (Smartphone’s, Tablets), online non-mobile (PC’s, Laptops) and in-store channels for shopping during the pre-purchase and purchase phase and their ‘actual use’ of mobile (Smartphone’s, Tablets), online non-mobile (PC, Laptops) and in-store shopping channels during the pre-purchase and purchase phase. According to Dholakia et al. (2010) several studies have used consumers channel appropriateness (Konus, Neslin and Verhoef, 2008) or actual use (Thomas and Sullivan, 2005) as the basis for their multichannel segmentation and have therefore proved the effectiveness of these variables for segmentation purposes. Derived from these earlier segmentation studies, this research tries to identify segments both for consumer’s appropriateness and actual use of multiple

channels. Table 2 provides a short explanation on the segmentation indicators considered. Table 2. Segmentation indicators

Indicators Explanation

Appropriateness Consumer’s attitude of mobile, online non-mobile PC/Laptop and in-store channels during the pre-purchase and pre-purchase phase.

Use Consumer’s actual usage behaviour of mobile, online

non-mobile PC/Laptop and store channels during the pre-purchase and purchase phase.

(24)

_18!

3.3. Active Covariates

In order to profile and understand the identified segments, the conceptual model furthermore proposes to research consumer’s active covariates. In short, active covariates could be explained as consumer’s personal traits, motivations (etc.) that impact their channel choices for shopping and therefore have a final impact on segment membership as they influence the classification probabilities and explain/profile the different identified segments (Vermunt and Magidson, 2002). If all consumers within a certain segment are significantly profiled by a covariate, it could give managers important strategic information about the customers within that segment.

I borrowed from Ailawadi, Neslin and Gedenk (2001), Konus, Verhoef and Neslin (2008) and Heitz-spahn (2013) aspects of their used covariates to profile and understand the segments identified within this particular study. Ailawadi, Neslin and Gedenk (2001) and Konus, Verhoef and Neslin (2008), for example, state that consumers choice of certain channel(s) during the pre-purchase or purchase process (in-store or out of store promotions in the case of Ailawadi Neslin and Gedenk, 2001) depends on the utility they derive from the benefits and costs of search and purchase, which in turn may be influenced by customer characteristics (e.g. psychological and demographical traits). In this way particular groups of consumers with same traits will be attracted to particular costs and benefits. Heitz-Spahn (2013) also used this principle, but only focused on the channel choices due to consumer’s maximization of their utilities and benefits. As indicated by Ailawadi, Neslin and Gedenk (2001) this typology has a well-established tradition in the literature on shopping behaviours, and it has therefore also been considered in the context of this research, whereas this research is slightly different as it focuses on the benefits (not costs), that are derived from multichannel appropriateness/ use considering the different identified segments and segments specific consumer psychological and demographic characteristics, but does not investigate the link between those benefits and traits.

Findings from past research reports on multichannel use (online non-mobile and in-store) and mobile channel acceptance/ use helped to determine which expected benefits,

(25)

_19!

psychological and demographical traits have shown to be of key influence to profile segments. The most interesting findings have briefly been explained in the following paragraphs in order to make assumptions about their possibility to characterize the segments identified in this extended multichannel segmentation research.

3.3.1. Expected benefits

In order to understand and profile the identification of different multichannel shopping segments, previous research suggest looking at expected channel benefits. Heitz-Spahn (2013) suggest that consumers select channels at each stage of the decision process to fulfil utilitarian and hedonic needs at the lowest costs relative to benefits. These utilitarian and

hedonic needs are derived from Tauber (1972). He was the first to consider the question ‘why do consumers shop?’ Next to simply making a purchase out of need (instrumental/ financial related benefit), later revered to as utilitarian shopping motivations, Tauber (1972) reasoned that some motivations are considered unrelated to the actual purchase, but with satisfaction of shopping activities (fun and exploration related benefit). Later referred to as hedonic shopping motivations by Holbrook and Hirschman (1982). In the context of this research four utilitarian and hedonic related benefits that could profile/ explain our identified multichannel segments are:

Convenience. As indicated by Schroder and Zaharia (2008) and Heitz-Spahn (2013) convenience is a frequently studied motive for shopping within the multichannel domain. The actual selection of the channels in use is based on time and/ or psychical and mental effort savings. Bellenger and Korgaonkar (1980) looked into which physical store to select considering convenience. Rohm and Swaminathan (2004) and Schroder and Zaharia (2008), on the other hand suggested the importance of convenience in online non-mobile channel choices due to the fact that location becomes irrelevant. Especially during the pre-purchase phase, convenience in searching on personal computers has appeared to be effective due to goal directed shopping and to ability to switch between suppliers (Children et al. 2000).

Chong (2013) and Yang and Kim (2012) furthermore indicates that mobile channels offer even more advantages in convenience shopping as it allows users to be free from physical

(26)

_20!

requirements of wired connections in exploring information and buying with a couple of touches. Consistent with these research findings I suspect that convenience will have a significant effect on the segment membership based on multichannel behaviour.

Price savings. Falk (1997) stated that economic value for money was one of the most important factors in determining which channels to use during different stages of the shopping process. With other words, ‘where can I get the best price’? In a multichannel setting, Balasubramanian et al. (2005) and Konus, Verhoef and Neslin (2008) assert that one of the key advantages of multichannel shopping behaviour is finding good deals by recognizing attractive offers and compare prices across channels. According to

Balasubramanian et al. (2005), information-search costs are generally the lowest in online non-mobile channels. Yang (2010), furthermore indicated that mobile shopping channels offer even lower information-search costs as it enables users to compare product prices and obtain relevant (discount) deal information at user pinpoint locations (Murphy and Meeker, 2011). I therefore suspect the benefit price savings to have a significant effect on the segment membership based on multichannel behaviour.

Entertainment. As stated by Schroder and Zaharia (2008) and Rohm and Swaminathan (2004) the entertainment benefit within the multichannel domain stands for the emotional (sensory stimulus) and social (interaction) need for an enjoyable shopping experience independent of product specific or task-directed objectives. To, Liao and Lin (2007) and Dholakia et al. (2010) indicated that especially offline channels entail entertainment benefits due to service perception, sensory stimulus and social interaction. Still, Yang & Kim (2012) and Chong (2013), on the other hand indicate that entertainment could be derived from online non-mobile and mobile channels if consumers find them entertaining. Kim, Foire and Lee (2006) and Farag et al. (2005) identified that specifically searching in online non-mobile shopping channels delivers on the benefit of entertainment due to quick multiple website access, and Yang and Kim (2012) indicated that the benefits;

adventure (shopping for excitement), idea (collecting information about new trends) and gratification (special treat) shopping situated in the pre-purchase phase of mobile channels. I

(27)

_21!

therefore suspect entertainment to have a significant effect on segment membership based on multichannel behaviour.

Exploration. As indicated by Konus, Verhoef and Neslin (2008) the benefit of exploration involves finding new brands and new experiences and within the context of multi-channel shopping it could be stated that this benefit indicates switching multi-channels to gain access to a broader assortment and experience channel benefits during different stages of the shopping process (Heitz-Spahn, 2013). The benefit of exploration indicates that consumers are wiling to explore a new retail store in their neighbourhood, explore the internet for new web shops and with the emerging mobile shopping channels it could be seen

as an interesting play field for exploring new shopping services and features (Yang and Kim, 2012). I therefore suspect the benefit exploration to have an effect on segment membership based on multichannel behaviour. In sum;

H1a. Convenience (expected benefit) is significantly associated with segment membership based on multichannel behaviour.

H1b. Price savings (expected benefit) is significantly associated with segment membership based on multichannel behaviour.

H1c. Entertainment (expected benefit) is significantly associated with segment membership based on multichannel behaviour.

H1d. Exploration (expected benefit) is significantly associated with segment membership based on multichannel behaviour.

3.3.2. Psychological traits

Ailawadi, Neslin and Gedenk (2001), Konus, Verhoef and Neslin (2008) as well as Heitz-spahn (2013) have indicated in their research the link that situates between consumers expected benefits that situate in channels and certain consumer psychological traits. For example, a consumers channel benefit of savings suggests a consumer with a characteristic of high price consciousness (Konus, Verhoef and Neslin, 2008). In this section I will elaborate on four chosen psychological traits derived from prior research that might profile

(28)

_22!

the segments identified within the extended multichannel domain. The identified psychological traits considered are:

Time pressure. Time pressure indicates the degree to which a consumer feels that there is not enough time available for performing a specific task (Bruner 2009 p. 941). It is an exogenous variable capable of influencing consumer behaviour (Howard and Sheth, 1969). Lee et al. (2009) indicated that customers seek goods and services in ways that can save time. As has already been identified by Xu-priour, Cliquet and Fu (2012) high time pressured consumers have a favourable attitude towards online channels compared to offline channels for shopping due to convenience (Fisman et al. 2006). The Mobile shopping domain allows

consumers to use the Internet anywhere and anytime free from physical requirements of wired connections which could be even more favourable for high time-pressured consumers (Yang and Kim, 2012). I therefore suspect time pressure to have a significant effect on segment membership based on multichannel behaviour.

Price consciousness. As stated by Lichtenstein, Netemeyer and Burton (1990), price consciousness is about consumers concern to pay low prices, or as defined by Ailawadi, Neslin and Gedenk (2001) it is about the degree to which consumers find product price important. Research by Elliot, Fu and Speck (2012) indicated that most studies identifying multichannel segments have shown consumers price driven and cost savings drive. These consumers therefore maximize searching across channels to find the best purchase price. As identified by Konus, Verhoef and Neslin (2008) the Internet provides consumers with quick information and benchmark opportunities at a low cost to attain price comparison. Furthermore, Yang (2010) indicates that mobile shopping channels make it even able to get relevant promotion and price information in consumers hands anywhere and anytime. It is therefore that I suspect price consciousness to have an effect on segment membership based on multichannel behaviour.

Shopping enjoyment. Shopping enjoyment is the tendency of a consumer to derive pleasure and fun from shopping (Bruner, 2009). This pleasure and joy could be gained through the sensation stimulus and social setting in store (shopping with friends) or idea

(29)

_23!

generating shopping opportunities that situate in offline, online non-mobile and mobile channels (Konus, Verhoef and Neslin, 2008, Yang and Kim, 2012). Verhoef et al. (2007) states that in a multichannel domain, shopping enjoyment influences channel selection and Ailawadi Neslin and Gedenk (2001) and Konus, Verhoef and Neslin (2008) have indicated that consumers who enjoy shopping have found to be heavier users of multiple channels because those that really enjoy shopping, shop to try new things (purchase) and get new ideas anywhere and anytime (pre-purchase). With this knowledge I expect shopping enjoyment to have a significant effect on segment membership based on multichannel behaviour.

Innovativeness. Hirschman’s (1980) research quotes from Midgley and Dowling (1978, p236): ‘Innovativeness is the degree to which an individual is receptive to new ideas

and makes innovation decisions independently of the communicated experience by others’.

With other words, Innovative consumers are consumers that are able to adopt a new technology in a relatively early stage far before average others (Aldas-Manzano, Ruiz-Mafe & Sanz-Blas, 2008). Finding by Konus, Neslin and Verhoef (2008) have shown that multi-channel enthusiast tend to be more innovative and explore new multi-channel alternatives. They maximize their search through several channels, which results in the best experience channel to purchase. Eastlick and Lotz (1999) characterised innovators as heavy users of interactive electronic shopping media. Yang (2012) indicated in his research that consumers with higher technology innovativeness are more likely to adopt mobile shopping. I therefore suspect innovativeness to have an effect on segment membership based on multichannel behaviour. In sum;

H2a. Time pressure (psychological trait) is significantly associated with segment membership based on multichannel behaviour.

H2b. Price consciousness (psychological trait) is significantly associated with segment membership based on multichannel behaviour.

H2c. Shopping enjoyment (psychological trait) is significantly associated with segment membership based on multichannel behaviour.

(30)

_24!

H2d. Innovativeness (psychological trait) is significantly associated with segment membership based on multichannel behaviour.

3.3.3. Demographical traits

Next to the psychological traits, demographics could also profile segments. It is, for example, stated that youngsters are more likely to engage within the online channel domain then the elderly (Farag et al. 2005). In this last section I elaborate on five demographical traits and how they might profile the channel segments identified, namely;

Age. Miller (1996) already stated in his research that the use of age in the identification of segments is meaningful. But when specifically referring toward the

multi-channel domain Heitz-Spahn (2013) and Konus, Verhoef and Neslin (2008) indicated no explanatory power of age within their identified segments. Bhatnagar & Ghose (2004), Farag et al. (2005) and Liebermann and Stashevsky (2002) on the other hand have shown that that the cyberspace is the domain of young consumers and Chong (2013) indicated that younger’s engaged more in mobile pre-purchase and purchase shopping than older users. It is therefore that I suspect age to have a possible effect on segment membership based on multichannel behavior.

Gender. Contradictions situate within past research findings in identifying a relationship between multichannel segmentation and gender. Kaufman-Scarborough and Lindquist (2002) and Farag et al. (2005) for example stated that men are more likely to engage within the online shopping domain than women. NPD research (Bhatnagar and Ghose, 2004) on the other hand reasoned that female segments overstate male shoppers in all product categories within the online domain. Referring specifically to the mobile shopping domain, Chong (2013) saw no significant difference between women and men in their m-commerce activities, whereas Okazaki & Mendez (2012) indicated that males tend to engage more in mobile transactions; whereas women like to use mobile channels to explore shopping offers. It is therefore that I suspect gender to have a significant effect on segment membership based on multichannel behavior.

(31)

_25!

Educational status. Konus, Verhoef and Neslin (2008) and Heitz-Spahn (2013) indicated that the most consistent relationship emerges between multichannel behavior and education. Those with a high education level see the benefit of extensive search before purchase. Still, no significant relationship has been identified. Only researchers that looked into the online non-mobile Pc/Laptop shopping domain have found significant relationships indicating that Internet users have higher educational degrees compared to offline users (Farag et al. 2005, Bhatnagar & Ghose, 2004). Yang and Kim (2012) and Chong (2013) furthermore argued that consumers with higher educational levels have greater purchasing power, and therefore conduct transactions more frequently using all sorts of channels

including mobile. It is therefore that I suspect educational level to have an effect on segment membership based on multichannel behavior.

Marital status and Number of children. Farag et al. (2005), Bhatnagar & Ghose (2004) and Bellman, Lohse and Jonson (1999) stated that time pressured heads of households; those that are married or divorced with children, tend to multichannel shop more. They shop online for functional reasons (efficiency and time) and shop offline for recreational reasons, in which the children are taken along. According to Yang and Kim (2012), mobile shopping seems to have a solution for those consumers with children due to mobiles ability to shop anywhere and anytime. It is therefore that I suspect marital status as well as number of children to have an effect on segment membership based on multichannel behavior. In sum;

H3a. Age (demographical trait) is significantly associated with segment membership based on multichannel behaviour.

H3b. Gender (demographical trait) is significantly associated with segment membership based on multichannel behaviour.

H3c. Educational level (demographical trait) is significantly associated with segment membership based on multichannel behaviour.

H3d. Marital status and number of children (demographical trait) are significantly associated with segment membership based on multichannel behaviour.

(32)

_26!

4. Methodology

To investigate the existence and composition of multichannel customer segments based on consumers mobile, online non-mobile PC/Laptop and offline in-store channel appropriateness, actual use behaviour, and to test the proposed hypothesis on covariates, quantitative data has been collected. This chapter describes respectively the used sample, research design and procedure, the analysis method, measures and scales used to identify and profile distinct multichannel-shopping segments.

4.1 The Sample

In order to create segments based on consumer’s multichannel shopping behaviour, a sample has been derived from the 18-plus Dutch shopping population, which approximately consists of 12.983.249 million consumers (CBS, 2014). The sample selection method that has been used is a non-probability (quota) sampling technique distributed via social media (aim: respondents ages 18 - 29), mail and Linked-in (aim: respondents aged 30 <), because no sampling frame could be selected from this enormous population. Therefore

the main goal for data collection was to achieve a sample as large as possible to increase the chance of having a representative/ reliable sample including both younger’s and elder

people. In this way it was still able to generalise from my findings, only not on statistical grounds (Sanders, Lewis and Thornhill, 2009). Since past multichannel researches (Konus,

Verhoef and Neslin, 2008, Neslin et al. 2005) have indicated a response rate of 40-50%, aim of this study was to reach a sample of at least 200-250 potential participants, to assure a minimum sample size of 100 shopping consumers in the Netherlands in which both youngsters (18-29) and elderly (30 <) are present.

On the 12th _{of November 2014, a total of 279 respondents were present.}

Subsequently 75 of these have been deleted due to insufficient or non-answered questions, which represents a dropout rate of 26,8 per cent. So in total 204 questionnaires were able to use for analysis purposes of which eight still represented some missing values, which have been taken care of throughout the analysis with the listwise deletion. In total N=196 has been

(33)

_27!

used during the analysis. The aim to find a quota sample that represented both youngsters (18-29) and elderly (30<) has been almost fully accomplished. In total 101 respondents (49,5%) were ages between 18-29, 94 respondents (46,1%) were 30<, and one participant was aged 17 (0.5). And with respect to gender, 101 respondents were male (49,5%) and 95 respondents represented females (46,6%).

4.2 Research Design

Since the purpose of this research is to segment consumers based on their multi-channel appropriateness/use, and next profile them with the use of active covariates an

inductive quantitative research approach has been used. In specific, the strategy chosen contains a cross-sectional, self-administered (online questionnaire) survey, which enables the collection of structured data (increases validity/ reliability) and large amount of respondents (Sanders & Lewis, 2012). The reason for this approach is that directly after data collection and preliminary analysis a latent class cluster analysis (LCCA) has been executed to see if different clusters of consumers based on their channel appropriateness/use, profiled by the hypothesized covariates; expected benefits, psychological and demographic traits could be identified (Saunders, Lewis and Thornhill, 2009). This inductive research approach has led to the development of new multichannel segments ‘theory’ and is therefore considered a mix of descriptive and explanatory research.

4.2.1 Survey

The survey consisted of two parts (appendix 3). Part one consisted of multichannel behaviour questions and represented the indicators on which the actual segmentation is based. Respondents indicated their ‘use’ and ‘appropriateness’ of mobile, online non-mobile PC/Laptop and offline in-store channels during the pre-purchase (information search) and purchase (transaction) phase. For validation reasons and as past research has indicated that segments could differ per category (Konus, Verhoef and Neslin, 2008, Keen et al. 2004),

questions have been asked for three distinct categories respectively, namely clothing, consumer electronics and flight and entertainment (concert/ movie) tickets, selected in terms

(34)

_28!

of their differences in tangibility, purchase frequency and perceived monetary value. This incorporation of different categories is highly important as it appeared for example out of results from Konus, Verhoef and Neslins’ (2008) segmentation study that the multichannel enthusiast segment (those that shop in store, online and through catalogue) remained quite small in the clothing category but accounted for a majority of consumers in the electronics category. This indicates that segments could differ per category and therefore segments identified in one category are not generalizable for others. The identification of multichannel segments are therefore based on the three categories considered together. Furthermore, part two consisted of questions to test the hypothesis about the covariates; expected benefit,

psychological and demographical traits that enabled to profile the identified segments.

4.3 The Procedure

The online survey has been created and hosted on Qualtrics (2014) both in English and Dutch. To assure high quality of the distributed survey a pilot study has firstly been conducted in order to evaluate the study design prior to the main study.

4.3.1 Pilot study

The pilot study has been executed from the 29th_{of October until the 2}nd_{of November.}

Within the pilot study the reliability and validity of the questions has been tested by asking two master students, two HBO students’, two working professionals and two heads of households about their critics and improvement point with respect to the questionnaire (Sanders & Lewis, 2012). Some questions and scales were interpreted incorrect and therefore changes have been made before administration of the actual questionnaire.

4.3.2 Main study

The final questionnaire (including cover letter) has been administered on the 3rd_of

November through social media, mail and linked-in. To increase reliability of the survey, the snowball technique has been used, to gain more respondents outside my personal network.

Several friends and relatives have distributed the questionnaire through their business as well as personal networks via email or social media, which enabled the collection of a more

Extending multichannel customer segments : does mobile channel use matter?

MASTER THESIS