Convenience or Personalization?
Drivers of consumer adoption of mobile retail apps
in the grocery retail landscape
1
1st Supervisor:
Dr. Hans Berger, Faculty of Economics and Business
2nd Supervisor:
2
Table of Content
ABSTRACT ... 1
INTRODUCTION ... 1
CONCEPTUAL FRAMEWORK ... 5
Literature review and research contribution ... 5
App Features ... 7
Perceived ease of use ... 9
Control Variables ... 10 METHOD ... 13 Sampling ... 16 Statistical technique ... 16 RESULTS ... 18 Sample characteristics ... 18 Measurement model ... 18 Structural model ... 21 Hypotheses 1 ... 22 Hypotheses 2 ... 23 GENERAL DISCUSSION ... 24 Conclusion ... 24 Managerial implications... 24 Theoretical implications... 25
Limitations and future research ... 26
REFERENCE LIST ... 28
1
ABSTRACT
Recent developments in the grocery retail industry such as the entry of Amazon and the growing success of hard discounters, have put more and more pressure on managers to find innovative solutions to remain competitive. In this perspective, the aim of this research is to determine a successful design for grocery mobile apps, an effective solution to differentiate from competitors. The model proposed for this study is a development of the Technology Acceptance Model (TAM). This has been often used to assess the impact of new technologies on consumers’ behaviour, although many have pointed the need to develop it, providing more actionable guidance to managers and practitioners. This study replaces one of its main constructs, perceived usefulness, with three of the most popular retail app features: E-coupons, Price checks and Personalization. The analysis was conducted on the data obtained by a consumer survey with 264 respondents, mainly students. Through a partial least square regression, this study reveals how among the three proposed app features, Price checks (.407) has a higher impact on consumers adoption than E-coupons (.240) and Personalization (.190). No moderating effect of ease of use was found on the three constructs, although it showed a significant direct effect on the dependent variable (.119). These insights can definitely be valuable for managers and practitioners interested in finding the most desirable mobile app design. Furthermore, this study can also serve as a starting point for further research in this direction.
INTRODUCTION
The grocery retail industry is currently undergoing a massive disruption: in 2018 alone, a dramatic 8000 stores had to close (Green 2017). Several factors can be held responsible for this transformation, such as the constant rise of hard discounters, the advent of online retail giants and technological and digital innovations, among others (Kahn 2018).
2 retailers such as SuperValu, Kroger and Sprouts registered losses of respectively 14.4%, 9.2%, 6.3% as result (Rigby 2017). Amazon plans to introduce Amazon Go, a grocery store where you need only to scan your phone to be able to get the products and just walk out, a possibility that might change the experience of grocery shopping forever (Wingfield 2018). While the retail sector was already experimenting different solution to implement digital solutions alongside their physical presence, the news of Amazon’s entrance and its revolutionary plans put much more pressure on the process. However, outracing Amazon in terms of innovativeness is easier said than done, given the expensiveness of the process and the fact that many retailers have overlooked innovation funding for a long time, forming an important gap (Rigby 2017).
On the other hand, while technological and digital innovations are definitely reshaping the industry, they can also represent an opportunity desperately sought by conventional retailers to differentiate from such fierce competitors (BRP 2018). In fact retailers able to offer significant value and provide consumers with targeted information and offers can remain competitive by creating high engagement (Grewal et al 2017). According to a recent market research in the US, technology concerns are increasingly becoming more crucial for retailers, as 73.6% of respondents admitted they plan to increase their technology spending this year (Hofbauer 2018). Furthermore the same research highlights the growth of importance of some omnichannel services offered, with Mobile taking the top spot, expanding from 29.6% to 54.2% rate of adoption among retailers. Smartphones and mobile apps can act like bridges between digital and physical world, providing the consumer with high quality experience and personalization. Companies on the other hand can utilize apps to better identify consumers, tracking their behaviour and register better conversion through stored account and payment
credentials (Rowinski 20173). In particular shopper-facing technologies such as mobile
applications can be indispensable for retailers, providing many opportunities to increase revenues: for example by increasing quantity purchased by current customers, by attracting new ones, or from higher suppliers’ payments, due to superior targeting capabilities (Inman and Nikolova 2017).
The use of smartphones is constantly growing worldwide, just in 2014 there were an
estimated 2.5 billion in the world (eMarketer 2014). A new form of Ecommerce called M-commerce has come up as a consequence of the increase of smartphone users
3 features to mobile phones that make them interesting marketing tools: personal relationship with owners, fast-paced information, portability, salience of contents and availability multiple functions such as navigation and email (Larivière et al. 2013, Shankar 2009).
Smartphones contain several personal information such as conversations, photos, bank account, email and more features that make it a proper cultural object, highly social by nature. This makes it the perfect instrument to effectively develop the relationship between retailers and consumers (Shankar et al. 2010). From an organizational point of view, a research by mobile solutions firm DMI found that improving the mobile shopping experience would bring significant benefits to retailers, in terms of both brand performance and revenue drivers (DMI 2016).
Nonetheless, providing an adequate mobile shopping experience has proven to be a harder challenge than retailers thought. It can be said in fact that retailers are failing when it comes
to the in-store digital experience (Rowinski 20171). According to a market study, stores in the
U.S. and Europe had a benchmark rating of respectively 20 and 17 (out of 100) of in-store mobile experiences against customers’ expectations (DMI 2016). The report was mainly issued to address the lack of retailers’ understanding regarding what customers ask out of mobile in-store experience. Several researches have tried to explain what factors influence the adoption of mobile shopper marketing (McLean 2018), but the discussion has only been aimed at the more general retail landscape, not focusing on the particular situation of the grocery retail sector. The grocery retail sector represents a stand-alone case when it comes to technological innovation: many retailers have delayed innovation funding for a long time (Rigby, 2017) and their interest towards technological applications has only recently become a top concern (Hofbauer, 2018). The entrance of Amazon has definitely accelerated the digitalization process of the industry, which implies that grocers cannot afford anymore the luxury of calmly experimenting the introduction of digital innovation such as mobile apps (Rigby, 2017). Therefore, in such an increasingly unpredictable market, with many signs of retailers’ struggle to get their digital experience right, a more precise and detailed depiction of shoppers’ preferences for mobile apps is crucial. This is confirmed by the significantly poor results of supermarkets apps, compared to those of other retail industries such as drugstores, electronics and apparel (Gray, 2015).
4 Christopher et al. 2018), but none of these was referred to grocery retail’s shopper apps. Furthermore, different authors have claimed how existing literature does not provide with practical advices on how to successfully devise shoppers’ apps (Christopher 2018; Shankar 2016). Other authors have even claimed how the field of mobile shopping “is still in its infancy” (Groß 2015, p. 222).
In addition, in line with the purpose of improving the specificity of the inquiry on how to effectively design mobile apps, this research attempts to enhance the Technology Acceptance Model. This answers the call from several authors to improve the TAM, making it more specific and detailed, in order to provide more actionable guidance to practitioners (Lee et al. 2003; Roy et al. 2018). Nevertheless, this research takes a different approach to the issue of integrating the Technology Acceptance Model. In particular, much research has been aimed at finding antecedents to the two constructs of the model. This paper tries instead to replace the generic perceived usefulness with app features, used as more detailed and precise constructs for investigating the issue of how to effectively design a grocer mobile app.
This paper attempts to fill these gaps, addressing the singular grocery retail landscape and zooming in from the mobile shopper marketing topic, focusing on the drivers of successful implementation of mobile shopping apps. Therefore, the focus of this research is:
“What are the key features of shopper mobile apps in the grocery retail industry that increase customer adoption likelihood?”
5
CONCEPTUAL FRAMEWORK
Literature review and research contribution
Much research has tried to investigate the elements that determine the adoption of new technologies. One of the starting points of the whole research in the topic is the article by Davis et al. in 1989, who introduced the Technology Acceptance Model (TAM). The framework takes inspiration from earlier behavioural psychology studies on the relationship between people’s beliefs, attitudes and behaviour. More specifically, these are the Theory of Reasoned Action (TRA) and the Theory of Planned Behaviour (TPB) which propose how one’s behaviour depends on the attitude held towards that behaviour, the perceived control over that behaviour and a subjective norm (Fishbein, Ajzen 1975). The main difference between these two models and the TAM is that in the previous two perceived usefulness directly influences a person’s behaviour while in the TAM this effect is mediated by attitudes. The Technology Acceptance Model was developed in order to investigate the reason behind acceptance (or rejection) of computer technology in organizations, taking into account the benefits provided by computer systems (Davis et al. 1989). The model proposes two factors as the main influence on people’s attitudes towards new technology: perceived usefulness of the new technology and its perceived ease of use.
6 In spite of the fact that several studies have tried to explore the factors influencing the adoption of mobile applications (McLean 2018, Munoz-Leiva et al. 2017, Gupta 2017, Chouali et al. 2017, Christopher et al. 2018), there has not been a focus on the grocery retail sector. Aside from the current uniqueness of the industry, it is worth mentioning how despite a good percentage of retailers already offers a mobile app, the capabilities and features offered by those applications vary significantly among one another (Gray 2015). Some of them have few limited options such as store finders or ability to view weekly circulars, while others offer an actual omnichannel experience (Inman and Nikolova 2017). Several market researches have tried to discover consumers’ attitudes and preferences in terms of grocery
retail apps’ features (Gray 2015, Forrester 2015, Hofbauer 20171, Hofbauer 20172, Rowinski
20171, Rowinski 20172). However the existing relevant academic literature provides very
limited advice on how to effectively create and design mobile apps (Christopher 2018). For example in a recent review it was concluded how research in the field of mobile shopping “is still in its infancy”, despite its growing popularity (Groß 2015, p. 222). Also Shankar found that from an app design standpoint, there are still various open research questions such as: - “How can marketers optimize their mobile app design to best influence shoppers on their path to purchase?” – “How should apps be designed to deliver rich experiences across a wide range of devices?” (Shankar 2016, p.42).
7 and Bala 2008, p.276). While much research has focused on finding antecedents and additions to the two constructs, to the best of the author’s knowledge, no research has attempted to substitute them, with the aim to provide more detailed and context-specific information about the drivers of adoption of new technologies. Following the previously mentioned gaps in literature, this research aims to identify the key features (in place of perceived usefulness) of a retail grocery app that increase its adoption likelihood by consumers.
App Features
In a US report on the retail industry made by ARC, the quality of some of the most popular apps was rated by customers. The findings suggested that the retail apps economy is incredibly volatile, with continuous app redesigns, new features and very high customer expectations (Gray, 2015). Furthermore, the research indicated how grocers' apps were on average ranked at the lowest position in terms of consumer rating among other industries. Other research supports the view of how difficult it is to get consumers to download retailer/brand-specific applications on their phones. In another US consumers survey, 60% of respondents had two or less grocers' apps on their phones, while 21% did not have any, implying how there must be a compelling reason for them to download a grocer’s app (Forrester 2015). No wonder then how several market research have made efforts to outline consumers' preferences in terms of shopping apps features and capabilities, to better address the requests of customers. Among the various research on the main reasons to download and use an app, two themes were the most recurring: Convenience and Personalization (Forrester 2015). Going further into detail, a survey developed by Progressive Grocer among 2,000 smartphone users mobile apps unveiled how consumers generally use them for three reasons: reward points and coupons, price checks (both before and while shopping) and
personalization (Hofbauer R., 20171, Rowinski D., 20171). Mobile coupons or E-coupons
8 consumers’ cognition of to what extent they can save money from the use coupons (Dickinger 2008; Chen 2011; Liu et al. 2015).
According to recent surveys, 74% of shoppers use their mobile phone to compare prices (DMI 2016), while around 43% of app users checks prices either before shopping or in-store
(Hofbauer R., 20171). Also the possibility to check an item's price by scanning, taking a
picture or tapping the item on the phone ranked among the top three shoppers’ desired
features from mobile experience (Rowinski D., 20171, DMI 2016). The same research also
found how the presence of each of the most desired features was linked to a superior in-store shopping experience. The 98% of respondents declared to be more likely to use an app able to improve the in-store shopping experience, (DMI 2016).
Personalization can be defined as the customization of services and content to consumers’ preferences (Lee, 2011; McLean, 2018). Additionally, personalization refers to marketers' ability to customize the delivery of the right content, to the right person at the right time (Tam & Ho, 2005; Montgomery, 2009). The personalization feature of mobile apps allow customers to tailor their settings to their own specific needs and to be an active part in creating their own experience (Magrath, 2013; Chang et al., 2010). In a recent survey 58 percent of respondents claimed they want apps to let them choose reward types, while 30 percent shared their interest in having personalized offers delivered to them while in-store
(Hofbauer R., 20171). Another market report states how 86% of consumers say
personalization has some impact on their purchasing decisions. According to this research, shoppers would increase their in-store spending by around 4.7% in case a retailer was able to provide a more personalized experience (Haggerty 2018). These figures reflect the many advantages that personalized services offer to both consumers and firms. For example they provide consumers with products and services better suited to their preferences (Vesanen 2007). Offering relevant products and services also reduces the chance of cognitive overload of consumers and increases convenience (Ansari, 2003). Furthermore, from a firm point of view, personalization can grant competitive advantage over competitors (Murthi, 2003), allowing to charge higher prices (Vesanen 2007) and increase profitability (Zhang, 2009). In previous literature, personalization has also been associated with increased convenience, perceived usefulness and perceived ease of use (Zhao, Fang et al. 2018; Lee, 2011; Magrath, 2013; Chang et al., 2010).
9 320), the previously mentioned apps are used as substitutes of the construct theorized by Davis. Based on all the aforementioned assumptions, the following hypotheses are derived: Hypothesis 1a: The possibility to redeem E-coupons has a positive effect on the adoption likelihood of the retailer’s app.
Hypothesis 1b: The possibility to perform Price checks on products both before and while shopping has a positive effect on the adoption likelihood of the retailer’s app.
Hypothesis 1c: The possibility to receive and to choose Personalized offers or rewards based on the consumer’s preferences and shopping history has a positive effect on the adoption likelihood of the retailer’s app.
Perceived ease of use
10 a positive attitude and behaviour intention on using personalized systems, when these are perceived as more accessible and easy to use (Zhao et al. 2018). Either way, while the research of Zhao et al. succeeds in linking the themes of personalization and ease of use, the aim of this research differs from the previous one: while the previous authors use a broad definition of personalization, this paper focuses on the personalization features offered by grocers’ mobile apps, being represented by the possibility to receive personalized offers and rewards, based on the shoppers’ preferences and past purchases.
As a consequence, the following hypotheses are presented:
Hypothesis 2a: The effect size of E-coupons on adoption likelihood of the app is positively affected by the extent of perceived ease of use of the feature.
Hypothesis 2b: The effect size of Price checks on adoption likelihood of the app is positively affected by the extent of perceived ease of use of the feature.
Hypothesis 2c: The effect size of Personalized offers/rewards on adoption likelihood of the app is positively affected by the extent of perceived ease of use of the feature.
Hypothesis 2d: The perceived ease of use of the app has a positive effect on its adoption likelihood.
Control Variables
In order to effectively assess and isolate the impact of app features and perceived ease of use on adoption likelihood, the possible effect of other external variables needs to be addressed. These variables have been proven to have an influence on both adoption behaviour and preferences for specific types of apps’ capabilities in previous academic research and therefore they need to be controlled.
11 towards another retail digital application, namely smart retail technologies such as iBeacons and smart shelves (Roy et al., 2018).
The second control variable is technology readiness, defined as an “individual's propensity to embrace and use new technologies” (Parasuraman, 2000, p. 308). According to literature, respondents can present both positive and negative technological beliefs at the same time. The predominance of positive beliefs towards technology is more likely to result in an individual’s higher receptiveness to new technology, while those with negative view are likely to resist the acceptance of new technological applications (Rosenbaum and Wong, 2015). Technology readiness has been demonstrated to have a direct impact on both perceived ease of use and perceived usefulness (Ferreira et al. 2014) and it has also been showed to acts as a psychological barrier for adoption of new technologies (Laukkanen et al. 2008).
The third control is gender, which has been used in previous similar studies as a control variable (Roy et al. 2018). Furthermore, a recent survey developed by Progressive Grocer in the US has showed how men and women typically use grocers’ apps for different purposes
(Hofbauer, 20171). In particular, women were more interested in receiving and choosing
rewards, while men considered scan and pay to be a more attractive feature. Also women mostly used the price check function before shopping, while men mainly used it already in-store. This, in addition to the two previously referred control variables, should help isolating the effect of app features and perceived ease of use on adoption likelihood, emphasizing the relationship among them.
13
METHOD
The research was conducted in form of a single cross-sectional design. A computer structured survey provided the data to test the research model. This simplified the administration of the questionnaire and made the data more reliable, given the reduced variability provided by the fixed-response questions (Malhotra, 2010). At first the control variables were addressed: to account for the possible effect of store reputation, the survey revolved around a generic grocer’s mobile app, with no links or references to any known retail brands; technology readiness was measured and then controlled statistically, with questions pertaining technology beliefs such as optimism, innovativeness, discomfort and insecurity, taken from previous research by Rosenbaum and Wong (2015) and Parasuraman (2000); gender was easily determined before the start of the actual questionnaire.
A brief introduction on what the purpose of the research is and a description in detail of each feature were provided. Given the non-intrusive nature of the survey, this was expected to help overcoming both the respondent’s inability and unwillingness to answer (Malhotra, 2010). Respondents were also informed that their response would help finding the most desired design and features for a generic grocer’s mobile app and how each of these features contributes to their intention to adopt said application. E-coupons was defined as the app capability to provide discounts for products, which can be scanned at the check via mobile app. The second feature was illustrated as the capability to check prices of products among all the brands sold at the retailer, both before shopping and in-store via QR scan. Personalization was described as the possibility to receive offers and rewards based on shoppers’ preferences and past purchases (Lee, 2011; McLean, 2018). Furthermore this feature allows some degree of customization to the customer, in terms of freedom of choice among various discounts and rewards.
14 then be asked to rank the three features in order of importance, as to obtain a straightforward insight on customers’ preferences and which of these app capabilities best drives the adoption behaviour.
Perceived ease of use was tested with respect to each of the three proposed app feature and directly to adoption likelihood. A 3-item scale was used to measure perceived ease of use, adapted from the research of Venkatesh and Davis (2000). Responses to all questions (except for the ranking of app features) were measured using a 5 point Likert scale ranging from “strongly disagree (1)” to “strongly agree (5)”. The questionnaire was divided in four small sections, with the first one (technology readiness) related to the control constructs, while the other three contribute to different research purposes:
₋ Technology readiness: Technology beliefs and perceptions are classified in four distinct dimensions, specifically: optimism, innovativeness, discomfort, and insecurity (Parasuraman, 2000). Respondents can demonstrate both positive and negative beliefs about new technology at the same time. It has been showed how the prevalence of one of the two beliefs can significantly impact the adoption of new technologies such as mobile apps (Roy et al. 2018). The following items are derived and adapted from the research by Rosenbaum and Wong (2015):
o Products and services that use the newest technologies are more convenient to use
o In general, I am among the first of in my circle of friends to acquire new technology when it appears
o It is embarrassing when I have trouble with a high-tech gadget while people are watching
o I do not consider it safe giving out personal information over a mobile phone
15 o The possibility to [Feature] is a positive addition to a grocer’s mobile app o The possibility to [Feature] makes a retail mobile app more attractive o I would be more likely to download and use the app if it allowed me to
[Feature]
₋ The second small section of the survey concerned asking respondents their ranking of app features, accordingly to what they perceive as most important in a grocer’s app. This item was not used in the data analysis model, but it only serve as an additional way to confirm the conclusions eventually obtained in the results section. This item in fact only provided the research with a clear and direct insight on which were the consumers’ favourite features in a grocery shopping app. Respondents were asked to rank the three features:
o E-coupons o Price checks
o Personalized offers/rewards
₋ Perceived Ease of use: the first item was taken as an adjustment from previous research by Venkatesh and Davis (2000) and helped measuring the importance of perceived ease of use for app features. The last two items were instead directly aimed at the purpose of this research: that is to measure the effect of perceived ease of use on the relationship between app features and adoption likelihood. More specifically, the last two questions assessed respectively the predicted positive impact of a feature’s ease of use and the negative effect in case it was confusing or difficult to use. The questions in this section are as follows:
o Using the app features should be easy, not requiring a lot of my mental effort
o I would be more willing to download and utilize the app if using its features was clear and effortless
o If the app features were somehow unclear or confusing, I would not download the app, or stop using it
16 way respondents were provided with a chance to actively participate in the study, possibly also providing interesting insight and directions for future research.
Sampling
The ideal sample for this research would be a completely random sample. Unfortunately the limited scale and resources of this research make it impossible to obtain a completely random sample. Convenience sampling is therefore used, in order to easily access cooperative respondents (Malhotra, 2010). Most of the respondents will be close to the personal network of the researcher. This means that a large part of the sample population will be comprised of students aged between 20 and 30. Respondents will have to complete a structured online questionnaire via computer/phone, in order to collect primary data.
Statistical technique
One of the most relevant features of the proposed conceptual model is the presence of a moderating variable (perceived ease of use), which impacts on the other three constructs. A moderator can influence both the strength and the direction of a relationship between two constructs (Malhotra 2010). In this case one of the aims of this paper is to verify if perceived ease of use has an impact on the relationship between the three app features and the dependent variable, adoption behaviour. In this sense, moderation should be seen as a mean to account for heterogeneity in the data (SmartPLS 2018): perceived ease of use accounts for heterogeneity in the features-adoption link, implying that this relationship is not the same for all customers, but instead it differs depending on the level of perceived ease of use.
17 variable scores, repeated until convergence has been obtained; -estimation of outer weights/loading and path coefficients; -estimation of location parameters. PLS is the method of choice for success factor studies in Marketing (Albers, 2009). It has also become increasingly popular in empirical research in international marketing. Many researchers confirm in fact how the goal of their research is in line with some key strengths of PLS path modelling. More into detail, this model is most effective in an early stage of theoretical development, as it allows testing and validating exploratory models. (Henseler et al., 2009). The primary objective of PLS consists in maximizing the explanation of variance in the dependent constructs. Also Joreskog and Wold (1982, p.270) state how “PLS is primarily intended for causal-predictive analysis in situations of high complexity but low theoretical information”. This method is also particularly suitable for the characteristics of this research as it effectively allows working with small samples. It has in fact been proved how its methodology is supported by the majority of factor analyses in international marketing research (Henseler et al., 2009).
Regarding the construct used, here is a list of their function within this model and their measure:
Construct Function Measure
Gender Control variable Binary (dummy coded)
Technology readiness Control variable Scale (Likert 1-5)
E-coupons Independent variable Scale (Likert 1-5)
Price checks Independent variable Scale (Likert 1-5)
Personalization Independent variable Scale (Likert 1-5)
Ease of use Moderator Scale (Likert 1-5)
Adoption Likelihood Dependent variable Scale (Likert 1-5)
18
RESULTS
Sample characteristics
A total of 264 respondents completed the survey. The sample contained 110 men and 154 women. A large part of respondents was aged between 18 and 24 (138), with also a significant presence of the age group between 45 and 54 (58). The youngest respondents belong to the under 18 category (4), while the oldest to over 65 (4). The vast majority of participants’ nationality was Italian (189), with a considerable presence of residents in other EU countries, outside from Italy, UK and The Netherlands (42). A large part of respondents has obtained a high school diploma. It is worth mentioning though how the vast presence of young participants implies that several of those who have not obtained a university degree yet but might be currently enrolled. In particular 113 have obtained a high school diploma, 65 a bachelor degree and 75 a master degree. Only 3 respondents have a PHD or higher.
Measurement model
19 Table 2: Outer loadings
As shown in the table, all the loadings of Easy_Uninstall show values below the threshold of 0.700, both towards the latent variable Ease of use and in the three moderation effects. While also Tech Readiness and Adoption present some indicators with non-sufficient values, as formative variables, they represent an exception. In formative models in fact, each indicator is to be considered a dimension of meaning of the latent variable. The indicators are assumed as “reality” and their collective set represents all the dimensions of the variable. In a formative model therefore, eliminating an indicator can be seen as dropping a dimension of meaning, altering the meaning of the latent variable (Garson, 2016).
Collinearity statistics were also assessed and the other two indicators of Ease of use, Mental_Effort (“Using the app features should be easy, not requiring a lot of my mental effort”) and Easy_Download (“I would be more willing to download and utilize the app if using its features was clear and effortless”) yielded several VIF scores > 4, especially when used in the moderation effect.
Adoption_Ease of use Easeofuse*Coupons_Easeofuse*Personalization Easeofuse*PriceChecks Ecoupons_Gender Personalization PriceChecks_ Tech Readiness_
20 Table 3: VIF scores
21
Structural model
As already mentioned, the structural model is composed by the factors and the arrows that connect them one another. The path loadings are standardized regression coefficients.
The resulting structural model was estimated with Adoption likelihood as the dependent variable and E-coupons, Price checks and Personalization as independent variables. Also the moderation effect of Ease of use on the three IVs has been tested, as well as the effect of Gender and Technology readiness on the DV, as control variables.
The effects of E-coupons and Price checks on Adoption likelihood are significant at 5% level, while the effects of Personalization and Ease of use on Adoption likelihood are significant at 10% level. No moderation effect was found to be significant and the same goes for control variables, Gender and Technology readiness (Appendix 1).
The model estimation produced the following results:
Path Coefficients Standard error P-value
Ease of use 0,119 0,068 0,081 Easeofuse*Coupons 0,040 0,088 0,654 Easeofuse*Personalization 0,064 0,096 0,504 Easeofuse*PriceChecks -0,107 0,076 0,158 Ecoupons 0,240 0,089 0,008 Gender 0,062 0,043 0,153 Personalization 0,190 0,105 0,070 PriceChecks 0,407 0,108 0,000 Tech Readiness 0,033 0,049 0,507
Table 2: estimation results (p-values: ***p ≤ 0.01; **p ≤ 0.05; *p ≤ 0.10)
Looking at the weights, it is clear how the most relevant effect on Adoption likelihood is that of PriceChecks (0,407), followed by E-coupons (0,240). Also Personalization (0,190) and Ease of use (0,119) showed meaningful effects on the outcome variable, although only significant at 10% level. Since path coefficients in PLS do not follow a normal, chi-square, or other known distribution, bootstrapped significance coefficients must be employed, instead of
***
22 usual significance levels (Garson, 2016). While no global fit value is available in PLS-SEM, SmartPLS still provides various coefficients related to model fit. Composite reliability for example is a preferred alternative to Cronbach's alpha a test of convergent validity, mainly because Cronbach's alpha usually underestimates scale reliability. Composite reliability varies from 0 to 1, with 1 being perfect estimated reliability. In a model adequate for confirmatory purposes, composite reliabilities should be equal to or greater than 0.70 (Henseler, Ringle, & Sarstedt, 2012: 269). All constructs in this case a composite reliability above the threshold, exception made for Adoption likelihood which is a formative construct (Appendix 2). Average Variance Extracted (AVE) is used to test both convergent and divergent validity. It reflects the average communality for each latent factor in a reflective model and should be greater than 0.5 (Garson, 2016). Again, all construct in the model show meaningful values above the threshold (appendix 3). The R-square (0.603) and R-square adjusted (0.589) both show high levels.
Hypotheses 1
Hypothesis 1a: The possibility to redeem E-coupons has a positive effect on the adoption likelihood of the retailer’s app.
From the results it can be concluded that there is a positive estimate for E-coupons. The effect is the second in terms of mean (0,240) and it is highly significant (0,008). This suggests that E-coupons is of high importance to shoppers, therefore hypothesis 1a can be accepted.
Hypothesis 1b: The possibility to perform Price checks on products both before and while shopping has a positive effect on the adoption likelihood of the retailer’s app. The results clearly show a positive correlation between Price checks and Adoption likelihood. The effect is both the highest of all constructs (0,407) and the most significant (0,000), therefore hypothesis 1b can be confirmed. From this, it can be stated that the possibility to perform Price checks is considered by respondents as the most important feature within a grocery mobile app.
23 Also Personalization shows a positive correlation with the outcome variable. Out of the three app features, its effect is the lowest (0,190) and it is only significant at 10% level (0,070). Nonetheless, both values are more than adequate, therefore also hypothesis 1c can be accepted. Therefore, out of the three presented app features it can be said that Personalization represents the least important one, although its effect on likelihood of adoption is still very relevant.
Hypotheses 2
Hypothesis 2a: The effect size of E-coupons on adoption likelihood of the app is positively affected by the extent of perceived ease of use of the feature.
Based on the results, no substantial moderation effect of Ease of use on E-coupons was found. Its mean was in fact found to be very low (0,040) and also non-significant (0,654). Hypothesis 2a is therefore rejected.
Hypothesis 2b: The effect size of Price checks on adoption likelihood of the app is positively affected by the extent of perceived ease of use of the feature.
Also no moderation effect of Ease of use was found on Price checks. Its effect (-0,107) was found to be non-significant (0,158). Hypothesis 2b is rejected.
Hypothesis 2c: The effect size of Personalized offers/rewards on adoption likelihood of the app is positively affected by the extent of perceived ease of use of the feature. The same argument goes for the moderation effect of Ease of use on Personalization. Its mean value was quite low (0,064) and also non-significant (0,504). Again, hypothesis 2c is rejected.
Hypothesis 2d: The perceived ease of use of the app has a positive effect on its adoption likelihood.
24
GENERAL DISCUSSION
Conclusion
Following the results of the analysis, it can be stated that the first aim of this study was successfully achieved. All the three constructs of app features yielded in fact significant positive effects on Adoption likelihood. Among them, the strongest effect was that of Price checks, followed by E-coupons and lastly by Personalization. To answer therefore the question proposed as title for this study, the results of this research imply that shoppers prefer convenience over personalization. Further discussion on the possible motivation and interpretation of these results is presented in the limitations and future research section. However, the same cannot be said about the second aim of the research. No significant moderation effect of Ease of use was found on either of the three app features’ constructs, therefore hypothesis 2a, 2b and 2c were rejected. Nevertheless, a positive and significant direct effect of Ease of use on Adoption likelihood was found. Several research have proven in the past the significance of ease of use for consumers’ adoption of mobile apps and new technology in general (Christopher, 2018; Grob, 2015; McLean, 2018; Roy et al., 2018). Again, this study differentiated from previous ones in first place by zooming in from the Technology Acceptance Model, making it more specific. This, by replacing perceived usefulness with the specific features of the proposed app. The fact that perceived ease of use showed a significant effect on likelihood of adoption, while its interactions with the specific features are not significant, is therefore very likely dependent on the very attempt of this research to surpass the TAM: this could mean in fact that shoppers do take into consideration ease of use in deciding whether or not to download and use a grocery mobile app, but they just do not consider the ease of use of each feature, but more of the mobile app as a whole.
Managerial implications
25 Following the results provided by this study, the main takeaway is that consumers value very much convenience and simplicity when doing grocery shopping, more than they care about personalization. Price checks and E-coupons were in fact respectively the first and second most rated features among the three, all while being the only two to show significance at the confidence level of 1%. As a consequence, the main suggestion to managers would be to implement and focus on both these features in their mobile apps, as both have been shown to lead to higher likelihood of adoption. Also Personalization has shown a positive effect on consumer adoption, although it was both the lower and least significant of the three (confidence level of 10%). This implies that the majority of shoppers do value the possibility to receive personalized offers and rewards, but possibly not all of them feel the same way about it. More observations and interpretations on the meaning of this result will follow in the limitations and future research section. A similar discussion can be made about perceived ease of use: as Personalization, it showed a positive significant effect on adoption, although again only significant at 10% level. It has been mentioned in the previous section the possible explanation for its significant direct effect and at the same time non-significant interaction with app features: shoppers might not differentiate between the ease of use of a single feature and that of the app in general. Nonetheless, the importance of ease of use was proved by countless past studies and confirmed by the present one. Therefore, both managers and the software developer team should also put careful attention to developing a mobile app which is easy to use and does not require shoppers to put excessive mental effort into it.
Theoretical implications
26 usefulness, with three different app features. This, in order to provide managers and practitioners with practical advice and actionable guidance on how to effectively design a mobile app, based on shoppers’ requests. The proposed model used E-coupons, Price checks and Personalized offers/rewards as constructs instead of perceived usefulness, moving a step forward from the TAM. The first aim of the analysis was therefore to check whether these three features played a significant role on consumers’ behaviour, represented by likelihood of adoption of the mobile app, and which of them yielded the strongest effect. The second aim of the analysis was instead to check whether Ease of use influenced the effect of the three app features on consumers’ behaviour. This interaction effect was theorized following the moderation effect of perceived ease of use on perceived usefulness affirmed by the TAM.
Limitations and future research
27 the effectiveness of personalization is influenced by how firms collect consumer data: personalization based on overtly collected data is linked with higher adoption rate by consumers, while covertly collected data results in lower adoption rates (Aguirre, Mahr et al. 2015). Furthermore, privacy concerns have also been added to the Technology Acceptance Model as an additional construct. Privacy concerns were found to play a negative effect on both perceived usefulness and perceived ease of use (Fortes, Rita 2016). For these reasons, future research should account for the effects of privacy concerns on personalization, to test whether this influences personalization’s effect on adoption likelihood. This could be a possible explanation of why personalization was found to be the least relevant out of the three features and should therefore be verified in future studies.
One last direction from future research comes from the insight obtained through the last item of the survey, which asked respondents whether they would be interested in other features in a grocery mobile app, other than the ones proposed in the research. The question obtained 81 meaningful responses, who were then analysed to find the most recurring suggestions from respondents. 13 of these responses (16,05%) expressed interest in obtaining information/notifications on new products or offers in the app (i.e. circulars). 8 participants (9,88%) expressed an interest towards having information on how to locate a product within the store. The other two relevant responses obtained were the possibility to place an order online to the store through the mobile app and the possibility to scan items, to both receive product information and to allow self-checkout. Both these last two suggestions were registered 6 times (7,41% each). All these insight suggest that consumers have various and different requests from a grocery mobile app and no dominant design has been found yet. Future research could replicate this study with these last features requested by respondents. Another suggestion could be to attain a comprehensive analysis of all possible app features and their different impact on customer adoption.
28
REFERENCE LIST
Agarwal, R. and Prasad, J. (1998), “A conceptual and operational definition of personal innovativeness in the domain of information technology”, Information Systems Research, Vol. 9 No. 2, pp. 204-215.
Agrebi S and Jallais J (2015) “Explain the Intention to Use Smartphones for Mobile Shopping,” Journal of Retailing and Consumer Services, 22, pp. 16–23. doi: 10.1016/j.jretconser.2014.09.003.
Albers S. (2010) PLS and Success Factor Studies in Marketing. In: Esposito Vinzi V., Chin W., Henseler J., Wang H. (eds) Handbook of Partial Least Squares. Springer Handbooks of Computational Statistics. Springer, Berlin, Heidelberg.
ANSARI, A. and MELA, C.F., 2003. E-Customization. Journal of Marketing Research, 40(2), pp. 131-145.
Bao, Y., Bao, Y., Sheng, S., 2011. Motivating purchase of private brands: effects of store image, product signatureness, and quality variation. J. Bus. Res. 64 (2), 220–226.
Boulding, W., Kirmani, A., 1993. A consumer-side experimental examination of signaling theory: do consumers perceive warranties as signals of quality? J. Consum. Res. 20 (1), 111–123.
BRP Consulting. 2018. Personalization is Key to Customer Loyalty - BRP Consulting. [ONLINE] Available at: https://brpconsulting.com/download/2018-special-report-customer-loyalty/. [Accessed 27 September 2018].
Business Insider (2013), “Mobile coupons smash paper coupon redemption rates”, available at:
www.businessinsider.com.au/mobile-commerce-and-mobile-coupon-growth-2013-4. [accessed 23 September, 2018]. Business Insider, 2015. German store Lidl plans US expansion - Business Insider . [ONLINE] Available at: https://www.businessinsider.com/german-store-lidl-plans-us-expansion-2015-11?international=true&r=US&IR=T.
[Accessed 27 September 2018].
Cait Lamberton, Adrew T. Stephen (2016), “A Thematic Exploration of Digital, Social Media, and Mobile Marketing: Research Evolution from 2000 to 2015 and an Agenda for Future Inquiry”, Journal of Marketing, 80 (November), 146-172. Chang, W. L., Yuan, S. T., & Carol, W. (2010). Creating the experience economy in ecommerce. Communications of the AMC, 53(7), 122e127.
Chaouali, W., Souiden, N., & Ladhari, R. (2017). Explaining adoption of mobile banking with the theory of trying, general self-confidence, and cynicism. Journal of Retailing and Consumer Services, 35, 57e67.
Chen, M.-F. and Lu, T.-Y. (2011), “Modeling e-coupon proneness as a mediator in the extended TPB model to predict consumers’ usage intentions”, Internet Research, Vol. 21 No. 5, pp. 508-526.
Collier, J.E. and Barnes, D.C. (2015), “Self-service delight: exploring the hedonic aspects of self-service”, Journal of BusinessResearch,Vol.68No.5,pp.986-993.
29
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1992). Extrinsic and intrinsic motivation to use computers in the workplace. Journal of Applied Social Psychology, 22, 1111e1132.
Davis, F. D., Bagozzi, R. P. and Warshaw, P. R. (1989) “User Acceptance of Computer Technology: A Comparison of Two Theoretical Models,” Management Science, 35(8), pp. 982–1003.
Dickinger, A. and Kleijnen, M. (2008), “Coupons going wireless: determinants of consumer intentions to redeem mobile coupons”, Journal of Interactive Marketing, Vol. 22 No. 3, pp. 23-39
DMI, 2016. M3 Mobile Maturity Model: 2016 U.S. RETAIL. [ONLINE] Available at:
http://pages.dminc.com/USA-17-retail-maturity-model.html. [Accessed 29 September 2018].
EMARKETER, 2014-last update, 2 Billion Consumers Worldwide to Get Smart(phones) by 2016 Over half of mobile phone users globally will have smartphones in 2018. Available: https://www.emarketer.com/Article/2-Billion-Consumers-Worldwide-Smartphones-by2016/1011694.
Faulds, D. J., Mangold, W. G., Raju, P. S., & Valsalan, S. (2018). The mobile shopping revolution: redefining the consumer decision process. Business Horizons. https://doi.org/10.1016/j.bushor.2017.11.012.
Ferreira, J.B., da Rocha, A., da Silva, J.F., 2014. Impacts of technology readiness on emotions and cognition in Brazil. J. Bus. Res. 67 (5), 865–873.
Forrester (2015). “The State Of Mobile Apps For Retailers.”. Available at:
https://www.retailmenot.com/corp/static/filer_public/78/9c/789c947a-fe7c-46ce-908a-790352326761/stateofmobileappsforretailers.pdf
Garson, G. D. (2016). Partial Least Squares: Regression and Structural Equation Models. Asheboro, NC: Statistical Associates Publishers.
GOOGLE SHOPPER MARKETING COUNCIL, 2013-last update, Mobile In-Store Research How In-store Shoppers are Using Mobile Devices. Available: https://www.thinkwithgoogle.com/advertisingchannels/mobile/mobile-in-store/.
Gray, Ben (2015), The Best- and Worst-Rated Retail Apps,. Applause Analyt-ics ARC Report, [accessed online: http://go.applause.com/rs/539-CKP-074/images/ARC-The-Best-and-Worst-Rated-Retail-Apps-2015.pdf]
Green, Dennis, and Megan Harney (2017), “More Than 8,000 Store Closures Were Announced in 2017 — Here’s the Full List,” Business Insider, http://www.businessinsider.com/stores-closures-announced-in-2017 -2017-12/#radioshack-1430-stores-1.
Grewal, Dhruv, Yakov Bart, Martin Spann, Peter Pal Zubcsek (2016), “Mobile Advertising: A Framework and Research Agenda”, Journal of Interactive Marketing, 34 (May), 3-14.
Grewal, Dhruv, Anne L. Roggeveen, and Jens Nordfält, (2017). "The future of retailing." Journal of Retailing 93, no. 1: 1-6. Groß, M., (2015) Exploring the acceptance of technology for mobile shopping: an empirical investigation among Smartphone users, The International Review of Retail, Distribution and Consumer Research, 25:3, 215-235, DOI: 10.1080/09593969.2014.988280.
30
Haggerty D., 2018. Bringing the Golden Rule of Retail into an Omnichannel World. Applause Blog. [ONLINE] Available at:
https://www.applause.com/blog/golden-rule-of-retail-omnichannel-world/. [Accessed 13 September 2018].
Hair, J. F., Hult, G. T. M., Ringle, C. M., and Sarstedt, M. (2017). A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM), 2^nd^ Ed., Sage: Thousand Oaks.
Henseler, J., Ringle, C. M., and Sinkovics, R. R. (2009). The Use of Partial Least Squares Path Modeling in International Marketing, in Advances in International Marketing, R. R. Sinkovics and P. N. Ghauri (eds.), Emerald: Bingley, pp. 277-320. Henseler, Jörg; Ringle, Christian M.; & Sarstedt, Marko (2012). Using partial least squares path modeling in international advertising research: Basic concepts and recent issues. Pp. 252-276 in Okzaki, S. Handbook of partial least squares: Concepts, methods and applications in marketing and related fields. Berlin: Springer.
Hoehle, H. and Venkatesh, V. (2015), “Mobile application usability: conceptualization and instrument development”, MIS Quarterly,Vol.39 No.2, pp.435-472.
Hofbauer R., 20171. Grocers’ Mobile Apps Need to Do More: Survey | Progressive Grocer. [ONLINE] Available at: https://progressivegrocer.com/grocers-mobile-apps-need-do-more-survey. [Accessed 13 September 2018].
Hofbauer R., 20172. Personalization, Respecting Data Key to Loyalty Program Success | Progressive Grocer. [ONLINE] Available at: https://progressivegrocer.com/personalization-respecting-data-key-loyalty-program-success. [Accessed 15 September 2018].
Hofbauer R., 2018. Tech's Now a Top Concern for Grocers – Here's Where They're Investing in It: Annual Report | Progressive Grocer. [ONLINE] Available at: https://progressivegrocer.com/techs-now-top-concern-grocers-heres-where-theyre-investing-it-annual-report. [Accessed 13 September 2018].
Holmes A, Byrne A and Rowley J (2014) “Mobile Shopping Behaviour: Insights into Attitudes, Shopping Process Involvement and Location,” International Journal of Retail and Distribution Management, 42(1), pp. 25–39. doi: 10.1108/IJRDM-10-2012-0096.
Hong, S., & Tam, K. Y. (2006). Understanding the adoption of multi-purpose information appliances: The case of mobile data services. Information Systems Research, 17, 162–179.
Hsiao, C.-H., Chang, J.-J. and Tang, K.-Y. (2016) “Exploring the Influential Factors in Continuance Usage of Mobile Social Apps: Satisfaction, Habit, and Customer Value Perspectives,” Telematics and Informatics, 33(2), pp. 342–355. doi: 10.1016/j.tele.2015.08.014.
Im,H., & Ha, Y. (2012).Who are the users of mobile coupons? A profile of US consumers. Journal of Research in Interactive Marketing, 6(3), 215–232.
Inman, J.J. & H. Nokolova, 2017, "Shopper-Facing Technology: A Retailer Adoption Decision Framework Incorporating Shopper Attitudes and Privacy Concerns". Journal of Retailing, vol. 893, issue 1, 7-28.
Joreskog, K.G.,& Wold, H.O.(Eds). (1982). The ML and PLS technique for modelling with latent variables: Historical and comparative aspects. In: Systems under indirect observation, Part I (pp. 263–270). Amsterdam: North-Holland.
31
Kang, J.-Y. M., Mun, J. M., & Johnson, K. K. P. (2015). In-store mobile usage: Downloading and usage intention toward mobile location-based retail apps. Computers in Human Behavior, 46, 210e217.
Kaushik A.K and Rahman Z (2015) “An Alternative Model of Self-Service Retail Technology Adoption,” Journal of Services Marketing, 29(5), pp. 406–420. doi: 10.1108/JSM-08-2014-0276.
KIM, C., MIRUSMONOV, M. and LEE, I., 2010. An empirical examination of factors influencing the intention to use mobile payment. Computers in Human Behavior, 26(3), pp. 310-322.
Koufaris, M. (2002). Applying the technology acceptance model and flow theory to online consumer behavior. Information Systems Research, 13, 205–223.
Kumar, Ashish, Ram Bezawada, Rishika Rishika, Ramkumar Janakiraman, P.K. Kannan (2016), “From Social to Sale: The Effects of Firm-Generated Content in Social Media on Consumer Behavior”, Journal of Marketing, 80 (January), 7-25. Kumar, V., Anand, A. and Song, H. (2017) “Future of Retailer Profitability: An Organizing Framework,” Journal of Retailing, 93(1), pp. 96–119. doi: 10.1016/j.jretai.2016.11.003.
Larivière, B., Joosten, H., Malthouse, E.C., Marcel, V.B., Aksoy, P., Kunz, M.H. and Minghui Huang, 2013. Value fusion: The blending of consumer and firm value in the distinct context of mobile technologies and social media. Journal of Service Management, 24(3), pp. 268-293.
Laukkanen, P., Sinkkonen, S., Laukkanen, T., 2008. Consumer resistance to internet banking: postponers, opponents and rejectors. Int. J. Bank Mark. 26 (6), 440–455.
Lee, C. H., & Crange, D. A. (2011). Personalisation - privacy paradox: The effects of personalisation and privacy assurance on customer responses to travel web sites, Tourism Management. 32, 987e994.
Lee, Thaemin. (2005). The impact of perceptions of interactivity on customer trust and transaction intentions in mobile commerce. Journal of Electronic Commerce Research. 6.
Lee, Y., Kozar, K. A., & Larsen, K. (2003). The technology acceptance model: Past, present, and future. Communications of the Association for Information Systems, 12(50), 752–780.
Liu, F., Xuefeng Zhao, Patrick Y.K. Chau, Qing Tang, (2015) "Roles of perceived value and individual differences in the acceptance of mobile coupon applications", Internet Research, Vol. 25 Issue: 3, pp.471-495, https://doi.org/10.1108/IntR-02-2014-0053.
Magrath, V., & McCormick, H. (2013). Marketing design elements of mobile fashion retail apps. Journal of Fashion Marketing and Management: An International Journal, 17, 115e134.
Malhotra, N. K. (2010) Marketing research : an applied orientation. 6th edn. Boston: Pearson.
McLean, G. (2018) “Examining the Determinants and Outcomes of Mobile App Engagement - a Longitudinal Perspective,” Computers in Human Behavior, 84, pp. 392–403. doi: 10.1016/j.chb.2018.03.015.
32
MONTGOMERY, A.L. and SMITH, M.D., 2009. Prospects for Personalization on the Internet. Journal of Interactive Marketing, 23(2), pp. 130-137.
Munoz-Leiva, F., Climent-Climent, S., & Liebana-Cabanillas, F. (2017). Determinants of intention to use the mobile banking apps: An extension of the classic TAM model. Spanish Journal of Marketing - ESIC, 21, 25e38.
MURTHI, B.P.S. and SARKAR, S., 2003. The Role of the Management Sciences in Research on Personalization. Management Science, 49(10), pp. 1344-1362.
Parasuraman, A., 2000. Technology Readiness Index (TRI) a multiple-item scale to measure readiness to embrace new technologies. J. Serv. Res. 2 (4), 307–320.
Rigby, Darrell. “The Amazon - Whole Foods deal Means Every Other Retailer’s Three-Year Plan Is Obsolete”, Harvard Business Review, June 2 (2017).
Ringle, C.M., Wende, S. and Becker, J.M., 2015. SmartPLS 3. Boenningstedt: SmartPLS GmbH, http://www. smartpls. com. Rowinski D., 20171. How Retailers Are Failing The Mobile In-Store Shopping Experience. Applause Blog [ONLINE] Available at: https://www.applause.com/blog/mobile-shopping-experience/. [Accessed 13 September 2018].
Rowinski D., 20172. Retailer's Biggest Obstacle To Personalization Is Payments. Applause Blog [ONLINE] Available at:
https://www.applause.com/blog/retail-personalization-obstacles/. [Accessed 13 September 2018].
Rowinski D., 20173. The Pros And Cons Of Native Apps And The Mobile Web For Ecommerce. Applause Blog [ONLINE] Available at: https://www.applause.com/blog/ecommerce-apps-and-mobile-web-faq/. [Accessed 13 September 2018]. Roy, S. K., Balaji, M. S., Quazi, A. and Quaddus, M. (2018) “Predictors of Customer Acceptance of and Resistance to Smart Technologies in the Retail Sector,” Journal of Retailing and Consumer Services, 42, pp. 147–160.
Shankar, V. and Balasubramanian, S., (2009) “Mobile Marketing: A Synthesis and Prognosis,” Journal of Interactive Marketing, 23(2), pp. 118–129. doi: 10.1016/j.intmar.2009.02.002.
Shankar V, Venkatesh A, Hofacker C and Naik P (2010) “Mobile Marketing in the Retailing Environment: Current Insights and Future Research Avenues,” Journal of Interactive Marketing, 24(2), pp. 111–120. doi: 10.1016/j.intmar.2010.02.006. Shankar, V., Kleijnen, M., Ramanathan, S., Rizley, R., Holland, S. and Morrissey, S. (2016) “Mobile Shopper Marketing: Key Issues, Current Insights, and Future Research Avenues,” Journal of Interactive Marketing, 34, pp. 37–48. doi: 10.1016/j.intmar.2016.03.002.
SmartPLS. 2018. Moderation | SmartPLS. [ONLINE] Available at: https://www.smartpls.com/documentation/algorithms-and-techniques/moderation. [Accessed 03 November 2018].
STATCOUNTER, 2018-last update, Desktop vs Mobile vs Tablet Market Share Worldwide. Available:
http://gs.statcounter.com/platform-market-share/desktop-mobile-tablet.
Tam, K. Y., & Ho, S. Y. (2005). Web personalization as a persuasion strategy: An elaboration likelihood model perspective. Information Systems Research, 16(3), 271e291.
33
Taylor, S. and Todd, P. A. (1995) “Understanding Information Technology Usage: A Test of Competing Models,” Information Systems Research, 6(2), pp. 144–176. doi: 10.1287/isre.6.2.144.
The Guardian. 2015. Morrisons sacks supermarket boss Dalton Philips | Business | The Guardian. [ONLINE] Available at: https://www.theguardian.com/business/2015/jan/13/morrisons-chief-dalton-philips-quits-supermarket. [Accessed 27 September 2018].
Venkatesh, V. (2000). Determinants of perceived ease of use: Integrating perceived behavioral control, computer anxiety and enjoyment into the technology acceptance model. Information Systems Research, 11, 342–365.
VENKATESH, V. and DAVIS, F.D., 2000. A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management science, 46(2), pp. 186-204.
Venkatesh, V. and Bala, H. (2008) “Technology Acceptance Model 3 and a Research Agenda on Interventions,” Decision Sciences, 39(2), pp. 273–315. doi: 10.1111/j.1540-5915.2008.00192.x.
Verhoef P.C., P.K. Kannan & J.J. Inman, 2015, "From Multi-Channel Retailing to Omni-Channel Retailing. Introduction to the Special Issue on Multi-Channel Retailing". Journal of Retailing, Vol. 91 Issue 2, pp. 174-181.
VESANEN, J., 2007. What is personalization? A conceptual framework. European Journal of Marketing, 41(5), pp. 409-418.
Wang, W.T., Li, H.M., 2012. Factors influencing mobile services adoption: a brand-equity perspective. Internet Res. 22 (2), 142–179.
Wang R.J.H, Krishnamurthi L and Malthouse E.C (2015) “On the Go: How Mobile Shopping Affects Customer Purchase Behavior,” Journal of Retailing, (2015 01 01). doi: 10.1016/j.jretai.2015.01.002.
Wingfield, N., 2018-last update, Inside Amazon Go, a Store of the Future. Available:
https://www.nytimes.com/2018/01/21/technology/inside-amazon-go-a-store-of-the-future.html.
Wu, I.-L., Li, J.-Y. and Fu, C.-Y. (2011), “The adoption of mobile healthcare by hospital’s professionals: an integrative perspective”, DecisionSupportSystems,Vol.51No.3,pp.587-596.
ZHANG, J. and WEDEL, M., 2009. The Effectiveness of Customized Promotions in Online and Offline Stores. Journal of Marketing Research, 46(2), pp. 190-206.
Zhao, J., Fang, S. and Jin, P. (2018) ‘Modeling and Quantifying User Acceptance of Personalized Business Modes Based on TAM, Trust and Attitude’, Sustainability. MDPI AG, 10(2), p. 356. doi: 10.3390/su10020356.
34
APPENDICES
Appendix 1 - Model
Appendix 2 – Composite reliability
