The perfect fit: the role of shopping mode and ad type for mobile ad effectiveness

The perfect fit: the role of

shopping mode and ad type

for mobile ad effectiveness

Celine Buri

Faculty of Economics and Business

2020 Dr. J. (Hans) Berger: j.berger@rug.nl

The perfect fit: the role of

shopping mode and ad type

for mobile ad effectiveness

Master’s thesis

Dr. J (Hans) Berger: j.berger@rug.nl 2nd _supervisor:

Dr. O.K. (Outi) Lundahl: o.k.lundahl@rug.nl

Management summary

This study contributes to the existing literature of mobile in-store shopper marketing by investigating whether there are differences in consumer’s shopping motivation on the preference of ad type and ultimately on purchase intention depending on the perceived regulatory fit.

A positive connection between regulatory fit and ad type on purchase intention was confirmed, examining the results of questioning 190 participants. However, a definite relationship between shopping motivation, ad type, and regulatory fit could not be provided.

Regulatory fit significantly influences purchase intention, but utilitarian shoppers seem to prefer neither pull nor push messages.

The hypothesis that hedonic shoppers perceive similarly high levels of regulatory fit in push- and pull-based advertising could not be verified. However, some indications of this relationship were found. Additional research remains to be done on this topic.

Furthermore, regulatory fit fully mediates the effect of shopping motivation on purchase intention for hedonic shoppers but not for utilitarian shoppers.

Examining and differentiating between different consumer groups is becoming more vital with the rise of personalization. Not all consumers like to receive push messages during their shopping trip. Generally, pull messages were perceived as better likeable than push messages across all shopping motivations and regulatory focus types.

Keywords: mobile advertising, in-store marketing, ad type, pull messages, push

Preface

I wrote this thesis as part of my master’s degree in Marketing Management at the Faculty of Economics & Business of the University of Groningen. In-store shopper marketing has become more and more relevant in recent years, which is especially true regarding the rise of new technologies like apps, automatic check-out, or face-scanning to increase

convenience and personalization. A guest lecture in the course ‘Retail & Omnichannel Marketing’ further sparked my interest to dive deeper into the topic. This thesis provided me with the opportunity to deepen my knowledge in what the future of retail will look like. I hope you, as a reader will enjoy reading and learning about this as much as I did.

Furthermore, I would like to thank Dr J. Hans Berger for his support, input, and

understanding during the whole conceptualizing and writing process of my thesis in this exceptional situation.

Table of content

Management summary ... III Preface ... IV Table of content ... V List of abbreviations ... VII Tables ... VIII Figures ... IX

1. Introduction ... 1

2. Literature review and theoretical framework ... 5

2.1. Regulatory focus ... 7

2.2. Shopping mode and regulatory fit ... 8

2.3. Push-based and pull-based advertising ... 9

3. Research design ... 12

3.1. Procedure ... 12

3.2. Measures ... 14

3.3. Plan of analysis ... 15

4. Results ... 17

4.1. Data cleaning and demographics ... 17

4.2. Factor and reliability analysis ... 18

4.3. Correlation check ... 21

4.4. Regulatory fit and purchase intention ... 23

4.5. Shopping mode, ad type and regulatory fit ... 23

4.6. Regulatory focus, fit and ad type ... 25

4.7. Mediating effect of regulatory fit ... 26

4.8. Discussion ... 26

5.1. Recommendations for theory and practice ... 29

5.2. Limitations and directions for future research ... 30

References ... 31

Appendix A ... 37

List of abbreviations

et al.: et alii/et aliae (and others)

LBA: location-based advertising

sm: shopping motivation

promo_focus: promotion focus

prevent_focus: prevention focus

regfit: regulatory fit

Tables

Table 1: Constructs ... 15

Table 2: Quota ... 18

Table 3: Dimensions ... 18

Table 4: Correlation matrix ... 21

Table 5: Results moderated regression ... 24

Figures

1. Introduction

92% of the population in the Netherlands and 75% of the population in Germany owns a

smartphone in 2020 (O’Dea 2020a; O’Dea 2020b).Mobile advertising spending is

expected to reach $280 million worldwide in 2022. In 2019, the market already amounted to $190 million (Guttmann 2019). The new channel allows marketers to reach consumers differently. Nowadays, nearly every person owns a smartphone and carries it with them throughout the day. Consumers become accessible for companies more frequently and in every situation. They not only text or call with their mobile device but also browse on the Internet or use the apps on their phone. This increase in touchpoints offers a better

targeting of personal consumer needs. Companies can utilize this behaviour and advertise in apps, mobile internet, or via text. With that, highly personalized ads can increase consumer’s purchase intention. Especially targeting consumers at their point of purchase has proven to be a very compelling tactic for retailers (Baxendale, Macdonald, & Wilson 2015). Therefore, it is crucial to obtain more insights into how to maximize in-store mobile advertising effectiveness.

Mobile advertising also offers marketers the possibility to gain more information about consumer’s pain points and needs, since they retrieve valuable data. The business model of online giants like Facebook and Google consists of precisely this information. They are visited by millions of users every day and have access to a vast amount of information about their customers. Facebook states it had over 2.6 billion active monthly users in the first quarter of 2020 (Clement 2020). These companies understood very early that having access to this amount of consumer data is a highly valuable and demanded product. The public painfully learned this during the now-famous Cambridge Analytica scandal, where the consulting firm became known for acquiring Facebook data of millions of users for political advertising and targeting (Confessore 2018).

In the perspective of the subsequent rise of suspicions towards mobile advertising and their intrusiveness, it becomes vital to persuade consumers of its benefits to increase ad

acceptance. One way to achieve this is ad personalization (Tong, Luo, & Xu 2020). Providing content specifically designed for the customers generally increases product liking and purchase intention. However, personalization could have the opposite effect and decrease ad acceptance, when they fail to disclose any information about the data

needs to be balanced by retailers to achieve a successful strategy. Trust and transparency can soften the blow (Xu et al. 2012).

In the last years, the trend of personalized mobile push-coupons became extremely popular among retailers, and its effectiveness in mobile in-store marketing is researched by a vast majority of the existing literature (e.g. Andrews et al. 2016; Fong, Fang, & Luo 2015; Li et al. 2017; Luo et al. 2014; Hui et al. 2013). Push-coupons or push messages are sent to the consumer without them actively asking for it. However, pull-based advertising also

presents a potentially successful approach in this context. This advertising type needs to be actively searched for by the consumer, for example, with an app. Compared to its

alternative, it elicits a higher perceived value and does not risk consumers getting annoyed of constantly receiving notifications that might not even fit their shopping trip goal (Shieh, Xu, & Ling 2019; Unni & Harmon 2007). Looking at the differentiated effect of using pull- or push-based advertising for different consumers could create promising

implications for retailers aiming to maximize the effectiveness of their personalized mobile advertising strategies.

To the author’s knowledge, to this point, there exists only one relevant paper investigating the differentiated effect of push versus pull messages in location-based advertising, despite the call-to-action from Grewal et al. (2016) to investigate this further in their central paper. Shieh, Xu, and Ling (2019) proved that people generally prefer pull-based over push-based advertising. The relationship is further moderated by time consciousness. These findings suggest that people under time pressure are more annoyed when they receive push-notifications.

Moreover, the existing literature focuses mainly on location-based advertising (LBA). LBA uses consumer’s GPS information for better targeting.

New potential influence factors on its effectiveness, like crowdedness (Andrews et al. 2016) or weather (Li et al. 2017) are examined rather than linking in-store mobile advertising with well-founded psychological constructs.

want to achieve positive outcomes and develop themselves, whereas consumers with a prevention focus tend to avoid risks, losses and negative results (Higgins 1997).

Subsequently, Das and colleagues established that consumers with a prevention goal prefer utilitarian products and the ones with a promotion focus prefer hedonic products. It is striking that the characteristics of a prevention focus seem comparable with pull-based advertising and the ones of promotion focus with push-based advertising. In more detail, pull-based advertising attracts consumers who actively search for information and is therefore perceived as more personalized. This group wants to conclude their shopping trip as efficient as possible which is why push messages can interrupt their shopping process and pose an unwelcome distraction for consumers with a prevention focus (Shieh et al. 2019). Based on this assumed connection, it would be interesting to see whether consumers with different shopping orientation prefer different kinds of ads.

This idea can also be connected to the increasingly important topic of personalized mobile marketing strategies. Utilitarian shoppers are more difficult to distract from their shopping goals than hedonic shoppers. Therefore, a personalized mobile marketing strategy is crucial to target this group (Tong et al. 2020).

Combining the highly relevant mobile advertising topic with a well-founded psychological construct like regulatory fit theory will lead to more general and applicable insights for marketers. Furthermore, it creates a better foundation for understanding consumer behaviour in the marketplace and can diminish consumer’s hesitation towards mobile advertising.

To broaden the literature on this topic and to provide more insights into when marketers should use pull or push ads to elicit purchase intention, this study is the first one to combine shopping mode, regulatory fit, and ad type. This thesis draws on regulatory fit theory to evaluate the effectiveness of different ad types and shopping modes on purchase intention.

• Does the type of ad used in mobile advertising influence purchase intention? • Does a regulatory fit between shopping motivation and ad type boost purchase

intention?

• Is the perceived regulatory fit for certain combinations of shopping motivation and ad type bigger than for others?

2. Literature review and theoretical framework

Researchers examined various drivers of the effectiveness of mobile advertising, such as context and consumer variables, market factors as well as ad goals, ad elements, and desired outcomes (Grewal et al. 2016).

One extensive research stream looks into environmental impact factors on push-based advertising for LBA. When consumers have to move across longer distances within a store for a coupon, it increases unplanned buying (Hui, Xu, & Ling 2013). Time is another relevant factor, proving that location-based advertising is more effective when consumers receive messages close to the store and when they plan to purchase (Danaher et al. 2015; Luo et al. 2014). Moreover, negative surroundings like crowdedness in subways increase the effectiveness of mobile advertising (Andrews et al. 2016). The authors argue that consumers see the ad as a welcomed distraction and stress relief measure from their negative environment. Additionally, weather influences the responsiveness on promotion messages. Generally, people used coupons more often and faster when the weather was better than when it was rainy and grey (Li et al. 2017). Going one step further, marketers can predict consumer’s behaviour by analyzing their routes within a shopping trip by individually examining the path they cover (Ghose, Li, & Liu 2019).

Another research stream investigates the role of technology. The ad effectiveness partially depends on the viewing device and whether the ad is watched on a first or a second screen. For example, Liaukonyte, Teixeira, and Wilbur (2015) used an extensive data set to prove that watching a TV advertisement while simultaneously shopping online can increase online traffic and purchases. Prior research argued that through cognitive depletion of the first task, multitasking harms mobile ad effectiveness (Segijn, Xiong, & Duff 2019). However, a rich amount of papers suggests this effect is weakened under certain circumstances, for example, when the brand is familiar, or screen goals overlap (Segijn & Eisend 2019).

for papers that investigate the differential effect of push-based and pull-based advertising. However, while analyzing the existing literature about in-store mobile advertising, it becomes clear that the main focus lays on push-based mobile promotions. Little literature investigates the effect of pull-based advertising on its own or in comparison to push-based advertising (Molitor et al. 2020; Shieh, Xu, & Ling 2019; Zubcsek, Katona, & Sarvary 2017). Additionally, although previous literature addresses interesting topics like environmental and technological impact factors with high-quality research, many topics are not tangible for marketers, as they relate to particular circumstances. High

specialization in these topics leads to a scarcity of papers that connect recent developments in technology and marketing to already existing theories and concepts

(Lamberton & Stephen 2016).

One of the most basic and commonly used consumer differentiations in research is the one between hedonic and utilitarian shopping motives (Babin, Dardin, & Griffin 1994).

Research proves that there is a positive relationship between consumer motivation and product features (Vieira, Santini, & Araujo 2018). A bundle of researchers confirmed that this relationship could be translated into regulatory fit theory, where consumers with hedonic motives have a promotion focus and are more likely to choose hedonic products, and consumers with utilitarian motives shop with a prevention focus and choose utilitarian products (Das et al. 2018; Khajehzadeh et al. 2014; Thongpapanl et al. 2018). Based on this idea, it might be appealing to see if this relationship also holds for different ad types, as suggested by Lu, Wu, and Hsiao (2019). Consumers with a promotion focus might prefer push-based advertising because it offers additional opportunities to achieve their shopping goal (Shieh et al. 2019).

In contrast, consumers with a prevention focus place much emphasis on concluding their shopping trip as efficient as possible. Receiving push messages are possible distractions from this shopping goal that steals valuable time and thus could be perceived negatively. Pull-based advertising does not distract them from their initial shopping goal, and because consumers actively search for it, it is perceived as the more personalized option

(Shieh et al. 2019). Hence, consumers with a prevention focus might prefer pull-based advertising.

different ad types in relationship with the well-established regulatory fit theory on purchase intention.

2.1. Regulatory focus

Regulatory focus theory distinguishes two approaches: prevention and promotion focus (Crowe & Higgins 1997). On the one hand, promotion focus describes the need to thrive for accomplishment, self-fulfilment, and the urge to approach the desired goal most effectively. On the other hand, prevention focus describes the desire for security and consequentially avoiding pain, loss, or other negative experiences. Regulatory fit occurs when the regulatory focus of a consumer matches with his strategy to achieve a goal (Motyka et al. 2014). This strategy consists of either eagerness for promotion focus or vigilance for prevention focus (Khajehzadeh et al. 2014; Motyka et al. 2014). Consumers who use eagerness (promotion focus) to achieve a goal are more open to other possibilities and opportunities that could be helpful along the way (Förster & Higgins 2005;

Pham & Avnat 2004). The exact opposite happens when consumers use vigilance

(prevention focus) to achieve a goal. Because they are afraid of potential risks that could avert them from reaching their aim, this group tends to avoid other opportunities at any cost (Förster & Higgins 2005; Herzenstein, Posavac, & Barkus 2007). A regulatory fit increases engagement and therefore supports and assures consumers into feeling confident

in their decision-making process (Higgins 2005).

Promotion focus indirectly translates into hedonic values and products, while prevention focus translates into utilitarian values and products (Arnold & Reynolds 2009; Das, Mukherjee, & Smith 2018). Thus, it is only logical that promotion focus is related to hedonic shopping motivations and prevention focus is related to utilitarian shopping motivations (Khajehzadeh et al. 2014; Thongpapanl et al. 2018). When examining mobile advertising attributes, regulatory fit positively influences ad attitude in mobile advertising (Kim 2019).

self-induced primes or situational ones, like making the participants write essays or mention their aspirations and duties (e.g. Haws, Dholakia, & Bearden 2010; Motyka et al. 2014). In this thesis, regulatory focus will be measured as a chronic state. This decision is based on the scope and limited time of this thesis, which does not allow for an extensive

experimental study, where different participant groups are distinctively primed. Moreover, a chronic focus allows the use of standardized and established measures, granting a better chance for external validity.

2.2. Shopping mode and regulatory fit

As already established earlier, literature proves that there is a strong connection between shopping mode and regulatory focus, to be more specific between hedonic/utilitarian shopping motives and promotion/prevention focus. Thus, in the following chapters, the focal point will be the differentiation between hedonic and utilitarian shopping motivation, inferring that they go hand in hand with promotion and prevention focus, respectively. The effect of different shopping modes on ad effectiveness is crucial for effective

targeting. However, research on this domain remains scarce, as researchers predominately focus on environmental and contextual factors (e.g. Andrews et al. 2016; Li et al. 2017). Although these topics need to be researched and offer insights into consumer behaviour in the modern world of big data, the focus on the consumer and the individual and

psychological aspects should not get lost.

Hedonic shopping motivations promote impulse buying since consumers stay in stores longer (Yim et al. 2014). Utilitarian shoppers, on the other hand, are more goal-related and only respond to a mobile promotion that fits their shopping goals.

Regulatory fit describes the process of consumers preferring outcomes or products that match with their motivation and regulatory focus. As an example, prevention-focused consumers enjoy tasks more when they have to resist temptations because it matches their regulatory focus of avoiding adverse outcomes (Freitas, Liberman, & Higgins 2003). In a cross-country analysis, the fit between regulatory orientation and shopping motivation significantly influenced the perceived value and trust of mobile retailing

product with the regulatory goal pursued by the consumer increases purchase intention and vice versa (Labroo & Lee 2006; Das et al. 2018). One could assume that this also means that a regulatory fit with ad type increases purchase intention. Hence, the following hypothesis will be tested:

H1: Regulatory fit positively influences purchase intention.

2.3. Push-based and pull-based advertising

In general, two ad types are differentiated. On the one hand, push-based or direct

advertising focuses on promotion, “pushing” the product to the consumers, for example, through special offer notifications or sending ads via text to a consumer’s phone. Pull-based advertising, on the other hand, “pulls” the consumer towards the product, either through apps or browsing the Internet.

Extensive research exists about the antecedents and drivers for successful push-based mobile advertising. Cognitive and affective aspects, more specifically involvement and interactivity of a product, both influence consumer behaviour in mobile push-based advertising. However, this should be expanded in terms of hedonic and utilitarian focus as well (Lu, Wu, & Hsiao 2019). The extent of redeemed mobile coupons increases with proximity to the store as well as with how early in the week or day the consumer receives the coupon (Danaher et al. 2015; Luo et al. 2014). Push-based advertising struggles with the personalization-privacy paradox, which can be lowered by focusing on ad relevancy and provided value (Edwards, Li, & Lee 2002; Inman & Nikolova 2017).

Literature comparing the effect of push- versus pull-based mobile advertising in this context remains scarce, although the importance of studying this aspect of mobile

advertising is stressed in various papers (Grewal et al. 2016; Tong et al. 2020). Consumers generally seem to prefer pull ads over push ads, especially if they are under time pressure (Shieh et al. 2019).

Receiving push messages would interrupt their goal pursuit and steal their time. It also possesses a higher privacy risk (Unni & Harmon 2007; Xu et al. 2011).

Pull-based advertising, on the other hand, is more personalized to the individual and thus seems more efficient and less intrusive. The perceived privacy intrusion is smaller if the consumer willingly accesses the information (Shieh et al. 2019; Unni & Harmon 2007). Hence consumers with a prevention focus might prefer pull-based advertising over push-based advertising. Because it is already proven in the literature that hedonic shopping motivation and promotion focus, as well as utilitarian shopping motivation and prevention focus, align with each other, it is justified to assume that these connections are also valid for different ad types (Khajehzadeh et al. 2014). Derived from this line of reasoning, the following hypothesis is extracted:

H2a: Utilitarian shoppers perceive more regulatory fit in pull-based advertising than in push-based advertising.

Push-based advertising is more likely to induce impulse buying, which it has in common with hedonic shopping motivation and therefore also promotion focus (Shieh et al. 2019; Xu et al. 2011; Yim et al. 2014). Hedonic shoppers are generally more open during their shopping trip than utilitarian shoppers. Without a specific goal, they also seem to be more open towards different kinds of product promotions. Because of that, they respond to both hedonic and utilitarian products, which implies that they could also be more receptive towards both push and pull messages. Hence, regulatory fit plays a less important role for hedonic shopping motives than for utilitarian ones and with that for consumers with a promotion focus (Khajehzadeh et al. 2014). Based on this, the following can be derived:

H2b: Hedonic shoppers perceive similar high levels of regulatory fit in push- and pull-based advertising.

hedonic shoppers are more open during their shopping trip and therefore would respond to both hedonic and utilitarian products as well as pull and push messages. A regulatory fit is not as essential for them (Khajehzadeh et al. 2014). Therefore, the following hypothesis is derived and visualized in Figure 1:

H3: Regulatory fit mediates the effect of shopping motivation on purchase intention for utilitarian shoppers but not for hedonic shoppers.

Figure 1: Conceptual model:

Shopping motivation:

hedonic vs. utilitarian Regulatory fit

3. Research design

The research was conducted in a mixed factorial design. Due to time and resource constraints, the data was collected with the help of an online survey.

Data was collected in the Netherlands and Germany, resulting in a sample size of 200. After data cleaning, this amounted to usable data of 190 participants.

For an optimal result, sampling should be completely random. However, due to the constraints of this research and the choice of an online survey as a means, this would not have been feasible. Therefore, the author relied on the next best alternative, convenience sampling. The data was obtained through the personal network of the author in the Netherlands and Germany as well as the platforms SurveySwap and SurveyCircle. As participants could complete the survey in either English or German, all questions were translated into German. In this process, spelling and grammar were checked to ensure the questions measure the same on both languages. Additionally, a test round was run among Germans to check whether the formulations feel natural.

3.1. Procedure

First, a short introduction to the survey was provided, including the purpose of the study and a brief description of what LBA means. Participants were also ensured that their answers are treated anonymized to increase response accuracy. To set the stage, the attendants were told that the study aimed to find the best mobile ad design for the shown product. In the next step, participants were supposed to imagine that they are shopping in their local grocery store. Every person was randomly assigned to see either a pull-based mobile advertisement (scenario one) or a push-based mobile advertisement (scenario two), which they were then asked to evaluate.

In scenario two, participants imagined that they are on their way to their local supermarket. Because they are in a designated radius of the supermarket, they automatically receive a message about current deals. The message is shown to the participants. A more detailed description of the scenarios can be found in Appendix A.

At the end of the survey, the participants were required to complete a questionnaire in order to sort them into either group 1 (hedonic shopping motivation) or group 2 (utilitarian shopping motivation). The questions regarding their shopping motive were intentionally put at the end of the survey to ensure no distortion due to social desirability. Furthermore, items checking regulatory focus were included in the questionnaire to control for

correlations between regulatory focus and shopping motivation. A set of questions measured purchase intention after receiving the mobile ad. At the end of the survey, demographics like age, gender, education level, and occupation were retrieved. The constructs shopping motivation, promotion focus, regulatory fit, and ad type were

measured using a 5-point Likert scale ranging from 1 (“strongly disagree) to 5 (“strongly agree”). Purchase intention was measured similarly, but one item was rescaled upside down for an attention check and to minimize annoyance with the multi-trait approach. All constructs of the model are multi-item scaled.

Supermarkets were chosen as a target location because deals are a commonly used

technique in this field, that has already been translated and analyzed in the mobile context in various literature (e.g. Grewal et al., 2018). Moreover, supermarkets represent a familiar location for almost every person. It is safe to say that the supermarket was more likely visited in the last three months than any other retailer, particularly in cases where other stores were closed due to the pandemic.

Product type is a variable, which is proven to influence regulatory fit

(e.g. Khajehzadeh et al. 2014). Product type effects are controlled through randomly assigning every participant to either a hedonic or utilitarian product in each scenario. Chocolate, wine, and ice cream are products classically associated as hedonic (Crowley, Spangenberg, & Hughes 1992; Kuikka & Laukkanen 2012) whereas kitchen paper, detergent, and deodorant spray are more often defined as utilitarian products

stock purchases in the corona crisis. However, it is essential to note that all product categories possess both hedonic and utilitarian attributes and are examined in the supermarket context, which is mostly utilitarian. Regarding this, a higher hedonic

assessment than usual is also expected because for the months’ March and April 2020, the supermarket visit was the only distraction from everyday life for the majority of the population, especially in Germany. Product prices were retrieved from reference prices online.

3.2. Measures

The wording of the questionnaire for regulatory fit is based on Khajehzadeh et al. (2014), who also investigated regulatory fit of shopping motivation and mobile coupons. The items are directed towards the liking of the encountered ad message.

There are numerous scales to measure regulatory focus. While testing for reliability, validity, and generalizability of the most commonly used ones, researchers concluded that the RFQ (regulatory fit questionnaire) is the most universally applicable measure.

However, it lacks affective components as well as items that are directed to the present or future. Based on this, the composite regulatory focus scale was developed by Haws, Dholakia, and Bearden (2010). They derived five items testing for promotion focus and five for prevention focus that have proven their effectiveness in other papers

(Das et al. 2018; Haws et al. 2010).

The 3-item scales for both pull-based and push-based ads were adopted from Shieh et al. (2019).

Shopping motivation is also tested to check whether it translates into regulatory focus. The scale is retrieved from established literature (Babin, Darden, & Griffin 1994;

Khajehzadeh et al. 2014).

Table 1: Constructs

variables dimensions literature

Shopping motivation Hedonic motives

Utilitarian motives

Babin, Darden, & Griffin (1994); Khajehzadeh et al. (2014)

Regulatory focus Promotion focus

Prevention focus

Haws, Dholakia, & Bearden (2010);

Haws et al. (2010); Das et al. (2018).

Regulatory fit Regulatory fit Khajehzadeh et al. (2014)

Ad type Pull-based

Push-based

Shieh et al. (2019)

Purchase intention Purchase intention Mukherjee, Jha, & Smith (2017)

Age, gender, education level, and occupation were included in the questionnaire for descriptive purposes.

3.3. Plan of analysis

Firstly, a confirmatory factor analysis is performed to ensure all items used from previous research are relevant in this study. After that, a reliability analysis is conducted to measure the internal consistency of the dimensions obtained in the previous step. Then, descriptive figures and correlations are examined. As a next step, it is controlled whether shopping mode translates into regulatory focus in this context with the help of a correlation matrix. Afterwards, the remaining hypotheses are tested.

All constructs from the conceptual model are collected with the help of a Likert Scale and can, therefore, be interpreted as continuous variables, although naturally being ordinal-scaled.

Thus, linear regression with a 90% confidence interval can be used to identify the strengths of the effects of the independent variables on the dependent variables in the model and the derived hypotheses.

4. Results

Firstly, issues of validity and representativeness are discussed in this chapter. After that, the data cleaning process is described, and the demographic data is analyzed.

Subsequently, a confirmatory factor analysis and reliability test are conducted, resulting in the constructs used in the further steps. Next, it is controlled whether the shopping mode and regulatory focus constructs translate into each other, as suggested in previous literature (Khajehzadeh et al. 2014; Thongpapanl et al. 2018). After this, a linear regression

including a moderation and mediation analysis is conducted to test the hypotheses derived in the conceptual model.

Internal validity was improved through randomization of both the questionnaire and the assignment of the extraneous variable product type. Dividing participants into groups ensured that no internal correlation between the questions would appear. However, it might be possible that the answers concerning the construct push-based advertising were

influenced by privacy concerns, which was not controlled for (Shieh et al. 2019).

Moreover, the study was conducted in a time where the perception of hedonic or utilitarian constructs might be biased due to the outbreak of the coronavirus. Out of 200 respondents, ten were excluded because of missing responses that were crucial for the analysis. The reason for that was most probably not the research design but technical problems that arose at the start of the collection process. Because the credit link for SurveySwap users did not work, participants stopped the survey early.

The data collection process lasted two weeks. External validity was improved by making sure that the tested scenarios were as realistic as possible. However, one cannot deny a selection bias in the study, as most of the participants were found through the personal network of the researcher.

4.1. Data cleaning and demographics

old (45%), representing students, another 30% were in the age group of 45-64 and mostly employed for wages. An overview of the quota can be found in Table 2.

Table 2: Quota

Quota (%) Gender

Age Male Female Total

12-17 years 0 (0,0) 2 (1,1) 2 (1,1) 18-24 years 12 (6,3) 42 (22,1) 54 (28,4) 25-34 years 19 (10.0) 15 (7,9) 34 (17,9) 35-44 years 10 (5,3) 5 (2,6) 15 (7,9) 45-55 years 14 ( 7,4) 15 (7,9) 29 (15,3) 55-64 years 21 (11,1) 10 (5,3) 31 (16,3) 65-74 years 19 (10,0) 2 (1,1) 21 (11,1) 75+ 4 (2,1) 0 (0,0) 4 (2,1) Total 99 (52,1) 91 (47,9) 190 (100,0)

The vast majority of participants holds at least a bachelor’s degree (67%). Almost half of the respondents are employed (42%). Striking when looking at the crosstabulation of gender and level of education is that the percentage of males holding a master or doctorate (33%) is double than that of females (16%). All demographic results can be found in Appendix B.

4.2. Factor and reliability analysis

The items used in the survey are derived from literature. However, it is necessary to check whether they hold in this study. For this purpose, a confirmatory factor analysis was conducted. After that, a reliability analysis was performed to measure the internal consistency of the dimensions. An overview of the results is depicted in Table 3. Further details can be found in Appendix B.

Table 3: Dimensions

Items Cronbach’s a KMO

Hedonic sm 3 .795 .747 Utilitarian sm 2 .652 .747 Promotion focus 4 .689 .703 Prevention focus 2 .681 .703 Push ad 2 .396 .5 Pull ad 3 .715 .677 Regulatory fit 4 .903 .832 Purchase intention 2 .868 .5

Usually, a confirmatory factor analysis is performed for the entire set of eight variables in a two-step approach (e.g. Anderson & Gerbing 1988). However, in this case, not all

participants were confronted with all variables. Therefore, a correlation coefficient could not be computed for all pairs of variables and was conducted per variable, which resulted in six individual measurement models (Khajehzadeh et al. 2014; Yim et al. 2014). The first model, shopping motivation includes two variables: hedonic and utilitarian shopping motivation. The second one, regulatory focus, contains promotion and prevention focus and the rest of the variables consist of themselves, as depicted in Table 3.

The procedure reduces items with high correlations into dimensions. In order to test if the factor analysis is appropriate, correlations are tested with the Kaiser-Mayer-Olkin measure of sampling adequacy (KMO) and Bartlett’s test of sphericity. KMO needs to be higher than 0.5, there has to be proof for correlations between the items, and all communalities need to be above 0.4.

A principal component analysis (PCA) is conducted to see whether the items combine into factors. This procedure is a broadly accepted measure to determine the weights of the variables. The best number of factors can be selected based on eigenvalues (>1), factors that explain more than 5% of variance each, the total variance explained (> 60%) and the scree plot. 5% of variance will be explained by almost every factor because of the small sample size. Thus, this criterion will not be considered in this case.

In the next step, the factor matrix is rotated orthogonally using VARIMAX to minimize high cross-loadings. Then, the internal consistency or the strength to proceed with these dimensions instead of original items will be measured. In the following, the results for each construct will be discussed.

For shopping mode, two factors are distinguished: hedonic and utilitarian shopping motivation. All communalities are higher than 0.4. The eigenvalues, total variance

what I need to buy and not go to other shops” is excluded, ultimately leading to the factor consisting of two variables.

Regulatory focus is divided into promotion and prevention focus. In the analysis, items “When it comes to achieving things that are important to me, I find that I don’t perform as well as I would ideally like to do” (.363) and “Not being careful enough has gotten me into trouble at times” (.409) share a small amount of variance with the other variables and are therefore excluded from further steps. Another anomaly is described by the item “I usually obeyed rules and regulations that were established by my parents.”. The rotated factor matrix suggests three factors with the third one only holding significant loadings for this one item (.909).

Although Fabrigar et al. (1999) suggest that high communalities can indicate that a smaller sample size is sufficient, this number seems too high. This might be due to a small number of participants attributing importance to this item, which leads to outliers in the data set. The factor analysis is rerun without the item to test this assumption, which ultimately results in the best number of two factors.

Items 1-4 distinctively load on factor 1 “promotion focus” and items 5-7 on factor 2 “prevention focus” (KMO = .703, p = .00). The reliability analysis shows for promotion focus internal consistency (Cronbach’s a = .689). Prevention focus can be improved by deleting “I see myself as someone who is primarily striving to become the self I “ought” to

fulfil my duties, responsibilities and obligations.” (Cronbach’s abefore = .674;

Cronbach’s aafter = .681). The overall internal consistency for prevention focus is relatively

low. Hence, it is reasonable to exclude one more item, even if it only slightly increases the value.

delivered to me without my consent” explains the higher amount of variance and thus will stay in the analysis (62.241% > 37.529%).

Pull-based advertising is adequately measured by the items derived from previous literature; all communalities and loadings meet the necessary criteria (KMO = .677, p = .00). Moreover, the factor is internally consistent (Cronbach’s a = .715).

The same holds for the construct regulatory fit, which has a very high reliability (KMO = .832, p = .00; Cronbach’s a = .903).

For purchase intention, the variable “How likely would you be to purchase the advertised product, given the information shown?” shares a low amount of variance with the other variables (.315) leading to its exclusion. Performing the analysis again with the two remaining factors leads to small sampling adequacy (KMO = .50, p = .00). Cronbach’s a suggests a high internal consistency of the chosen items (.868).

4.3. Correlation check

Possible correlations between the independent variables are checked for with the help of a Pearson correlation coefficient and a bivariate correlation matrix. No variables correlate higher than .05 with each other, however hedonic shopping motivation and pull messages show a comparable high correlation (.409). Based on this, no extensive multicollinearity was found. Push- and pull-based advertising were measured in different groups; therefore, no correlation value exists. The results are presented in Table 4.

Table 4: Correlation matrix

Utilitarian_sm Hedonic_sm Prevent_focus Promo_focus Push Pull

Utilitairan_sm 1 -.395** -.018ns -.013ns .047ns -.294** Hedonic_sm -.395** 1 .182* .339** .102ns .409** Prevent_focus -.018ns .182* 1 .023ns -.036ns -.021ns Promo_focus -.013ns .339** .023ns 1 .066ns .320** Push .047ns .102ns -.036ns .066ns 1 na Pull -.294** .409** -.021ns .320** - 1

Note: na: not applicable; all values ≥ 0.5 must be excluded

promotion focus and utilitarian shopping motivation prevention focus

(Khajehzadeh et al. 2014; Thongpapanl et al. 2018). The correlations between the

independent variables were closely examined to make sure this is also right for this study. The results are shown in Table 4.

Hedonic shopping motivation translates into promotion focus to some extend (.339**). However, there is no significant relationship between utilitarian shopping motivation and prevention focus (-.018ns). Instead, hedonic shopping motivation also correlates with prevention focus (.182*). As suspected, the correlation of hedonic shopping motivation and promotion focus remains higher and more significant, although no correlation is higher than 0.5. These outcomes suggest that the theory of shopping motivation and ad type resulting in regulatory fit only holds for the hedonic shopping motivation path.

Another assumption of the model is that consumers with hedonic shopping motivation and promotion focus prefer push messages and consumers with utilitarian shopping motivation and prevention focus prefer pull messages. However, it is already established that there is no significant correlation between utilitarian shopping motivation and prevention focus. Therefore, it can be presumed that the relationship with push messages is also not significant. Verifying these connections is vital to determine the veracity of the derived hypotheses. An overview of the results can also be found in Table 4.

Hedonic shopping motivation has no significant effect on the attitude towards push messages (.102ns). Nevertheless, the variable correlates with pull messages. Thus, there seems to be a preference of customers with hedonic shopping motivation for these (.409**).

Utilitarian shopping motivation has no significant effect on attitude towards push messages, but significantly and negatively correlates with pull messages (-.294**). This relationship infers that consumers with utilitarian shopping motivation tend to reject pull messages. It is important to note that this effect seems to be much smaller than the effect of hedonic shopping motivation on pull messages (-.294** vs .409**).

the previous results that (a) hedonic shopping motivation translates into promotion focus and (b) consumers with a hedonic shopping motivation prefer pull messages as well. These relationships contradict H2b, which assumes that consumers with hedonic shopping

motivation perceive a similar high regulatory fit of push and pull messages.

Utilitarian shopping motivation does not seem to have any relationship with prevention focus nor ad type. Subsequently, prevention focus also does not have significant

connections.

Crucial to mention is that the factor push messages only consists of one item, which might be the reason for the lack of significant results. Because of that, these results should be interpreted with caution.

4.4. Regulatory fit and purchase intention

The construct regulatory fit significantly influences purchase intention (R2 _{= .421,}

𝛽 = .649, p = .000). Thus, with an increase of one unit of regulatory fit, purchase intention increases by 65%, which is a huge proportion. This relationship proves H1, regulatory fit increases purchase intention, to be true. Furthermore, because regulatory fit refers back to the ad type, this outcome indicates that the type of ad chosen indirectly influences purchase intention.

4.5. Shopping mode, ad type and regulatory fit

Table 5: Results moderated regression R2 _R2_{adjusted 𝜷 𝜷} std p VIF Hedonic_sm .194*** .168*** -.152ns -.158ns .400 4.057 Push messages “ “ .197 .233 .014 1.012 Hedonic_push_interaction “ “ .195 .480 .012 4.063 Hedonic_sm .399*** .379*** .351 .335 .000 1.238 Pull messages “ “ .438 .442 .000 1.450 Hedonic_pull_interaction “ “ .152ns .146ns .110 1.208 Utilitarian_sm .420*** .150*** .344ns .351ns .109 5.304 Push messages “ “ .385 .386 .000 1.018 Utilitarian_push_interaction “ “ -.185 -.481 .029 5.341 Utilitarian_sm .275*** .251*** -.045ns -.044ns .915 1.117 Pull messages “ “ .503 .507 .000 1.210 Utilitairan_pull_interaction “ “ .010ns .010ns .642 1.108 DV: regulatory fit

The Variance Inflation Factor is a popular tool to assess multicollinearity. Several rules of thumb exist in the literature, ranging from 4 to 10, 20 or even higher (O’Brien 2007). In the regression analysis, higher values of VIF can be found for the models related to push messages. The highest value for the interaction effect of utilitarian shopping motivation and push messages is 5.341. There is no consensus about a substantial threshold value. However, it is pointed out that the interpretability depends on other aspects that influence this relationship (O’Brien 2007). No high cross-correlations for push messages on other variables can be detected when looking at the correlation matrix. Moreover, the variable push messages consists of only one item, so further combining correlating factors is not possible. Thus, the moderate multicollinearity is acceptable.

For regulatory fit, the main effect of hedonic shopping motivation is not significant,

however, push messages and the interaction effect are (𝛽std=.233** and 𝛽std= .480**).

Pull messages do not moderate the relationship between hedonic shopping motivation and

regulatory fit. However, the main effects play an important role (𝛽std = .335, p = .000 for

hedonic shopping motivation; 𝛽std = .442, p = .000 for pull messages).

Push messages moderate the relationship between utilitarian shopping motivation and

regulatory fit in a negative way (𝛽std = -.481, p = .029). This outcome suggests that

consumers with a utilitarian shopping motivation want to avoid push messages and that there is no regulatory fit between these two factors. The positive interaction effect between hedonic shopping motivation and push messages further strengthens this interpretation.

However, there is no overall effect of utilitarian shopping motivation, but just a crossover

interaction. Just as with hedonic shopping motivation, push messages have a significant main effect on regulatory fit (𝛽std = .386, p = .000).

For utilitarian shopping motivation and pull messages, no interaction exists and therefore also no regulatory fit. However, pull messages also have a significant main effect in this

relationship (𝛽std = .507, p = .000).

4.6. Regulatory focus, fit and ad type

Separately, the same analysis was also conducted with promotion and prevention focus to check whether there are discrepancies between the results of shopping motivation and regulatory focus. The results of this analysis can be found in Appendix B.

Interestingly, the only significant interaction effect was between promotion focus and pull messages. However, hedonic shopping motivation holds an interaction effect with push messages and translates into promotion focus at the same time (𝛽std = .203, p = .038). One

reason for this might be that hedonic shopping motivation influences both the preference for push and pull messages, which would also cover the reasoning in the literature part. The interaction effect between hedonic shopping motivation and pull messages is just slightly insignificant but has a similar Beta than the interaction with push messages

(𝛽std = .146, p = .110). This relationship could become significant with a bigger data set, an

4.7. Mediating effect of regulatory fit

Regulatory fit could influence the effect of shopping motivation on purchase intention and thus act as a mediator. In order to test this, a mediation analysis was conducted using the Hayes’ PROCESS macro. It investigates the total effect of shopping mode on purchase intention (c), the effect of shopping mode on regulatory fit (a), the effect of regulatory fit on purchase intention (b) and the conditional direct effect of shopping mode on purchase intention (c’). According to Hayes (2017), c, a and b need to be significant and including regulatory fit as a mediator has to make c’ either insignificant or substantially lower than c. The results can be found in Table 6.

Table 6: Mediation effect size

a b c c’ ab

Hedonic_sm .3694*** .6370*** . 2672*** .0319ns .2353***

Utilitarian_sm -.1140ns .6358*** -.1864*** -.1139*** -.0725ns

The confidence interval does not include zero; therefore, regulatory fit fully mediates the relationship between hedonic shopping motivation and purchase intention. The total effect is .2672.

For utilitarian shopping motivation, regulatory fit does not meditate the relationship with the total effect on purchase intention being -.1864. The whole mediation analysis can be found in Appendix B.

4.8. Discussion

H1: Regulatory fit increases purchase intention.

This hypothesis was confirmed. One unit increase of regulatory fit heightens purchase intention by 65%.

H2a: Utilitarian shoppers perceive more regulatory fit in pull-based advertising than in push-based advertising:

Push messages negatively influence regulatory fit, and for pull messages, there exists only a positive main effect, no interaction effect with utilitarian shopping motivation. No proof of a positive regulatory fit between utilitarian shoppers and any other construct was found. H2b: Hedonic shoppers perceive similarly high levels of regulatory fit in push- and pull-based advertising.

The hypothesis was not verified. However, a few signs hint on the established hypothesis holding some truth. Regulatory fit increases purchase intention, and because the construct regulatory fit was measured with items asking for the liking of the encountered ad

messages, a connection to the ad type prevails. Hedonic shopping motivation translates into both promotion- and prevention focus, and there is a positive interaction effect

between both hedonic shopping motivation and push messages as well as promotion focus and pull messages. Nevertheless, the interaction of hedonic shopping motivation itself and pull messages remains insignificant. The same is true for the interaction effect of

promotion focus and push messages.

Overall, pull messages have a more significant effect on regulatory fit and thus purchase intention than push messages. These results align with previous literature, stating that pull messages lead to a higher perceived value than its counterpart (Shieh et al. 2019;

Unni & Harmon 2007).

H3: Regulatory fit mediates the effect of shopping motivation on purchase intention for utilitarian shoppers but not for hedonic shoppers.

5. Conclusions and recommendations

The purpose of this study was to examine whether a regulatory fit between shopping motivation and ad type positively influences consumer’s purchase intention.

To summarize, no sustainable evidence was found for consumers with specific shopping motivations to prefer different ad types. Some indications point to purchase intention being increased through both pull- and push-based advertising for hedonic shoppers. However, no clear conclusions can be drawn here. For utilitarian shoppers, only one significant connection exists, the negative interaction effect with push messages on regulatory fit. In this context, no significant main effect was found; push messages only directly and positively influence the perceived regulatory fit of the message type. This result implies that receiving push messages when shopping with a utilitarian focus further decreases regulatory fit and ultimately purchase intention. The results of the model check suggest a similar relationship to pull messages. Having a utilitarian shopping motivation

significantly decreases the liking of pull messages.

Therefore, according to this study, utilitarian shoppers do not want to get distracted of their goal – efficiently buying the necessary items and minimizing time spent in-store- at all. They do not perceive a regulatory fit with either push- or pull-based advertising. One reason for this could also be privacy concerns that affect utilitarian shoppers more than hedonic/promotion focus shoppers (Andrews et al. 2016; Shieh et al. 2019; Van Noort, Kerkhof, & Fennis 2008).

Hedonic shoppers do not perceive similar high levels of regulatory fit in push- and pull-based advertising. However, hints suggest that this hypothesis holds some truth, like the interaction effect with push messages implying that perceived regulatory fit increases for hedonic shoppers when they receive push messages. The same holds for consumers with a promotion focus who receive pull messages. For all types of shopping motivation and regulatory focus, the main effect of pull messages on regulatory fit was higher than for push messages, indicating that pull messages have a higher value (Shieh et al. 2019; Unni & Harmon 2007).

fit between ad type and another construct boosts purchase intention. It could not be proven that this construct is in fact, shopping motivation (or regulatory focus).

Moreover, regulatory fit mediates the effect of shopping motivation on purchase intention for hedonic shoppers but not for utilitarian shoppers, which stays in contrast with the initially assumed connection. It coincides, however, with the findings that regulatory fit only seems to be essential for consumers with hedonic shopping motivation. It might be that utilitarian shoppers avoid any kind of mobile ad they could receive during their

shopping trip in their want of not getting distracted. Hedonic shoppers and consumers with promotion focus embrace all types of mobile advertising and opportunity to make the most out of their shopping trip.

Overall, it is noticeable that classically categorizing consumers became extremely difficult. They do not either have a hedonic or a utilitarian shopping motivation or even shop with a promotion or a prevention focus. Instead, consumers can switch between them, which makes a classification a challenge (Shankar et al., 2016).

5.1. Recommendations for theory and practice

Although this study did not prove a definite relationship between shopping motivation, ad type and regulatory fit, it contributes to the existing literature in confirming a positive connection between regulatory fit and ad type on purchase intention. On top of that, the study provides insights into mobile advertising from a consumer perspective and combines it with a well-founded psychological construct. Researchers should differentiate consumers more on a psychological level and acknowledge that push messages might not be

beneficial for all groups.

It is important to note for managers that not all consumers like to receive push messages during their shopping trip. Examining and differentiating between different consumer groups is becoming more and more critical. Understanding which consumers under which circumstances should be targeted with push messages or in-app is crucial for a

individual consumer data to target different consumer groups better and gain a competitive advantage.

5.2. Limitations and directions for future research

The study did not deliver proof that a connection between regulatory fit of shopping motivation and ad type (a) exists and (b) significantly increases purchase intention. However, throughout the analysis, hints of such an existing relationship, at least for hedonic shopping motivation, were found. It might be beneficial to research on a bigger scale to gain more representative and significant results.

As the dimension push messages only consists of one item, all findings related to this need to be interpreted with caution. In future research of the differences between push- and pull-based advertising, more reliable items for the push construct need to be developed.

Additionally, the study was conducted mainly in the personal environment of the researcher, which limits both generalizability and external validity of the outcomes. On top of that, the data collection took place during a two weeks’ time frame in the middle of the global coronavirus pandemic. This external influence potentially skewed the results towards an emphasis on hedonic shopping motives, which is especially true in the

supermarket research setting. Some participants of the study stated in an additional statement that the items testing for hedonic shopping motivation were mostly correct for them during the time in the pandemic when doing the groceries was the highlight and only outdoor activity of the day.

Moreover, privacy concerns are a crucial influence factor in the context of mobile advertising and should be included in future research.

References

Anderson, James C. and Gerbing, David W. (1988), “Structural equation modeling in practice: A review and recommended two-step approach,” Psychological Bulletin, 103 (3), 411-423.

Andrews, Michelle, Luo, Xueming, Fang, Zheng, and Ghose, Anindya (2016), “Mobile ad effectiveness: Hyper-contextual targeting with crowdedness,” Marketing Science, 35 (2), 218-233.

Arnold, Mark J. and Reynolds, Kristy E. (2009), “Affect and Retail Shopping Behavior: Understanding the Role of Mood Regulation and Regulatory Focus,” Journal of Retailing, 85 (3), 308-320.

Austin, Peter C. and Steyerberg, Ewout W. (2015), “The number of subjects per variable required in linear regression analyses,” Journal of Clinical Epidemiology, 68 (6), 627-636. Babin, Bary J., Darden, William R., and Griffin, Mitch (1994), “Work and/or Fun:

Measuring Hedonic and Utilitarian Shopping Value,” Journal of Consumer Research, 20 (4), 644-656.

Baxendale, Shane, Macdonald, Emma K., and Wilson, Hugh N. (2015), “The Impact of Different Touchpoints on Brand Consideration,“ Journal of Retailing, 91 (2), 235-253. Clement, J. (2020), “Number of Facebook users worldwide 2008-2020,” (accessed June 13, 2020), https://www.statista.com/statistics/264810/number-of-monthly-active-facebook-users-worldwide/.

Confessore, N. (2018, November 14), “Cambridge Analytica and facebook: The Scandal and the Fallout So Far,” (accessed June 13, 2020),

https://www.nytimes.com/2018/04/04/us/politics/cambridge-analytica-scandal-fallout.html. Crowe, Ellen and Higgins, E. Tory (1997), “Regulatory focus and strategic inclinations,” Organizational Behavior and Human Decision Processes, 69 (2), 117-132.

Danaher, Peter J., Smith, Michael S., Ranasinghe, Kulan, and Danaher, Tracey S. (2015), “Where, when, and how long: Factors that influence the redemption of mobile phone coupons,“ Journal of Marketing Research, 52 (5), 710-725.

Das, Gopal, Mukherjee, Amaradri, and Smith, Ronn J. (2018), “The Perfect Fit: The Moderating Role of Selling Cues, on Hedonic and Utilitarian Product Types,“ Journal of Retailing. 94 (2), 203-216.

Edwards, Steven, Li, Hairong, and Lee, Joo Hyun (2002), “Forced Exposure and

Psychological Reactance: Antecedents and Consequences of the Perceived Intrusiveness of Pop-Up Ads,” Journal of Advertising, 31(3), 83-95.

Fabrigar, Leandre, MacCallum, Robert C., Wegener, Duane T., and Strahan, Erin J. (1999), “Evaluating the Use of Exploratory Factor Analysis in Psychological Research,” Psychological Methods, 4 (3), 272-299.

Fong, Nathan M, Fang, Zheng, and Luo, Xueming (2015), “Geo-Conquesting: Competitive Locational Targeting of Mobile Promotions,” Journal of Marketing Research, 52, 726-735. Förster, Jens and Higgins, E. Tory (2005), “How global versus local perception fits

regulatory focus,” Psychological Science, 16 (8), 631-636.

Freitas, Antonio L., Liberman, Nira, and Higgins, E. Tory (2003), “Regulatory Fit and Resisting Temptation during Goal Pursuit,” Journal of Experimental Social Psychology, 38 (3), 291-298.

Ghose, Anindya, Li, Beibei, and Liu, Siyuan (2019), “Mobile Targeting Using Customer Trajectory Patterns,” Management Science, 65 (11), 5027-5049.

Grewal, Dhruv, Bart, Yakov, Spann, Martin, and Zubcsek, Peter Pal (2016), “Mobile Advertising: A Framework and Research Agenda,” Journal of Interactive Marketing, 34 (Special Issue), 3-14.

of Marketing Research, 44 (2), 251-260.

Higgins, E. Tory (1997), “Beyond Pleasure and. Pain. American Psychologist,” 52 (12), 1280-1300.

Higgins, E. Tory (2005), “Value from regulatory fit. Current Directions in Psychological Science,” 14 (4), 209-213.

Hui, Sam, Inman, J. Jeffrey, Huang, Yanliu, and Suher, Jacob (2013), “On unplanned spending: Applications to mobile promotion strategies,” Journal of Marketing, 77 (2), 1-16.

Inman, J. Jeffrey and Nikolova, Hristina (2017), “Shopper-Facing Retail Technology: A Retailer Adoption Decision Framework Incorporating Shopper Attitudes and Privacy Concerns, “Journal of Retailing, 93 (1), 7-28.

Khajehzadeh, Saman, Oppewal, Harmen, and Tojib, Dewi (2014), “Consumer responses to mobile coupons: The roles of shopping motivation and regulatory fit,” Journal of Business Research, 67 (11), 2447-2455.

Kim, Mikyoung (2019), “Determinants of Young Consumers’ Attitude toward Mobile Advertising: The Role of Regulatory Focus,” Journal of Promotion Management, 26 (2), 186-206.

Kuikka, Anna and Laukkanen, Tommi (2012), “Brand loyalty and the role of hedonic value,” Journal of Product & Brand Management, 21 (7), 529-537.

Labroo, Aparna A. and Lee, Angela Y. (2006), “Between Two Brands: A Goal Fluency Account of Brand Evaluation,” Journal of Marketing Research, 43 (3), 374-385.

Lamberton, Cait and Stephen, Andrew T. (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 (6), 146-172.

Li, Chenxi, Luo, Xueming, Zhang, Cheng, and Wang, Xiaoyi (2017), “Sunny, rainy, and cloudy with a chance of mobile promotion effectiveness,” Marketing Science, 36 (5), 762-779.

Liaukonyte, Jura, Teixeira, Thales, and Wilbur, Kenneth C. (2015), “Television Advertising and Online Shopping,” Marketing Science, 34 (3), 311-330.

product loyalty through mobile advertising: Affective and cognitive perspectives,” International Journal of Information Management, 47(1), 101-111.

Luo, Xueming, Andrews, Michelle, Fang, Zheng and Phang, Chee Wei (2014), “Mobile Targeting,” Management Science, 60 (7), 1738-1756.

Guttmann, A. (2019), “Mobile advertising spending worldwide from 2007 to 2022” (accessed 13 June, 2020), https://www.statista.com/statistics/303817/mobile-internet-advertising-revenue-worldwide.

Molitor, Dominik, Spann, Martin, Ghose, Anindya, and Reichhart, Philipp (2020), “Effectiveness of Location-Based Advertising and the Impact of Interface Design,” Working Paper (Available at SSRN 2645281),

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2645281.

Motyka, Scott, Grewal, Dhruv, Puccinelli, Nancy M., Roggeveen, Anne L., Avnet, Tamar, Daryanto, Ahmad, de Ruyter, Ko, and Wetzels, Martin (2014), “Regulatory fit: A meta-analytic synthesis,” Journal of Consumer Psychology, 23 (3), 394-410.

O’Brian, Robert M. (2007), “A Caution Regarding Rules of Thumb for Variance Inflation Factors,” Quality & Quantity, 41 (5), 673-690.

O’Dea, S. (2020a), “Forecast of smartphone users in the Netherlands from 2018 to 2024,” (accessed June 13, 2020), https://www.statista.com/statistics/494636/smartphone-users-in-netherlands/.

O’Dea, S. (2020b), “Smartphone users in Germany 2018-2024,” (accessed June 13, 2020), https://www.statista.com/statistics/467170/forecast-of-smartphone-users-in-germany/. Pham, Michel and Avnet, Tamar (2004), “Ideals and Oughts and the Reliance on Affect versus Substance in Persuasion,” Journal of Consumer Research, 30 (4), 503-518. Segijn, Claire M. and Eisend Martin (2019), “A Meta-Analysis into Multiscreening and Advertising Effectiveness: Direct Effects, Moderators, and Underlying Mechanisms,” Journal of Advertising, 48 (3), 313-332.

Segijn, Claire M., Xiong, Shili, and Duff Brittany R.L. (2019), “Manipulating and Measuring Media Multitasking: Implications of Previous Research and Guidelines for Future Research,” Communication Methods and Measures, 13 (2), 83–101.

and Morrissey, Shawn (2016), “Mobile Shopper Marketing: Key Issues, Current Insights, and Future Research Avenues,” Journal of Interactive Marketing, 34 (Special Issue), 37-48.

Shieh, Chih-Hui, Xu, Yingzi, and Ling, I-Ling (2019), “How location-based advertising elicits in-store purchase,” Journal of Services Marketing, 33 (4). 380-395.

Statistical Solutions (2020), “Sample Size Formula” (accessed June 13, 2020), https://www.statisticssolutions.com/sample-size-formula/.

Thongpapanl, Narongsak, Ashraf, Abdul R., Lapa, Luciano, and Venkatesh, Viswanath (2018), “Differential effects of customers' regulatory fit on trust, perceived value, and m-commerce use among developing and developed countries,” Journal of International Marketing, 26 (3), 22-44.

Tong, Siliang, Luo, Xueming, and Xu, Bo (2020), “Personalized mobile marketing strategies,” Journal of the Academy of Marketing Science, 48 (1), 64-78.

Unni, Ramaprasad and Harmon, Rpbert (2007), “Perceived Effectiveness of Push vs. Pull Mobile Location Based Advertising,” Journal of Interactive Advertising, 7 (2), 28-40. Van Noort, Guda, Kerkhof, Peter, and Fennis, Bob (2008), “The persuasiveness of online safety cues: The impact of prevention focus compatibility of Web content on consumers' risk perceptions, attitudes, and intentions,” Journal of Interactive Marketing, 22 (4), 58-72. Vieira, Valter, Santini, Fernando Oliveira, and Araujo, Clécio Falcao (2018), “A meta-analytic review of hedonic and utilitarian shopping values,” Journal of Consumer Marketing, 35 (4), 426-437.

Wan, Echo Wen, Hong, Jiewen, and Sternthal, Brian (2009), “The Effect of Regulatory Orientation and Decision Strategy on Brand Judgments,” Journal of Consumer Research. 35 (6), 1026-1038.

Xu, Heng, Luo, Xin (Robert), Carroll, John M., and Rosson, Mary Beth (2011), “The personalization privacy paradox: An exploratory study of decision making process for location-aware marketing,” Decision Support Systems, 51 (1), 42-52.

Appendix A

Figures A 1: Scenarios used in the survey 1. Push-based advertising

Hedonic product: Utilitarian product:

1. Participants presented with scenario 1:

Because they are in a designated radius of the supermarket, they receive notifications about current deals on their smartphone. 2. Mobile advertisement is shown.

3. Rate advertisement

Hedonic product: Utilitarian product:

1. Participants are told that their local supermarket has a very popular app, which they downloaded before the shopping trip.

2. Next, they see the home screen of the app, where there is the possibility to click “Offers near you”. On the next page, they are presented with a variety of current deals; they could select.

3. Rate advertisement

Table A 1: Constructs and items used

variables dimensions items literature

Shopping motivation Hedonic motives

Utilitarian motives

1. During a supermarket visit, I want to enjoy myself.

2. A supermarket visit should make me feel better.

3. During a supermarket visit, I relieve my sense of boredom. 1. During a supermarket visit, I want

to purchase only the necessary items that I need in the least amount of time.

2. While visiting a supermarket, I want to get my shopping tasks