Is there an emotional connection?
The role of emotions on the relationship between a customer
brand experience and loyalty intentions in an online purchase
customer journey
MASTER THESIS
MSc Marketing Intelligence
The role of emotions on the relationship between a
customer brand experience and loyalty intentions in an
online purchase customer journey
Talitha la Macchia
ABSTRACT
PREFACE
Dear reader,
Thank you for taking the time to go through the final project of my master program. The process of writing a thesis is challenging, though it did allow me to apply some of the
knowledge I gained during my study program as well as I learned a lot of things about myself. I have been working on this project from February till July 2018 in order to fulfill the
requirements of the Master Marketing of the University of Groningen.
There are some people I would like to thank. First of all, I would like to thank my supervisor dr. Lisette de Vries for her guidance and making me feel comfortable throughout the process in one of the most important periods of my study career. And I want to thank some my fellow study mates of which some I got to know at the beginning of this master program or activities of the MARUG with whom I build valuable friendships and spend several days and nights sharing thesis insights, tips and tricks at one the coffee corners of the university library. Wherever our paths may lead us, I am grateful for their support.
Talitha la Macchia
TABLE OF CONTENTS
1. Introduction ... 5
2. Theoretical framework ... 9
2.1 Overview of the major constructs ... 9
2.1.1. Customer experience ... 9
2.1.2. Brand-owned touchpoints as an element of the customer experience ... 9
2.1.3. Emotions ...10
2.1.4. Loyalty intentions...10
2.2 Conceptual Framework and Hypotheses ...11
2.2.1. The effect of a customer brand experience on loyalty intentions ...11
2.2.2. Emotions as a mediator ...12
2.2.3. Control variables ...14
3. Research Design ...16
3.1 Data ...16
3.2 Operationalization of Variables ...17
3.3 Cleaning and preparing structured data...17
3.4 Descriptive Statistics ...18 3.5 Combining constructs ...18 3.5.1. Factor analysis ...18 3.5.2. Reliability analysis ...21 3.6 Research method ...21 3.6.1. Mediation ...21 3.6.2. PROCESS ...22 4. Results ...23
4.1 Model without control variables (model 1) ...23
4.2 Model with control variables (model 2) ...24
5. Discussion ...26
6. Conclusion and Implications ...29
7. Implications for practitioners ...30
8. Limitations and Further Research ...32
REFERENCES ...34
1. Introduction
Digital innovations and advances in analytics, automation and artificial intelligence (AI), create new opportunities for businesses and the economy in the online environment (McKinsey, 2017). Among of these, AI is disrupting and reshaping digital marketing, which has resulted in a large number of companies (88%) that have adopted AI solutions (e.g. one-to-one interaction with customers via chatbots, translating speech to text). In order not to fall behind on these digital innovations and advances, companies are forced to explore the possibilities of interaction with individual customers with the aim of differentiating themselves in the future (Ramani & Kumar, 2008). In the Netherlands, online retailers are striving more than ever to differentiate themselves from competitors through enhancing customer experience, order and delivery options and accessibility (ING: Online aandeel in detailhandel stijgt, 2018). According to the economic bureau of ING, online retail is expected to increase with 25% in the upcoming six years. Currently, the media and entertainment sector has the largest online share of its industry (75%), followed by gaming (65%) and electronics (50%) (Online shoppen nu goed voor €25 miljard, 2018).
Alongside these developments, customers feel comfortable interacting with brands seamlessly, switching channels effortlessly and allowing their data to be used with the aim of receiving more personalized experiences within their customer journey (Atkins et al. 2016). As customers have become more comfortable, they have also become smarter as they have become more aware of the availability of other options as well as they have learned to set a good experience apart from a bad experience. These customer developments have resulted in greater expectations in terms of how customers want to receive marketing efforts of the firm and a less tolerant attitude towards brands that are not able to deliver a positive customer experience that meet customer expectations (Hyken, 2017; Atkins et al. 2016). If a company is not able to deliver on the entire customer experience across multiple interactions, this will result in a decrease of customer loyalty towards the firm (Maynes & Stone, 2014).
experience that is personalized and different from other offers. Furthermore, effective customer journeys should contain automation, proactive personalization, contextual interaction and journey innovation in order to create customer experiences that are more likely to stick, draw and capture customers for a longer period (Edelman & Singer, 2015).
Today, there has not been a single agreement on what is the best measure on a good customer experience. Yet, attempts have been made to measure the customer experience on loyalty, word-of-mouth and customer satisfaction (Maklan & Klaus, 2011; Brakus et al. 2009). Furthermore, while previous research has already shown the positive impact of customer equity drivers, such as value equity (preference for price, quality and convenience of the product or service), brand equity (strength and innovativeness of the brand) and relationship equity (perceived quality of the relationship between the firm and the customer) on loyalty intentions (Ou et al., 2014; Ou & Verhoef, 2017), these studies focused on the effect of both the objective and subjective assessment on loyalty intentions. Present research focuses solely on the subjective assessment of the customer experience, because it is argued that subjective assessments drive consumer decision making in low involvement decision processes. This means that in a purchase journey in which decisions are routinized, require little attention and little involvement of the customer (e.g. frequently purchased consumer goods in an online environment), the effect of subjective assessments of the customer experience on loyalty intentions matters the most (Lemon et al. 2014, p.3).
Moreover, Ou and Verhoef (2017) stress the importance of examining whether idiosyncratic or distinctive events that generate positive and negative emotions, lead to independence of positive and negative emotions. This is the reason why in present study, positive and negative emotions are examined independently.
Hence, present research explores the opportunity of creating another customer experience scale and tests whether the dimensions also positively impact loyalty intentions. Furthermore, while the positive link between customer equity drivers and loyalty intentions is empirically well supported (Ou et al. 2014; Ou & Verhoef, 2017), what remains uncertain is whether this effect also exists when focusing solely on the subjective assessment of the customer experience in an online purchase environment. Furthermore, the mediating role of emotions that occur independently as a result of an idiosyncratic or distinctive event is examined. This has not been done before. In addition, current research makes a distinction by asking specifically for brand-owned touchpoints as part of a so-called ‘customer brand experience’, which is a construct created based on the literature of Lemon and Verhoef (2016) and Brakus et al. (2009). This means that, present study does research upon the interaction of customers with touchpoints created by the brand or firm.
In order to measure the aforementioned constructs, research is conducted through an online survey among 249 respondents who made an online purchase in the past half year. Participants are asked to indicate to what extent the customer brand experience evoked their senses, sentiment, thoughts and actions after which they are asked to indicate their loyalty intentions to the firm. Furthermore, participants are asked to indicate to what extent they feel joy, happiness, enthusiasm, anger, regret or distrust towards the experience with the brand. All in all, this study has four main contributions. Firstly, the construct of the customer experience and more specifically digital brand-owned touchpoints as part of this experience and its effect on loyalty intentions are explored for which it combines literature found on customer experience and brand management. Next to this this, a new scale measuring a customer brand experience that evokes senses, sentiment, thoughts and actions is developed. Furthermore, present study focuses specifically on experiences purchases made online. Finally, the study provides empirical evidence of the potential mediating role of emotions and therefore it explores a rather subjective assessment of a good customer experience and its effect on loyalty intentions of which is argued that it provides a good foundation to (future) decision making and is a driver of specific behavioral outcomes.
2. Theoretical framework
2.1 Overview of the major constructs
2.1.1. Customer experience
In a study of Verhoef et al. (2009) with the aim to enhance our understanding of the concept customer experience, the authors propose a conceptual model on the determinants of customer experience. The authors argue that customer experience is a customer’s cognitive, affective, emotional and social response to elements of the experience created by retailers as well as by non-retailers. It is a holistic process that includes the search for a product or service to the time after the sale has been made.
A broader perspective on customer experience is taken by Verhoef and Lemon (2016) a couple of years later, who propose that customer experience is “a multidimensional construct focusing on a customer’s cognitive, emotional, behavioral, sensorial, and social responses to a firm’s offerings during the customer’s entire purchase journey” (p.71). By drawing upon literature found on customer experience in other studies, the authors extend our knowledge on customer experience by proposing a model that consists of various stages, containing different levels, which they call “touchpoints” (p.76). An example of a touchpoint could the webshop of a brand that is used by a customer. Verhoef and Lemon (2016) explain that “a customer might interact with each of these touch point categories in each stage of the experience” (p.76).
2.1.2. Brand-owned touchpoints as an element of the customer experience
Touchpoints may be brand-owned (e.g. brand-owned media and brand-controlled elements such as packaging or service), partner-owned (interactions initiated by marketing agencies), customer-owned (e.g. instructional videos made by customers on Youtube) or social/externally-owned (e.g. reviews). In addition, Baxendale et al. (2015) state that touchpoints are “an episode of direct or indirect contact with the brand” (p. 236). As a customer may also indirectly interact with the brand, one may infer that touchpoints are not solely created by the firm or the brand, but could also be created by other stakeholders that a firm does not control.
as the firm’s marketing efforts. They argue that a customer might encounter these touchpoints throughout its purchase journey (respectively previous experience, prepurchase stage, purchase stage, postpurchase stage and future experience), as well as they argue that the intensity of strength and importance of these touchpoints might differ across the purchase journey.
These touchpoints relate to the concept of brand-stimuli found in the literature on brand experience, which is defined as “subjective, internal consumer responses and behavioral responses evoked by brand-related stimuli that are part of a brand’s design and identity, packaging, communications, and environments” (Brakus et al. 2009, p. 53). The authors argue that brand-stimuli evoke sensorial, affective, behavioral and intellectual responses. Similar to Verhoef and Lemon (2016), Brakus et al. (2009) also argue that these responses vary in strength (strong versus weak brand experience) and intensity (e.g. positive versus negative brand experience).
Hence, in present article brand-owned touchpoints are considered to be part of the customer experience, which may elicit sensorial, affective, behavioral and intellectual responses from the customer during interaction due to brand-stimuli, that may vary in strength and intensity. In present article, the brand-owned touchpoints that customers interact with as a part of the customer experience are called a customer brand experience.
2.1.3. Emotions
Morrison and Crane (2007) define emotions as “a state of physical and mental readiness that involves valence (directional force), evaluative appraisal, a target (or object or stimulus) and behavioral tendencies” (p.412). DeWitt et al. (2008) argue that the intensity of emotion that are felt is dependent on the “nature and the meaning for the person having it” (p.271). This condition is also discussed by Zeelenberg et al. (2008), who state that “emotions typically arise when one evaluates an event or outcome as relevant for one’s concerns or preferences” as well as they argue that emotions are “acute” and “momentary experiences” (p.20).
In present paper, emotions are defined as an acute, momentary internal feeling state that becomes apparent through mental, bodily or behavioral tendencies which are directed at something (e.g. feeling happy about something), involves evaluative appraisal and that arise when the event or outcome is evaluated as important.
2.1.4. Loyalty intentions
commitment is the “desire to maintain a relationship,” as well as they suggest that, “various sources could create a stickiness that keeps customers loyal to the brand or company even when satisfaction might be low” (p.211).
As current study focuses specifically on the loyalty intentions towards a brand after interaction with brand-owned touchpoints, the definition of Oliver (1999) on brand loyalty is chosen which is defined as “a deeply held commitment to rebuy or patronize a preferred product/service consistently in the future [..] despite situational influences and marketing efforts having the potential to cause switching behavior” (p.3).
2.2 Conceptual Framework and Hypotheses
The conceptual framework for the variables affecting loyalty intentions is presented in Figure 1. It is argued that interaction with brand-owned touchpoints affect loyalty intentions under the influence of emotions. Additionally, age, gender and income are controlled for (Figure 1).
2.2.1. The effect of a customer brand experience on loyalty intentions
As mentioned earlier, present article proposes that brand-related stimuli are considered to be brand-owned touchpoints that elicit sensorial, affective, behavioral and intellectual responses as part of the customer brand experience. As one dimension is not isolated from another dimension, a consumer’s interaction with a touchpoint or brand-related stimuli can be experienced on multiple dimensions (Brakus et al. 2009).
the importance of an unique experience, that gives novel stimuli each time and that the value of a good customer experience is that it cannot learn from previously given responses.
Hence, according to this line of thought, customers desire unique and novel experiences in which knowledge and sensation acquisition are central.
Furthermore, the empirical support for behavioral outcomes as a result of a customer brand experience are to be found in Brakus et al. (2009), who argues that brand-related stimuli result in a positive direct impact on brand loyalty. Baxendale et. al (2015), elaborate on these findings by arguing that affective response has been shown to impact on spending and repeated purchase intentions.
Affective response finds its origins in the affect regulation theory, which explains how people process information based on their affective state as well as it explains how people regulate current and ideal affect and create behavioral strategies accordingly. The reason why people regulate their affective states stems from the hedonic principle that “people are motivated to approach pleasure and avoid pain” (Arnold & Reynolds, 2009, p.309).
Following this line of reasoning, it explains the relationship why customers are motivated to commit to rebuy or reuse a product or service as a result of a positive customer brand experience that elicited sensorial, affective, behavioral and intellectual responses. This means that a positive customer brand experience, such as customer interaction with the webshop of the focal brand, leads to an increase of loyalty intentions towards the focal brand. Therefore hypothesis 1 is stated as follows:
Hypothesis 1: A positive customer brand experience has a positive impact on loyalty intentions
2.2.2. Emotions as a mediator
From a customer experience perspective, various emotions may be at play during the stages of a customer experience as well as brand-owned touchpoints may elicit different emotions such as happiness, joy, enthusiasm, anger, regret and distrust (Ou & Verhoef, 2017). While we can’t control having emotions, people unconsciously regulate emotions on which we base most of our decision making (Zeelenberg et al. 2008). From the appraisal theory one has learned that emotions arise as a result of a person’s subjective interpretation and cognitive assessment of a particular event or situation being either positive and negative as well as these emotions form the foundation of future assessments (Scherer et al. 2001). The authors argue that certain types of appraisal or assessment patterns lead to certain emotions.
Furthermore, the aim of people during interaction is becoming closer to one another and maximize gains while minimizing losses. These gains and losses may be emotional, resulting in either positive or negative feelings. This is explained by the interdependence theory which focuses on the interaction between the partners as the essence of a close relationship (Thibaut & Kelley, 1959).
From these findings one may conclude that 1) people interact with the aim of becoming closer to one another in order to increase one’s own gains and minimize losses and 2) during these interactions people subjectively interpret and cognitively assess an event as being either positive or negative from which respectively either positive or negative emotions arise. This means that a positive customer brand experience (e.g. positive webshop experience), leads to positive emotions. As opposed to a negative customer brand experience, which leads to negative emotions.
Based on this line of argumentation, the following hypotheses are derived:
Hypothesis 2: A positive customer brand experience positively impacts positive emotions Hypothesis 3: A negative customer brand experience positively impacts negative emotions
Emotions are a driver of action (Pansari & Kumar, 2017). Moreover, emotions are better in predicting behavior than cognitive evaluations (Allen et al. 1992). This might be explained due to the theory of reasoned action, which argues that emotions affect behavioral intention, which in turn affects behavioral action (Engel et al. 1995). Another explanation might be found in the “hierarchy of effects” model of consumer behavior which explains that a consumer’s feelings, interest or desire (affect) result in behavior (action) (Lavidge & Steiner 1961).
in decision making. Whereas positive emotions result in approach tendencies, negative emotions result in avoidance tendencies (Zeelenberg and Pieters, 2004). This means that on the one hand, the perception of acquiring a goal, leads to a positive assessment and therefore consumers choose to continue their interaction, which results in higher loyalty intentions. On the other hand, perception of failure of acquiring a goal, leads to a negative assessment, which result in the decision to discontinue interaction and leads to lower loyalty intentions.
Based on these findings, the following hypotheses are derived:
Hypothesis 4: Positive emotions have a positive impact on loyalty intentions Hypothesis 5: Negative emotions have a negative impact on loyalty intentions
2.2.3. Control variables
3. Research Design
The following chapter discusses the research design. First, the target population is defined. Second, a sampling technique and sample size are established. Third, the execution of the sampling process is explained. This is followed by a specification of the measurement and scaling procedures. Then the operationalization of the variables (table 1, Appendix) is discussed. Subsequently, some descriptive statistics are provided. And finally, in order to test the hypotheses, the research method is presented.
3.1 Data
The target population for present study are males or females who have used a website or mobile application in order to buy clothing from a particular brand. The sampling frame are households in which potential respondents are over the age of 18. The sampling technique is a simple random sampling and the sample size is a minimum of 200 respondents. Empirical data is collected through the survey method. By means of this method, respondents are asked a variety of questions on a non-comparative itemized rating scale, such as the Likert scale regarding their customer brand experience, emotions and loyalty intentions as well as some questions are asked regarding their age, gender, income, involvement and technological readiness. Furthermore, the survey makes use of a direct approach, which means that the purpose of the research is disclosed to the respondents from the questions that are asked. The sampling technique used for present study is a nonpanel recruited sampling method, which means that respondents are recruited via Social Media platforms (Facebook, LinkedIn and Whatsapp). These potential respondents are provided a link to the survey after which they are directed to the Qualtrics survey platform, an online survey software tool (Malhotra, 2010).
the survey. An example of the form design and the scenario is to be found in table 2 of the Appendix.
3.2 Operationalization of Variables
In order to measure loyalty intentions, present study has slightly adopted the idea of self-reported probabilities of engaging in repurchase behavior in the future as created by Rust et al. 2004. However, as respondents are asked to imagine a specific customer brand experience during on online customer journey of a particular brand, they are asked the likelihood of engaging in repurchase behavior at the same brand on a 7-point Likert scale (1 = “not very likely”, 7 = “very likely”).
Furthermore, in order to measure the customer brand experience present study has adopted four items that each measure one of the brand experience dimensions (sensorial, affective, behavioral and intellectual) as created by Brakus et al. (2009). These items are selected due to their loadings on a specific dimension. An overview of these items may be found in table 1 (Appendix). The customer brand experience is measured according to a 7-point Likert scale in which respondents are asked to rate the extent to which they disagree or agree with these statements (1) “very much disagree” or (7) “very much agree”.
Subsequently, in order to measure positive or negative emotions respondents are asked to what extent they feel six specific emotions as a customer of brand at which they have previously purchased a product online via the website or mobile application. Regarding positive emotions, happiness, joy and enthusiasm are selected which are experienced during all phases of the customer journey. Negative emotions include anger, regret and distrust, which are likely to occur in a service context (Ou & Verhoef, 2017). These six emotions are measured according to a 7-point Likert scale (1 = “not at all”, 7 = “strongly”).
Furthermore, one pretest is conducted. This pretest investigates whether the scenario and questions are understandable. In addition, the reliability and face validity of completely filled in questions is examined. With the help of fellow students and the first supervisor, the questions are assessed on (1) clarity and wording and (2) style and structure.
3.3 Cleaning and preparing structured data
First step is to remove respondents that did not make any online purchases online via a website or mobile application in the past half year. The following step is to look for impossible values, out of range values or missing data that affect the quality of the data. As some of the respondents did not complete the survey (completion up to 66,3%), these participants are deleted. Furthermore, age as a variable is recoded to age groups for analytical purposes.
3.4 Descriptive Statistics
A total of 278 respondents participated in the study, of which there are 249 eligible responses. Of these respondents 39,4% is male and 60,6% is female and the average age is 21 years old. The majority (50,2%) earns less than €10.000 per year.
3.5 Combining constructs
3.5.1. Factor analysis
In pursuance of conducting a multiple mediation analysis, data is reduced by means of a factor analysis using the principal component method with the aim to enhance interpretability of the data. By conducting a factor analysis, one seeks to verify underlying dimensions due to common variance in the variables. Since these variables are related, they may be taken together. As a first step, the factorability of 15 items must be examined. Once it is established that a factor analysis may be conducted, one may continue the analysis and establish the amount of factors after which items are taken together and relabeled.
Before conducting the principal component method, several criteria must be met in order to be certain of the factorability of the data. First for each item is checked whether it correlates with at least 0.3 on other item in the correlation matrix. Second, the Kaiser-Meyer-Olkin measure must be at least be above 0.5 in order to continue as well as the Bartlett’s test of sphericity must be significant, as we may then assume that variables are correlated. Current analysis shows a Kaiser-Meyer-Olkin measure of sampling adequacy of .774 and the Bartett’s test is significant (χ2 (105) = 1375.005, p < .05). Finally, the communalities of all items are at least above 0.3, indicating that they share some common variance.
Given these results, it is decided to continue with the analysis of 15 items.
components indicate a value higher than one, while the fifth component being .974. Second, the total explained variance for four components is 62.8%, whereas the total explained variance of five components is 69.3%. Third, the components each respectively explain 26.6%, 16.5%, 11.7% and 8.0%. As the fifth component shows a value above the threshold of 5% (6.5%), this is also taken into account. Finally, inspection of the scree plot indicates that either four or five components may be extracted as the graph become flatter after the fifth dimension, meaning that successive components account for less variance. Inspection of the rotated component matrix shows that both the items measuring the customer brand experience and the items measuring positive emotions load on one factor, which indicates that a number of four components should be extracted. However, as the total variance is best explained extracting five components, each of the components explain at least 5% and it is theoretically founded to consider customer brand experience and positive emotions as different constructs as the customer brand experience measures senses, sentiment, action and thoughts about a brand or firm and positive emotions measure the extent to which someone feels happiness, joy and enthusiasm. Therefore, the final decision is to extract five components.
Table 3: Total variance explained
Comp. Initial Eigenvalues
Ex. on Sums of Squared Loadings Rot. on Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance 1 3,971 26,471 26,471 3,971 26,471 26,471 3,468 23,119 2 2,479 16,53 43,001 2,479 16,53 43,001 2,454 16,361 3 1,762 11,744 54,745 1,762 11,744 54,745 1,996 13,307 4 1,203 8,021 62,766 1,203 8,021 62,766 1,497 9,979 5 0,974 6,495 69,26 6 0,79 5,264 74,525 7 0,703 4,688 79,212 8 0,579 3,857 83,069 9 0,542 3,614 86,683 10 0,516 3,439 90,123 11 0,419 2,79 92,913 12 0,359 2,393 95,306 13 0,264 1,762 97,068 14 0,254 1,696 98,764 15 0,185 1,236 100
Table 4: Rotated component matrix
Rotated Component Matrix
Component 1 2 3 4
Buying from this brand/firm makes a strong impression
on my visual sense or other senses 0,556
Buying from this brand/firm activates feelings and
sentiment 0,641
I engage in physical actions and behaviors when I buy from this brand/firm
0,528
I engage in a lot of thinking when I encounter this
brand/firm 0,386 Happiness 0,849 Joy 0,877 Enthusiasm 0,850 Anger 0,864 Regret 0,877 Distrust 0,878
How important are the products or services in this
industry to you? 0,852
How interested are you in the products or services in
this industry? 0,720
Technology gives people more control over their daily
lives 0,743
Technology gives you more freedom of mobility 0,755
Products and services that use the latest technologies are
much more convenient to use 0,753
3.5.2. Reliability analysis
A reliability analysis of all items of every construct is conducted to measure the internal consistency of the construct. None of the cases required to remove items for a higher Cronbach’s alpha and therefore it is decided to maintain the underlying items. The following step involved mean centering the items into five constructs after which they are labeled (table 5).
Construct Cronbach’s alpha Label
CBX .698 Eliciting sensations and sentiment and engaging in
thinking and behavioral action
POS_EMO .891 Feeling joy, happiness and enthusiasm
NEG_EMO .861 Feeling anger, regret and distrust
INV .551 Products and services in this industry are of interest
and are important to me
TECH .676 Believe that technology are gives people more
control of their lives, freedom of mobility and are much convenient to use
Table 5: Reliability scores and labels 3.6 Research method
3.6.1. Mediation
In a mediation model it is
hypothesized that the
independent variable changes the mediator, which in turn changes the dependent variable, the outcome. The mediated effect is the effect of the mediator serving as an intervention on the outcome (Baron and Kenny, 1986).
In figure 2a, the total effect of X on Y is described (path c). This is followed by figure 2b which represents the direct effect of X on Y (path c’) as well as it represents the indirect
effect of X on Y through j mediators, which described as the specific indirect effect of X on Y
via mediator j and is built up through the product of two unstandardized paths. Therefore, the specific indirect effect of X on Y through M1, is equal to the product of a1b1 (1). In order to
calculate the total indirect effect of j mediators, one must take the sum of the specific indirect effects (2). Naturally, the total effect is the sum of the direct effect and the specific indirect effects Preacher and Hayes (2008). An example of these equations can be found below:
Σ(𝑎𝑖𝑏𝑖) (1)
𝑐 = 𝑐)+ Σ𝑖(𝑎𝑖𝑏𝑖) (2)
3.6.2. PROCESS
The PROCESS 3.0 macro as written by Hayes as a path analysis modeling tool for SPSS is utilized in order to continue with the mediation analysis. However, there are four assumptions that have to be met in order to continue with the analysis. Firstly, all variables must be measured on a continuous scale. Secondly, all variables must follow a normal distribution. Thirdly, variables must be independent from each other. Finally, all variables are assumed to be linear. In order to test the underlying effects of the multiple mediation model, the first step is to confirm the significance of the effect of the mediators on the independent variable (a path). The following step is to confirm the significance of the effect of the mediator on the dependent variable, while controlling the independent variable (b path). Thirdly, the significance of the effect of the independent variable on the dependent variable must be confirmed, the total effect (c path). The final step is to confirm the insignificance or the notable reduction in effect of the independent variable on the dependent variable, while accounting for the effect of the mediators in the model, the direct effect (c’ path).
Furthermore, as the normality assumption in small sample sized is questionable, the bootstrap method as discussed by Preacher and Hayes (2008) is also conducted. In this method, samples of the same size as the original with replacement are drawn randomly. By doing so, more data on the variability of the parameter estimates becomes available.
4. Results
Current chapter examines the results of the indirect effect of two mediators, total effect and direct effect in terms of significance and size of the effect, followed by an examination of the bootstrapping results of the indirect effect of both positive emotions and negative emotions. By doing so, the model is run twice, first without the control variables age, gender, income, involvement and technological readiness, which is followed by a model that does contain these variables.
4.1 Model without control variables (model 1)
First, the a path which illustrates the indirect effect of customer brand experience (CBX) to positive emotions appears to be relatively large and very significant. However, the b path which illustrates the indirect effect of positive emotions on loyalty intentions is considered not to be significant (p > .05). Furthermore, the indirect effect of CBX on negative emotions is not significant (p > .05), but the indirect effect of negative emotions on loyalty intentions is relatively large and
significant. Thirdly, the total effect (c path) portrays the effect of customer brand experience on loyalty intentions without accounting for the effect of positive and negative emotions as mediators. In the model without control variables, this effect seems to be significant (p < .05). Furthermore, the c’ path which shows the direct effect and controls for positive and negative emotions as mediators is smaller and significant (p < .05) and has an effect of .2268. Finally, a considerately low amount of the variance is explained by the dependent variable (R2 = 0.0455). In addition, the results of the bootstrapping method are examined as these provide a set of confidence intervals in which true indirect effects that are very likely to occur. The true total
indirect effect of both the positive and negative emotions (.0152) is 95% likely to range from -.0840 (LLCI) and .1128 (UPCI) and flows from the independent variable through the mediator to the dependent variable, however this effect is not significant as zero lies between the two confidence intervals.
4.2 Model with control variables (model 2)
Similar to the previous model, the indirect effect (a path) of customer brand experience on positive emotions is relatively large (.4835) and significant (p < .05). Again, the indirect effect of positive emotions on loyalty intentions is not significant (p > .05). Furthermore, the indirect effect of CBX on negative emotions is in this model significant (p < .05) as well as the indirect of negative emotions on loyalty intentions is relatively large and significant. The c path, which
illustrates the total effect of customer brand experience on loyalty intentions is .1639 (p < .05), which is a smaller effect than was found in the previous model and less significant. Furthermore, the direct effect which is the effect of customer brand experience on loyalty intentions while accounting for the effect of the mediators is larger than the total effect (.1903, p < .05). Finally, the variance which is explained by the dependent variable is higher (R2 = 0.1230).
In addition to these results, the bootstrap results are also examined. The true total indirect effect (-.0264) is 95% likely to range from -.1302 (LLCI) and .0742 (UPCI) and flows from the independent variable through the mediator to the dependent variable, however this effect is not significant as zero lies between the two confidence intervals.
When comparing both models, a slight higher amount of significant effects are found when control variables such as age, gender, income, involvement and technological readiness are included. Furthermore, the dependent explains more of the variance in the second model with control variables. Of these control variables, only involvement appears to have a relative large (.2524) and significant effect on loyalty intentions (p < .05). On top of this, the bootstrapping method indicates that the indirect effect of positive and negative emotions in both models is small and not significant.
5. Discussion
Over the years, researchers have been developing scales to measure constructs such as a customer experience (Maklan & Klaus, 2011) or brand experience (Brakus et al. 2009) on loyalty, word-of-mouth and customer satisfaction. While the link between these constructs is well established, there is still not a single answer to what could be drivers of a customer experience that positively impact loyalty intentions.
Furthermore, elaborate research has been done into the effect of customer equity drivers such as value equity (preference for price, quality and convenience of the product or service), brand equity (strength and innovativeness of the brand) and relationship equity (perceived quality of the relationship between the firm and the customer) equity on loyalty intentions, however less is known about the sole subjective assessment of customers on touchpoint interaction within a customer brand experience and its effect on loyalty intentions towards the brand or firm (Ou et al. 2014; Ou & Verhoef, 2017).
On top of this, even though the effect of positive and negative emotions on loyalty intentions has been researched (Ou & Verhoef, 2017), it has not been researched yet how emotions might mediate the relationship between a customer brand experience and loyalty intentions, even though it has been argued that emotions form a strong foundation for decision making and important driver for certain behavioral outcomes in the present and in the future.
Therefore, present study makes use of emotions such as joy, happiness, enthusiasm, anger, regret and distrust as these are considered to be suited measures in consumption-related situations (McColl & Smith, 2006) or are proven to be felt in a service related context (Ou & Verhoef, 2017).
Yet, after running the mediation analysis as proposed by Baron and Kenny (1986) and conducting the bootstrap method according to Preacher and Hayes (2008), the findings indicate that emotions do not mediate the relationship between the customer brand experience and loyalty intentions in an online purchase environment.
their own affective regulation (Arnold & Reynolds, 2009). By repeatedly exposing themselves to experiences that are pleasurable, customers feel motivated to commit, to rebuy or reuse products or services.
Furthermore, a significant indirect effect is found in the relationship between a customer brand experience and positive emotions, which means that hypothesis 2 is supported. This result justifies the idea that people may feel positive or negative emotions as a result of own subjective interpretation and assessment of the customer brand experience after which customers decide whether the experience is truly maximizing a gain and minimizing a loss (Scherer et al. 2001; Thibaut & Kelley, 1959). In addition, a significant positive impact of customer brand experience on negative emotions is found, which means that hypothesis 3 is also supported.
However, an insignificant and relatively small effect is found in the relationship between positive emotions and loyalty intentions. This result conflicts the ideas proposed by Pansari and Kumar (2017) that emotions are a driver of actions and are a better predictor than cognitive evaluations (Allen et al. 1992). The results suggest that positive emotions are a relative weak predictor of higher loyalty intentions and therefore hypothesis 4 is not supported. On the contrary, feeling negative emotions such as anger, regret and distrust towards the brand or firm does negatively impact loyalty intentions, which means that hypothesis 5 is
supported. On may assume that feeling negative emotions results in avoidance tendencies due
to the perception of failure of acquiring a goal.
Additionally, there is also a need to discuss some of the unanticipated findings:
This means that positive emotions such as happiness, joy and enthusiasm are emotions that might not even be associated with loyalty intentions. Another cause that might be the reason behind this finding is that it was not possible to indicate mixed emotions, which Zeelenberg et al. (2008) call “emotional ambivalence” (p.21). According to the authors, emotions cannot be split into groups of good and bad or positive and negative, rather the question of what emotions influence certain behavior should be researched.
Secondly, as the results of the bootstrap method according to Preacher and Hayes (2008) are considered to be leading, it is concluded that both positive and negative emotions do not mediate the relationship between customer brand experience and loyalty intentions. This finding does not directly contrast ideas or concepts proposed by other authors, though it does confirm that emotions are not the best mediator in present model. Constructs such as brand personality, brand community, brand trust, brand attachment and brand love might be constructs that are of better use in this context as they have proven to be antecedents of brand loyalty (Brakus et al. 2009).
6. Conclusion and Implications
Digital advances are reshaping the market and together with higher expectations among customers with regard to their customer journey and customer experience as a result of becoming more advanced in their online purchase behavior, it becomes harder to have loyal customers to your brand, especially if a large part of the customer journey and therefore a big part of the customer brand experience occurs online.
Next to these developments, services online in The Netherlands are growing and emotions are seen as a way to connect with customers in an online environment and attach them to your services
Previous research has already established a link between emotions and loyalty behavior, but there is little known on whether emotions could play a mediating role. In other words, whether brands can truly connect via emotions or that there are other constructs and potential relations at play?
Current study shows that emotions do not mediate the relationship between a good customer brand experience and loyalty intentions
The reasons behind these findings are: 1) Each emotion might have a different behavioral outcome as explained by the feeling-is-doing approach which implies that loyalty intentions are not necessarily an outcome that consumers experience after having felt one or a couple of emotions. Furthermore, 2) consumers might feel both positive and negative emotions, which indicates that they feel emotions that are ambivalent. Current research does not test for ambivalent emotions and categorized these as either positive or negative. And 3) other constructs such as brand personality, brand community, brand trust, brand attachment and brand love could have been included as they have proven to be antecedents of brand loyalty. This means that there is still a lot of model improvement concerning the role of emotions. For example, while it might not be a good mediator, it might be a good predictor.
7. Implications for practitioners
It is clear that one of the most important implications for practitioners are that a well performing customer brand experience which evokes customer’s senses, sentiment, thoughts and action should be considered as one of the performance measures of brand-owned touchpoints. This means that practitioners who are responsible for this, should include the performance measure in their marketing strategy in order to gain more insights into the loyalty intentions among customers. Based on these insights, practitioners could segment loyalty intentions into customer groups (e.g. low, medium to high) and have more arguments to target a specific customer group.
In order to create interactions that evoke senses, sentiment, thoughts and action, practitioners could make attempts to create interactions that are personalized and very distinct from other interactions. By doing so, customer expectations might be exceeded, which may result in an increase of loyalty intentions. A better representation of how such a scenario may be performed is explained by De Bruijn (2018), who argues that a personalized customer experience consists of four steps. First, practitioners have to learn what drives customers. Or in other words, learn what makes them happy. Second, practitioners should explore what information customers are willing to share about themselves in return for a better customer experience. Therefore, it is suggested that practitioners should test what customers want to share or rather keep to themselves through A/B testing or through customer research. By doing so, practitioners must continue to explain why customer data is needed. Furthermore, data must be more centralized in order to create a better image of the customer. Also, by applying profiling techniques, practitioners can create an image of a particular customer group. Through AI and machine learning it is possible to extract patters from customer behavior of which practitioners can predict preferences or buying behavior in the future. Finally, De Bruijn suggests that surprising customers with a gift or an unexpected turn in the experience is a strong move to evoke senses, sentiment, thoughts and action.
In practice, these implications and suggestions could be incorporated in the firm’s current marketing efforts, for example by benchmarking each brand-owned touchpoint with the customer brand experience scale in order to make sure the brand-owned touchpoint is indeed evoking senses, sentiment, thoughts and action.
8. Limitations and Further Research
The following chapter discusses some of points of limitations and weaknesses that might have affected the validity of the results. In addition, it presents an outlook to future research directions or ideas.
First of all, negatively and positively worded customer brand experience scale questions (e.g. I engage in a lot of negative/positive thinking) could have maybe prevented positive emotions to load on the same factor as the customer brand experience, which means that the results could have looked different.
Furthermore, if a customer is highly involved with the product or service he or she bought online, this person may also process other touchpoints (e.g. partner-owned touchpoints) differently. As current research took a rather micro approach, it did not take into account any interplay or spillover effects of experiences and expectations encountered through other touchpoints.
Thirdly, participants were asked to reflect on their latest online purchase. However, while some may have thought about their experience in the pre-purchase phase, others might have reflected on the post-purchase phase. Current study did not account for this, even though the influence on loyalty intentions might be significantly different. In other words, touchpoints may influence customers in various phases of the customer journey differently (e.g. chatbots on a website). Current study made an attempt to narrow the scope to brand-owned touchpoints in the entire customer journey, however other touchpoints may have had a bigger influence on pushing a particular customer down the funnel (e.g. social/external touchpoints).
In a similar line of thought, participants were not asked to solely think of the online touchpoints while making the purchase. Therefore, participants might have also reflected on touchpoints that occurred in an offline context (e.g. fast or slow delivery). Current research did not account for this.
Finally, as the context is an online environment, one may argue that it might be relatively difficult to create touchpoints that evoke emotions. Comparing these results with a study in an offline environment could provide more clarification.
customer delight or brand attachment. Finally, the measure of customer experience is very complex and much broader than the measure presented in this research. Therefore, future researchers may also want to explore and define a measure that is specific to their context (Maklan & Klaus, 2011).
REFERENCES
Atkins, F., Fenech, C., Perkins, B. and Thomson R. (2016). The Deloitte consumer review, CX marks the spot: Rethinking the customer experience to win. Retrieved from:
https://www2.deloitte.com/content/dam/Deloitte/uk/Documents/consumer business/deloitte-uk-consumer-review-customer-experience.pdf
Allen, C. T., Machleit, K. A., & Kleine, S. S. (1992). A comparison of attitudes and emotions as predictors of behavior at diverse levels of behavioral experience. Journal of
Consumer Research, 18(4), 493– 504.
Arnold, M. J., & Reynolds, K. E. (2009). Affect and retail shopping behavior: Understanding the role of mood regulation and regulatory focus. Journal of Retailing, 85(3), 308-320. Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
personality and social psychology, 51(6), 1173.
Baxendale, Shane, Emma K. Macdonald, and Hugh N. Wilson (2015), “The Impact of
Different Touchpoints on Brand Consideration,” Journal of Retailing, 91 (2), 235–53. Bloemer, J., & De Ruyter, K. (1999). Customer loyalty in high and low involvement service
settings: The moderating impact of positive emotions. Journal of Marketing Management, 15(4), 315–330.
Brakus, J. Josko, Bernd H. Schmitt, and Lia Zarantonello (2009), “Brand Experience: What Is It? How Is It Measured? Does It Affect Loyalty?” Journal of Marketing, 73 (May), 52–68
Chaudhuri, A., & Holbrook, M. B. (2001). The chain of effects from brand trust and brand affect to brand performance: the role of brand loyalty. Journal of marketing, 65(2), 81 93.
De Bruijn, M. (2018). Breng emotie terug in de customer experience. Retrieved from https://www.frankwatching.com/archive/2018/06/12/breng-emotie-terug-customer experience/
DeWitt, T., Nguyen, D. T., & Marshall, R. (2008). Exploring customer loyalty following service recovery: The mediating effects of trust and emotions. Journal of service research, 10(3), 269-281.
Edelman, D. C., & Singer, M. (2015). Competing on customer journeys. Harvard Business
Review, 93(11), 88-100.
Engel, J. F., Blackwell, R. D., & Miniard, P. W. (1995). Consumer Behavior, 8th. New York: Dryder.
Gustafsson, A., Johnson, M. D., & Roos, I. (2005). The effects of customer satisfaction, relationship commitment dimensions, and triggers on customer retention. Journal of marketing, 69(4), 210-218.
Hyken, S. (2017). Ten customer service and CX predictions for 2018. Retrieved from https://www.forbes.com/sites/shephyken/2017/12/31/ten-customer-service-and-cx predictions-for-2018/#2b299511352a
Iglesias, O., Singh, J. J., and Batista-Foguet, J. M. (2011). The role of brand experience and affective commitment in determining brand loyalty. Journal of Brand Management, 18(8), 570-582.
ING: Online aandeel in detailhandel stijgt door tot 25 procent in 2025’. (2018, April 23). Retrieved from https://twinklemagazine.nl/2018/04/ing-detailhandel/index.xml Johnson, D., & Grayson, K. (2005). Cognitive and affective trust in service relationships.
Journal of Business Research, 58(4), 500–507.
Lemon, K. N., Rust, R. T., & Zeithaml, V. A. (2001). What drives customer equity.
Marketing management, 10(1), 20.
Lemon, K. N., & Verhoef, P. C. (2016). Understanding customer experience
throughout the customer journey. Journal of Marketing. doi:10.1509/jm.15.0420. In Press
Lau, G. T., & Lee, S. H. (1999). Consumers' trust in a brand and the link to brand loyalty. Journal of Market-Focused Management, 4(4), 341-370.
Lavidge, R. J., & Steiner, G. A. (1961). A model for predictive measurements of advertising effectiveness. Journal of Marketing, 25(October), 59–62.
Liljander, V., Gillberg, F., Gummerus, J., & Van Riel, A. (2006). Technology readiness and the evaluation and adoption of self-service technologies. Journal of Retailing and
Consumer Services, 13(3), 177-191.
Machleit, K. A. & Wilson, R. D. (2013) Emotional Feelings and Attitude toward the
Advertisement: The Roles of Brand Familarity and Repetition, Journal of Advertising, 17:3, 27-35, DOI: 10.1080/00913367.1988.10673121
Malär, L., Krohmer, H., Hoyer, W. D., & Nyffenegger, B. (2011). Emotional brand
attachment and brand personality: The relative importance of the actual and the ideal self. Journal of Marketing, 75(4), 35-52.
Malhotra, N. K. (1999). An applied orientation. Marketing Research, 2.
Matzler, K., Bidmon, S., & Grabner-Kräuter, S. (2006). Individual determinants of brand affect: the role of the personality traits of extraversion and openness to experience. Journal of Product & Brand Management, 15(7), 427-434.
Mayes, J. and Stone, D. (2014). For customer loyalty, only the best will do. Retrieved from https://www.mckinsey.com/business-functions/marketing-and-sales/our-insights/for customer-loyalty-only-the-best-will-do
McColl-Kennedy, J. R. and A. K. Smith (2006). ‘Customer emotions in service failure and recovery encounters’ In W. J. Zerbe, et al. (eds), Research on Emotion in
Organizations. Vol. 2: Individual and Organizational Perspectives on Emotion Management and Display, pp. 237-268. Amsterdam: Elsevier.
McKinsey. (2017). What’s now and next in analytics, AI, and automation. Retrieved from https://www.mckinsey.com/global-themes/digital-disruption/whats-now-and-next-in analytics-ai-and-automation
Morris, J. D. & Boone, M. A. (1998) ,"The Effects of Music on Emotional Response, Brand Attitude, and Purchase Intent in an Emotional Advertising Condition", in NA
Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT: Association for Consumer Research, Pages: 518-526.
Morrison, S. and Crane, F.G. (2007) Building the service brand by creating and managing an emotional brand experience. Journal of Brand Management 14(5): 410–421.
Nelson-Field, K., Riebe, E., & Newstead, K. (2013). The emotions that drive viral video. Australasian Marketing Journal, 21(4), 205–211.
Oliver, R. L. (1999). Whence consumer loyalty?. the Journal of Marketing, 33-44.
Online shoppen nu goed voor €25 miljard, top 10 webwinkels van Nederland. (2018, April 9). Retrieved from https://www.consultancy.nl/nieuws/17175/online-shoppen-nu-goed- voor-25-miljard-top-10-webwinkels-van-nederland
Ou, Y. C., Verhoef, P. C., & Wiesel, T. (2017). The effects of customer equity drivers on loyalty across services industries and firms. Journal of the Academy of Marketing Science, 45(3), 336-356.
Palmer, A. (2010). Customer experience management: a critical review of an emerging idea.
Journal of Services marketing, 24(3), 196-208.
Pansari, A., & Kumar, V. (2017). Customer engagement: the construct, antecedents, and consequences. Journal of the Academy of Marketing Science, 45(3), 294-311. Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing
and comparing indirect effects in multiple mediator models. Behavior research
methods, 40(3), 879-891.
Ramani, G. & Kumar V. (2008), “Interaction Orientation and Firm Performance,” Journal of Marketing, 72 (January), 27–45
Richins, M. L. (1997). Measuring emotions in the consumption experience. Journal of consumer research, 24(2), 127-146.
Roe, D. (2017). The Role Of Emotion In Digital Customer Experiences. Retrieved from https://www.cmswire.com/digital-experience/the-role-of-emotion-in-digital-customer experiences/
Rust, R. T., Lemon, K. N., & Zeithaml, V. A. (2004). Return on marketing: Using customer equity to focus marketing strategy. Journal of Marketing, 68(1), 109–127.
Salesforce. (n.d.). How AI Software is Reshaping Digital Marketing Right Now. Retrieved from https://www.salesforce.com/products/marketing-cloud/best-practices/marketing ai/
Sahin, A., Zehir, C., & Kitapçı, H. (2011). The effects of brand experiences, trust and satisfaction on building brand loyalty; an empirical research on global brands. Procedia-Social and Behavioral Sciences, 24, 1288-1301.
Scherer, K. R., Schorr, A., & Johnstone, T. (2001, Eds.). Appraisal processes in emotion: Theory, methods, research. New York: Oxford University Press.
Sorrels, M. (2018, March 5). Voice tech in travel, part 1: Brand adoption. Retrieved from https://www.phocuswire.com/Voice-series-part-1-brand-adoption
Thibaut, J. W., & Kelley, H. H. (1959). The social psychology of groups.
Whan Park, C., MacInnis, D. J., Priester, J., Eisingerich, A. B., & Iacobucci, D. (2010). Brand attachment and brand attitude strength: Conceptual and empirical differentiation of two critical brand equity drivers. Journal of marketing, 74(6), 1-17.
Zeelenberg, M., Nelissen, R. M., Breugelmans, S. M., & Pieters, R. (2008). On emotion specificity in decision making: Why feeling is for doing. Judgment and Decision making, 3(1), 18.
APPENDICES
Measurement of variables
Measurement variable Latent variable
Dependent variable
LI Imagine you should buy this product or service again.
How likely are you to repurchase from the same brand? (Rust et al. 2004; Ou et al. 2014; Ou & Verhoef, 2017)
Loyalty intentions
Independent variables
SEN Buying this brand/firm makes a strong impression on
my visual sense or other senses
AFF Buying this brand/firm induces feelings and
sentiment
BEH I engage in physical actions and behaviors when buy
from this brand/firm
INT I engage in a lot of thinking when I encounter this
brand/frim
Customer brand experience
EMO Please indicate whether you feel the following
emotions as a customer of the brand/firm based on your experience with the brand
(1) happiness, (2) joy, (3) enthusiasm, (4) anger, (5) regret and (6) distrust
(Ou & Verhoef, 2017)
Emotions
Control variables
AGE What is your age? Age
GEN What is your gender? Gender
INC What is your approximate average annual income? Income
INV1 How important are the goods or services in this
industry to you?
INV2 How interested are you in the products or services in
this industry?
(Bloemer & De Ruyter, 1999)
TR1 Technology gives people more control over their
daily lives
TR2 Technology gives you more freedom of mobility TR3 Products and services that use the latest technologies
are much more convenient to use (Liljander et al. 2006)
Technological readiness
Table 1: Measurement and latent variables
Survey design Part A: Introduction
Hi!
Thank you for taking part in this survey in which I would like to gain insights surrounding your latest online purchase via a website or mobile application. The survey should only take 3-5 minutes to complete. Be assured that all answers you provide will be kept in the strictest confidentiality!
Your input is very much appreciated! Please click ‘Next’ to begin.
Part B: Selection
1. Have you bought a product or service online via a website or mobile application in the past half year?
Part C: Customer experience and loyalty intentions
Reflect back on the time when you bought a product or service online via a website or mobile application and try to answer the following questions as if you were to be in the same situation again.
2. Imagine you should buy this product or service again. How likely are you to repurchase from the same brand/firm?
3. Buying from this brand/firm makes a strong impression on my visual sense or other senses
4. Buying from this brand/firm activates feelings and sentiment
5. I engage in physical actions and behaviors when I buy from this brand/firm (e.g. write a review or talk about the brand/firm with others)
6. I engage in a lot of thinking when I encounter this brand/firm
Part D: Emotions
Please indicate whether you feel the following emotions as a customer of the brand/firm based on your experience with the brand/firm
7. Happiness 8. Joy 9. Enthusiasm 10. Anger 11. Regret 12. Distrust
Part E: Involvement and technological readiness
Please indicate to what extent you agree with the following statements 13. How important are the products or services in this industry to you? 14. How interested are you in the products or services in this industry? 15. Technology gives people more control over their daily lives 16. Technology gives you more freedom of mobility
17. Products and services that use the latest technologies are much more convenient to use
Part F: Demographic information
18. What brand/firm did you have in mind 19. What is your age?
20. What is your gender?
21. What is your approximate average annual income?
Dear participant,
Thank you for taking the time to complete this survey! Your input has been recorded. Best wishes,
Talitha la Macchia
