1. Introduction
For the firm to survive in the highly competitive industrial environment, customer satisfaction has been a very important topic both in academia and in business. In many empirical findings, customer satisfaction gives a positive impact on consumer behaviors. But there were also researches reporting that mere satisfaction is not enough and something more is needed than just mere satisfaction. For example, Rust and Oliver (2000) suggested that customers want to be delighted than to be merely satisfied for greater loyalty. Also, delighted customers reveal higher retention and loyalty to the firm (Oliver et al., 1997; Hicks et al., 2005). Therefore, not only in business but also in academia, focus on customer satisfaction began to expand to customer delight. Despite the rise of attention on the issue of customer delight, not much research has been made regarding customer delight.
Present research suggest that customer delight gives positive impact on consumer perception, intention, and behavior (Paterson, 1997; Finn, 2005; Chitturi et al., 2008). But no research raised question in what circumstances do consumers get more delighted and reveal strong behavioral intention. This research focuses on customer delight and adapted the influence of surprise and attribute importance. Among the research of customer delight, couple of issues were mentioned. The first issue is whether customer delight can be elicited without surprise. That is, people can be delighted at the first experience with the emotion of surprise, but also can still be delighted at the second experience even though they already know the full process of the experience. For example, in watching a famous opera, Kumar, Olshavsky, and King (2001) found that people were delighted when they watched the opera for the first time, but also they were delighted at the same opera when they watched it for the second time. In that case, people feel the emotion of delight with less surprise but still the emotion is delight with pleasure. Not much study about customer delight paid attention to the role of surprise. In this research, I focused on the role of surprise on the effect of delight on behavioral consequences such as repurchase intention and recommendation intention.
The second issue is about the role of attribute importance in customer delight. In the research that elicited the emotion of delight via an experiment, reported statistically insignificant effect of the emotion of delight and mentioned that the insignificant effect of delight might be due to the delight element that might have been trivial to the main service (Vanhamme, 2008). Specifically, respondents were provided with a comic voucher to elicit delight when they were experiencing a virtual museum. Certainly in every product or service consumers experience, they have different levels of importance or relevance. Obviously, some product/service is important and relevant to certain consumers and some product/service is less important or relevant to certain consumers. Therefore, this research empirically tests delight that was elicited in important attribute and how it gives impact on consumers' behavioral consequences such as repurchase intention and recommendation intention.
2. Literature Review and Hypothesis Development
2.1. Concept of Customer Delight
According to previous research, the concept of customer delight can be categorized into threefold. First, is ‘Delight-as-extreme satisfaction’. This perspective is viewing customer delight as a continuous spectrum of customer satisfaction. Here, delight lies on the right extreme of customer satisfaction. It means that delight comes when consumers are totally and completely satisfied (Matzler et al., 1996; Keiningham et al., 1999; Berman, 2005; Dixon, Freeman, & Toman, 2010; Vanhamme, 2008; Vanhamme & Snelders, 2003; Sivakumar et al., 2014). Second is ‘Delight-as surprise & joy’. This perspective views delight as a combination of the emotion of surprise and joy (Westbrook & Oliver, 1991; Keiningham & Vavra, 2001; Finn, 2005, 2012; Verma, 2003; Füller & Matzler, 2008; Chitturi et al., 2008). Third is the perspective viewing ‘Delight-as-joy’. Not much research raised the issue that delight can constitute of pleasant emotion such as joy. Only Kumar et al. (2001) suggested that the emotion of delight can be evoked with only joy.
2.2. Previous Studies of Customer Delight
Previous research have reported that customer delight positively affects firm’s performance. For example,
Reicheld (1994) noticed that merely satisfied customers may switch to other companies and only ‘completely satisfied’ customers become loyal customers. Also delighted customers show higher retention and keep loyal to the firm (Oliver et al., 1997). [Table 1] summarizes previous studies in customer delight.
[Table 1] Previous studies on customer delight
2.3. Contribution and Focus of Research
Among the research in customer delight, it was found that customer delight positively affects consumers’ loyalty toward the firm (Chitturi et al., 2008; Füller & Matzler, 2008; Hicks et al., 2005). Also, there were articles investigating the emotion of ‘surprise’ (Vanhamme, 2008; Vanhamme & Snelders, 2003; Alden et al., 2000; Valenzuela et al., 2010). To address the focus of this research and its contributions, [Table 2] shows that this research focuses on customer delight, surprise, and attribute importance.
[Table 2] Previous studies on delight, surprise and attribute importance
2.4. Surprise and Delight
Research in surprise was mainly performed in psychology and one of the issues was that whether surprise is a pure emotion, a cognitive process, or a combination of emotion and cognitive process. As an emotion, surprise is defined as a 'feeling' that is temporary and volatile (Izard, 1977). It is an emotion but the difference between the basic emotions such as joy or sadness is that surprise is a neutral emotion that does not reveal a positive valence nor a negative valence. But emotion such as joy or pleasure is completely a positive valence and hate or displeasure is definitely a negative valence. The perspective that defines surprise not as an emotion is that surprise is a mere automatic 'reactance' from a mismatch of personal schema.
Another characteristic of surprise besides surprise as a neutral emotion is that surprise plays an amplifying role in eliciting emotions. That is, if it was without surprise, the elicited positive emotion might have been joyful or pleasant, but with surprise the joyful emotion can be amplified into delight or into much greater joy. For negative emotion, if it was without surprise, the elicited negative emotion might have been displeasure or unpleasant, but with surprise the unpleasant emotion can be amplified into stronger negative feelings like anger or rage.
In marketing, research about surprise mainly dealt with the impact of surprise on dependent variables. Specifically, Valenzuela et al. (2010) found that when consumers were given an unexpected present, they were more pleasant than when they were expected to receive a present. Also, in the context of advertisement, Alden et al. (2000) found that a humorous advertisement that elicited the emotion surprise made the consumers think the advertisement as more humorous than when they were not surprised at the advertisement. As from previous research, surprise plays an amplifying role in eliciting certain emotions or outcome. Since stronger emotions lead to higher behavioral intentions, this research suggests that if the emotion of delight was elicited with surprise then it will lead to higher behavioral intention such as repurchase intention or recommendation intention than when the positive emotion was elicited without surprise. Thus, suggested hypothesis is:
[H1] Delight that is elicited with surprise will reveal higher (a) repurchase intention and (b) recommendation intention than when delight is elicited without surprise.
2.5. Attribute Importance and Delight
Generally, in making a decision to choose a product or a service or in a usage situation of a product or a service, people make evaluations whether the attribute of a product or a service is important or not important to themselves. These evaluations lead to attitudes and become an important standard in evaluating and making a decision to buy a product (Jaccard, Brinberg, & Ackerman, 1986). If these important attributes help a person to achieve his or her personal goal, the attribute is perceived to be more important. On the other hand, if the attribute does not help in achieving his or her personal goal, the attribute is perceived to be less important (Batra, Homer, & Kahle, 2001). According to the multiattribute model, an overall evaluation of a product is constructed as a combination of the attractiveness of each attribute (Green & Srinivasan, 1978; Khan & Meyer, 1991). In previous research, Vanhamme (2008) found that in the experience of a virtual museum, a comic book voucher was proposed unexpectedly in order to elicit the emotion of surprise. But the result did not reveal a significant difference between the experimental group and the control group. It might be because that the surprise element, which was a comic voucher in the experiment, was marginal and had less relevance to the main experience. Vanhamme (2008) suggested that the effect of surprise would be different according to marginality versus centrality in the relation with the main service in the research. Obviously, every attribute a consumer recognize is perceived differently according to its importance or relevance to personal goals. Relating this logic to customer delight, when consumers are delighted and the emotion of delight was aroused from an important and relevant attribute, then consumers will reveal stronger behavioral consequence than when they were delighted from a less important and less relevant attribute. Thus, suggested hypothesis is:
[H2] Delight that is elicited in important attribute will reveal higher (a) repurchase intention and (b) recommendation intention than when delight is elicited in less important attribute.
3. Methodology
3.1. Procedure
Since the element in eliciting the emotion of delight is important, a pretest was performed in order to select the delight element before the main experiment. The object for the main experiment was a virtual online book store, 2 delight elements were selected. First was a bookmark second was a coffee shop gift card and both had relevance with the book store. 22 university students were being asked about the preference of 2 bookmarks with gold and silver plated (total 5$ worth) and a well known coffee shop's gift card 5$ worth. I asked the respondents 'valuable' (7point scale) and the result was that the coffee shop gift card was significantly more valuable than the bookmark (Meangiftcard=5.5 vs. Meanbookmark=4.5, p<.1). Also, I asked the respondents 'Suppose that receiving a gift after the survey, I will be pleased to receive this gift' and the result was that the coffee shop gift card was higher than the book mark (Meangiftcard=6.0 vs. Meanbookmark=5.3, p<.1). Therefore the coffee shop gift card was selected as the element for eliciting the emotion of delight for the main study.
Next, the main study was performed with three groups: experiment group, reference group, and control group. Experiment group was the group in order to elicit delight with the emotion of surprise. Reference group was the group in order to elicit delight but without the emotion of surprise. Lastly, control group was the group with neither delight nor surprise. [Figure 1] describes the process of the study and the difference between groups. As in [Figure 1], for all the groups to have the same experience, respondents were informed that the purpose of the study was to find out the impression of a newly opened virtual online bookstore. The reason for not choosing a real brand was in order not to form a prior attitude toward the online book store. Respondents were presented with two pages of a colored printout. The first page was the main page of the online bookstore's website and the second page was about introduction of an e-book. The respondents looked carefully at the images of the online bookstore and answered the questionnaires. In experiment condition, after the first questionnaire, the respondents were provided a lottery draw and offered a coffee shop gift card 5$ worth only to the winners. In the study of Valenzuela et al. (2010), the authors found differences between the Asians and Westerners in reaction to an unexpected present. Asians usually reveal less feelings of emotion or pleasure out of an unexpected event, but if the Asians attribute the unexpected present to good luck, they revealed much positive surprise than the Westerners. Therefore, for the surprise to be more effective, a lottery draw was adapted in this experiment and told the respondents that the winning probability was 50%, but the actual winning rate was about 80%.
[Figure 1] Process of the study
For the reference group, in order not to elicit the emotion of surprise, the respondents were informed that they would be provided with a coffee shop gift card after the questionnaire is over and they were actually provided with a coffee shop gift card worth of 5$. Lastly for control group, the respondents were not informed that they would receive a gift card nor did they actually receive a gift card. The second questionnaire was common to all the groups, with little difference in the expression according to the group. The question items were joy, surprise and dependent variables such as purchase intention, recommendation intention and other demographic variables were included.
3.2. Measurement Variables
In order to measure the emotion of joy, the measurement items were 'I feel joy', 'I am pleased' (1='not at all', 7='very much') (Westbrook & Oliver, 1991; Izard, 1977). Because the experiment group received the gift card without prior notice, 3 questions were additionally asked, 'When I won the lottery for the gift card I was glad', 'When I won the lottery for the gift card I was very pleased', 'When I won the lottery for the gift card I felt very good', all in 7point scale. For the reference group that were noticed in advance that they would receive the gift card were asked additional 3 questions: 'When I received the gift card I was glad', 'When I received the gift card I was very pleased,' 'When I received the gift card I felt very good', all in 7point scale.
Next, to measure surprise, the item for control group was 'I was surprised before the second questionnaire' (1='not at all', 7='very much'). Since the characteristic of each group was different in expressing the emotion of surprise, only the basic expressions were remained. For experiment and reference group, 'I was surprised after receiving the gift card' (1=‘not at all’, 7=‘very much’). For experiment group, 'I was surprised when I won the lottery for the gift card', 'Before the second questionnaire, since I was informed about the gift card I was not surprised at the moment I won the lottery for the gift card'. For reference group, 'I was surprised when I received the gift card', 'Before the second questionnaire, since I was informed about the gift card I was not surprised at the moment I won the lottery for the gift card'
In order to measure attribute importance, Jaccard et al. (1986) measure was applied and modified into 'When buying a book at an online book store, how important is customer promotions (free gifts, gift cards, discount coupons, etc.) to you?' (7point scale, 1='not at all important', 7='very much important'). For dependent variables, repurchase intention was measured as 'When you have to buy a book next time, what is your intention to buy at this online book store?' (1='not at all buy at this store’, 5='very much willing to buy at this store', 5point scale). Recommendation intention was measured as 'When you recommend a book store, what is your intention to recommend this online book store? (1='not at all', 5='very much willing to recommend', 5point scale).
4. Results
4.1. Manipulation Check
Manipulation check was performed in order to check whether the groups were successfully manipulated in emotions of joy and surprise. The experiment group was highest in all the items about joy (see [Table 3]) and reference group was the second highest. Control group revealed the lowest among the groups (see [Table 3]). Independent sample t-test was performed to find significant difference between each group. As in [Table 4], all the items turned out to be significantly different among the groups (p<.1). The second manipulation check was to find out whether surprise was manipulated according to groups. The experiment group was highest in all the items about surprise (see [Table 5]) and reference group was the second highest. Control group revealed the lowest among the groups (see [Table 5]). Independent sample t-test was performed to find the significant difference between each group. As in [Table 6], all the items turned out to be significantly different among the groups (p<.01).
[Table 3] Mean of items of joy in each group
[Table 4] Differences in joy between groups (experiment group, reference group, control group)
[Table 5] Mean of items of surprise in each group
[Table 6] Differences in surprise between groups (experiment, reference, control)
4.2. Hypothesis Test
To test the hypothesis, repurchase intention for each group turned out to be Repurchaseexperiment=3.69, Repurchasereference =3.22, Repurchasecontrol=2.70 (See [Figure 2]). In order to test the difference between each group, LSD (Least Square Distance) post-hoc analysis of ANOVA was performed. The results was that each group had significant difference in repurchase intention (experiment vs. control: p=0.000, reference vs. control: p=0.008, experiment vs. reference: p=0.018) ([H1a] supported). Also, The recommendation intention for each group turned out to be Recommendationexperiment=3.53, Recommendationreference=3.06, Recommendationcontrol=2.38. LSD results were that each group had significant difference in recommendation intentions (experiment vs. control: p=0.000, reference vs. control: p=0.000, experiment vs. reference: p=0.006) ([H1b] supported). Therefore [H1] was supported. In order to find out whether there were differences between delighted-inimportant-attribute group and delighted-in-less importanceattribute group, first the experiment group and reference group were combined and the median point (5.5) of attribute importance was obtained. Then median split was done into attribute importance high group and attribute importance low group. In order to test the difference, an independent sample t-test was performed. For repurchase intention, the groups revealed significant difference (Repurchaseimportance_high=3.80 vs. Repurchaseimportance_low=3.13, t=3.553, p=0.001) (]H2a] supported). Recommendation intention was also significantly different between the groups (Recommendationimportance_high=3.46 vs. Recommendationimportance_low =3.13, t=1.971, p=0.052) ([H2b] supported). Therefore [H2] was supported.
[Figure 2] Results of repurchase intention (left), recommendation intention (right)
For further analysis, the experiment group that is with surprise was divided into attribute importance high (surprise high and importance high, SHIH) and attribute importance low (surprise high and importance low, SHIL). The reference group that was without surprise was also divided into attribute importance high (surprise low and importance high, SLIH) and attribute importance low (surprise low and importance low, SLIL). [Figure 3] shows the result of repurchase intention and recommendation intention of the 4 groups. As in [Figure 3] SHIH revealed the highest repurchase intention and recommendation intention among the groups. Also, SLIL revealed the lowest repurchase intention and recommendation intention among the groups.
[Figure 3] Results of surprise high/low and attribute importance high/low (Surprise High Importance
5. Discussion and Conclusion
5.1. Summary of Results
This research investigates the relationship between delight and behavioral consequences such as repurchase intentions and recommendation intentions. Also, the emotion of surprise and attribute importance was expected to play a moderating role in the relationship between delight and the dependent variables. The experiment was performed with virtual online bookstore to elicit the real emotion of surprise. Surprise was manipulated via experiment. The first group was experimental group which was delighted with surprise. Respondents were given a chance to draw a real lottery that the chance of winning was 50% (actually the real chance of winning was 80%). Some of the respondents who won the lottery did express facial surprise and joy and expressed their feelings by screaming in happiness. Second group was reference group that were delighted without surprise since they had notification to receive the gift in appreciation. And the last group was control group without delight nor surprise. As a result, the delighted with surprise group revealed higher repurchase intentions and recommendation intentions than the group that were delighted without surprise ([H1] supported). Also the delighted group that were elicited in important attribute had repurchase intention higher than the delighted group that were elicited in less important attribute. Also, recommendation intentions were higher for delighted group in important attribute than delighted group in less important attribute ([H2] supported).
5.2. Discussion and Implication
Despite the needs of research or information for customer delight from the fields and academia, there were not much research investigating the arena of the research of customer delight. This research gives several contributions by expanding the research of customer delight. Specifically, this research considered the assertion that delight has to be elicited with surprise vs. delight can be elicited without surprise. In this research, for consumers, it was found that the delighted emotion that was elicited with the emotion of surprise revealed higher repurchase intentions and recommendation intentions therefore backing up the assertion that the impact of delight to consumer behavior would be more effective when delight is accompanied with 'surprise'. In the experiment, the emotion of surprise was elicited by not giving prior information about the advantage that the respondents would get (e.g., winning the lottery and receiving the prize). Because the reference group in the experiment were noticed in advance that they would get the free gift, respondents actually had some expectation that they would receive the free gift. In real retail settings, firms can apply this information by providing consumers the advantage (for example coupon, special price discount, buy 1 get 1 free, free gifts, etc.) without notification for ‘unexpectedness’. An example to suggest is that stores may give out to their customers lotteries with high winning probability of a free gift.
For the second hypothesis, previous studies about customer delight did not consider whether delight was elicited in important attribute or less important attribute. Even though in Vanhamme (2008), suggested the consideration of the element of delight, whether the element is related to the main service or not, commented that the insignificant result was due to the irrelevance of the main product. This research tried to test the role of attribute importance on the impact of delight on behavioral consequences. For practical implication, in real retail settings, when marketers give advantage or special promotions to the customers, the certain objects should be important or relevant to customers. The gifts do not have to be high in monetary value as long as they are valuable to customers. To back up, recent research of hand made products that contain ‘love’ increases consumers’ intention to buy (Fuchs et al., 2015). What is thought as meaningful can also be thought as important. Therefore, there are various ways to provide for consumers in important attribute. The issue of customer delight can also apply to job satisfaction (Jung, 2018). Applying the result of this research, to make employees happier and more satisfied, tactics such as evoking surprisingness or focusing on important value would be effective in increasing job satisfaction.
5.3. Limitations and Ideas for Further Research
This research conducted an experiment to elicit the emotions with virtual online bookstore experience. Therefore it contains limitations that real situation and settings need to be explored. As mentioned in advance, a pretest was done to select the surprise element between a coffee shop gift card (e.g., Starbucks) and a bookmark. Strictly speaking, for the surprise element to be more relevant to the bookstore, the bookmark or other related surprise element should have been chosen. But, after the pretest result, though the gift card might be a less relevant surprise element, it was chosen because the respondents preferred it more. Therefore, a surprise element which is more relevant to the main product/service category should be considered in the future experiment. Also, the experiment was adapted and modified from Valenzuela et al. (2010) study to evoke the emotion of surprise precisely and effectively.
More Antecedents of customer delight should have considered such as interest, captivation, etc. For example, research in customer satisfaction have explored several antecedents of customer satisfaction (Coo, 2018). Therefore, in further research, investigating delight factors would be interesting and important in deepening the study of customer delight. Also, two moderating variables such as surprise and attribute importance were considered but other variables may be considered also. For example, variables controlling individual differences such as gender, personal tendencies such as regulatory focus may be considered. In this research, only moderating variables were considered but the psychological mechanism was not considered. Therefore mediating variables can be explored in further research. Strictly speaking, evoking surprise in other situations such as online shopping or TV home shopping (Lee et al., 2012) might not be viable than offline retail shopping. But marketing promotions that are original and attractive can evoke surprise and if the promotions are focused more on consumers wants then it would be effective in sustaining consumers.