Investigating the impact of surprise rewards on consumer responses

International Journal of Hospitality Management 50 (2015) 27–35

Contents lists available at ScienceDirect

International Journal of Hospitality Management journal homepage: www.elsevier.com/locate/ijhosman

Investigating the impact of surprise rewards on consumer responses Laurie Wu a,∗ , Anna S. Mattila b , Lydia Hanks c a

School of Tourism and Hospitality Management, Temple University, 1810 N 13th Street, Speakman Hall 303, Philadelphia, PA 19122, United States School of Hospitality Management, The Pennsylvania State University, 224 Mateer Building, University Park, PA 16802-1307, United States c Dedman School of Hospitality, The College of Business, Florida State University, 288 Champions Way, UCB 4114, Tallahassee, FL 32306-2541, United States b

a r t i c l e

i n f o

Article history: Received 23 November 2014 Received in revised form 3 June 2015 Accepted 10 July 2015 Keywords: Loyalty reward programs Reward type Discrete emotions Delight Cumulative satisfaction

a b s t r a c t Adopting an experimental approach, this research compared surprise rewards with membership discount rewards in terms of their impact on customer responses of delight, frustration and satisfaction. In addition, this research examined the circumstances under which surprise rewards should be offered in order to yield maximum benefits for hospitality firms. In particular, the study examined how the customer’s cumulative satisfaction (high vs. low) influences the effectiveness of surprise rewards (vs. membership discount rewards) in increasing customer delight and satisfaction and decreasing customer frustration. Consistent with the theoretical predictions, results show that surprise rewards are more effective than membership discount rewards for enhancing customer delight and satisfaction and attenuating customer frustration, particularly when the customer’s cumulative satisfaction is low. These findings have important implications for the hospitality industry. Hospitality managers and marketers could use this information to design effective loyalty reward programs. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Loyalty reward programs have become extremely popular in recent decades, and they are a subject of great interest to both practitioners and scholars (Henderson et al., 2011; Meyer-Waarden and Benavent, 2012; Shoemaker and Lewis, 1999; Yoo and Bai, 2013). According to Ferguson and Hlavinka (2007), in the United States, loyalty reward program participation has topped 1.3 billion, with the average household subscribing to 12 separate programs. By providing rewards such as discounts on purchases or points toward a free purchase, firms aim to entice their most valuable customers to make repeat purchases (Meyer-Waarden and Benavent, 2012). Despite their pervasiveness, academic research findings diverge on the value of loyalty reward programs (Dorotic et al., 2011; Dowling and Uncles, 1997; Leenheer et al., 2007; Shugan, 2005). On the one hand, previous research shows that loyalty reward programs can effectively enhance customers’ value perceptions, satisfaction, retention, willingness to pay price premiums and share of wallet (Bolton et al., 2000; Keh and Lee, 2006; Leenheer et al., 2007; Sharp and Sharp, 1997; Verhoef, 2003). On the other hand, empirical evidence indicates that loyalty reward programs are not that powerful in boosting market share (Meyer-Waarden and Benavent, 2006,

∗ Corresponding author. E-mail addresses: [email protected] (L. Wu), [email protected] (A.S. Mattila), [email protected] (L. Hanks). http://dx.doi.org/10.1016/j.ijhm.2015.07.004 0278-4319/© 2015 Elsevier Ltd. All rights reserved.

2009). Further, some research findings suggest that rewarding loyal customers with discount prices may erode future business profits (Shugan, 2005). Amid this debate, marketing scholars continue to pay attention to the effectiveness of program designs (Dowling and Uncles, 1997; Hu et al., 2010; Kivetz and Simonson, 2003; Leenheer et al., 2007; Liu and Yang, 2009; Nunes and Dréze, 2006; Wagner et al., 2009). The key question of concern is which type of loyalty rewards is most effective in yielding desired marketing outcomes (Jang and Mattila, 2005; Meyer-Waarden and Benavent, 2012). Previous research shows that reward structure factors such as reward type and reward tiers can greatly influence customer preferences and loyalty and that customers prefer direct rewards over indirect rewards (e.g., Keh and Lee, 2006; Tanford, 2013). In addition, Hu et al. (2010) revealed that, if customers are satisfied with hotel experience, immediate rewards are more effective than delayed rewards in building program loyalty and customer loyalty. This research examines yet another type of loyalty rewards: surprise rewards. Surprise rewards can be defined as unexpected incentives firms provide to their loyal customers. In the extremely competitive global hospitality industry, more and more firms are strategically aiming to delight, rather than merely satisfy their customers (Kim and Mattila, 2010; St. James and Taylor, 2004; Vanhamme and de Bont, 2008). Accordingly, it is critically important for hospitality firms to find an effective way to “woo” their loyal customers with rewards (Meyer-Waarden and Benavent, 2012) – that is, to delight loyal customers with surprise rewards.

28

L. Wu et al. / International Journal of Hospitality Management 50 (2015) 27–35

As such, this research argues that surprise rewards will boost loyal customers’ service experiences beyond their expectations, thus leading to an even higher level of customer satisfaction (Keiningham and Vavra, 2001). In addition, this research examines the circumstances under which surprise rewards are most effective. By linking a long-term relationship construct, cumulative satisfaction, to the research context, the current study aims to understand how cumulative satisfaction moderates the differential impact of surprise rewards (vs. membership discount rewards) on customers’ affective and evaluative responses. Cumulative satisfaction can be defined as a customer’s overall evaluation of a product or service provider to date (Johnson et al., 1995; Johnson and Fornell, 1991). When interacting with a particular service provider, customers who have experienced bad service in the past are more likely to feel frustrated and dissatisfied with their current experience. Will surprise rewards be able to reverse the influence of low cumulative satisfaction and reduce customers’ feelings of frustration and dissatisfaction toward the service provider? The current study explores these research questions in the context of hospitality loyalty reward programs. 2. Theoretical background The current research examines the joint impact of rewards type (surprise rewards vs. membership discount rewards) and cumulative satisfaction (low vs. high) on customers’ on-site responses of delight, frustration and satisfaction. In the next section, definitions of the three key dependent variables (i.e., delight, frustration and satisfaction) will be offered. Then, the anchoring and adjustment framework (Tversky and Kahneman, 1974) will be presented as a theoretical lens for this research. 2.1. Customer Delight, Frustration and Satisfaction Delight – In consumer behavior research, delight is often recognized as a positive emotional state that is above and beyond satisfaction (Füller and Matzler, 2008; Loureiro and Kastenholz, 2011; Plutchik, 1980, 2003). Compared with satisfaction, delight is unique in the sense that it involves the emotional responses of surprise and joy (Plutchik, 1980, 2003). A positive (pleasant) surprise in the consumer context is considered the necessary condition for, and is most often associated with, eliciting customer delight (Finn, 2005; Oliver et al., 1997; Rust and Oliver, 2000; Vanhamme and de Bont, 2008). The current research argues that, compared with a contract-based membership discount reward, a surprise reward includes a positive surprise element, thus leading to a delightful, rather than just a satisfactory experience (Füller and Matzler, 2008; Füller et al., 2006; Valenzuela et al., 2010). Frustration – By definition, consumer frustration is a highly negative emotion that can be attributed to interference with a potentially satisfying sequence of acts or behaviors (Berezan et al., 2015; Stauss et al., 2005). The notion of frustration is quite relevant to the context of service relationship management. Of particular relevance to this research is the work of Berezan et al. (2015) who discuss consumer frustration with loyalty rewards programs. Based on a content analysis of 1519 comments from members of five major hotel rewards programs, Berezan et al. (2015) found that managing customer frustration is paramount. Given a history of poor service episodes, customers are highly likely to experience frustration (Berezan et al., 2015). If these negative feelings are not effectively managed, problems may quickly escalate via the rapid spread of electronic word-of-mouth (Berezan et al., 2015). The current research builds on this work and argues that a surprise reward might reduce frustration, as unexpected incentives can break the

contextual consistency, thus providing an opportunity to establish new impressions of the service provider (Henderson et al., 2011; Ji and Wood, 2007; Wood and Neal, 2009). Satisfaction – While cumulative satisfaction is a customer’s overall evaluation of a service provider based on all previous consumptions (Johnson et al., 1995; Johnson and Fornell, 1991), the dependent variable of this research capture customers’ evaluations of the current consumption experience. Previous research show that emotions play a primary role during customers’ evaluation processes (Mattila, 2006; Wirtz et al., 2000; Zajonc, 1984). As such, it is argued here that the surprise element heightens the intensity of affective responses, thus leading to more favorable on-site evaluative responses such as satisfaction (Palmatier et al., 2009; Valenzuela et al., 2010). 2.2. The moderating impact of cumulative satisfaction Relying on Tversky and Kahneman’s (1974) anchoringadjustment framework, the current research further proposes that cumulative satisfaction will moderate the effects described above. Tversky and Kahneman (1974) postulate that customers use relevant information as an anchor for subsequent evaluations of the same stimulus. Prior research in customer satisfaction shows that, in repeat-consumption situations, consumers primarily rely on prior satisfaction judgments to evaluate their current experiences and engage in adjustment, or judgment updating processes, only when faced with unexpected service experiences (Mattila, 2003). Translated to the current research context, it is argued that contract-based discounts (e.g., a membership discount reward) are expected, and consequently, cumulative satisfaction serves as the anchor when evaluating the current experience. Conversely, surprise rewards are unexpected, thus reducing the consumer’s reliance on the anchor of cumulative satisfaction and hence an adjustment process is needed (Mattila, 2003). The nature of the adjustment process further hinges on the congruity between the anchor and the target of evaluation. Depending on the anchor-target congruity, the target evaluation will be adjusted either toward (i.e., assimilation effect) or away from (i.e., contrast effect) the anchor (Bohner et al., 2002; Chernev, 2011; Davis et al., 1986; McFerran et al., 2010; Wansink et al., 1998; Yadav, 1994). Previous research found that when the anchor and the target belong to the same domain (i.e., congruent by nature), the adjustment process reflects an assimilation pattern (Bohner et al., 2002; Chernev, 2011). When the anchor and the target belong to opposite domains (i.e., incongruent by nature), however, the adjustment process shows a contrast pattern (Bohner et al., 2002; Chernev, 2011). For example, in the context of vice vs. virtue foods, Chernev (2011) found that the sequential evaluation of calorie content is driven by an anchoring-adjustment process. When the target food and the anchor belong to opposite domains (e.g., evaluating a virtue food such as a green salad after being exposed to a vice food such as a chocolate cake), the calorie estimation of the target will be distanced away from the anchor (i.e., contrast effect). Conversely, when the target and the anchor belong to the same domain (e.g., evaluating French fries after being exposed to a chocolate cake), the calorie estimation of the target will be similar to that of the anchor (i.e., assimilation effect). Such an anchoring-adjustment framework also applies to the current research. In the current research, cumulative satisfaction serves as the anchor when evaluating the current consumption experience involving either a membership discount reward or a surprise reward. As a membership discount reward is expected, cumulative satisfaction serves as the anchor to guide the consumer’s evaluation of the current service experience. On the other hand, surprise rewards are unexpected (Ji and Wood, 2007; Wood and Neal, 2009), thus reducing the consumer’s reliance on

L. Wu et al. / International Journal of Hospitality Management 50 (2015) 27–35

cumulative satisfaction, leading to an adjustment process. The valence of cumulative satisfaction (high or low), in turn, will determine whether the adjustment follows an assimilation or a contrast pattern. When cumulative satisfaction is high, the positive nature of high cumulative satisfaction is congruent with the positive nature of surprise rewards and consumer responses will be assimilated toward the anchor. Similarly, the expected nature of membership discount reward will lead to evaluation processes heavily guided by high cumulative satisfaction (Johnson et al., 1995). Thus, when cumulative satisfaction is high, regardless of the type of reward received, customers will exhibit similar levels of responses such as delight, frustration or satisfaction. On the other hand, when cumulative satisfaction is low, the negative valence is incongruent with the positive nature of surprise rewards. Hence, consumer responses will be contrasted away from the anchor (Chernev, 2011; McFerran et al., 2010; Tversky and Kahneman, 1974), thus resulting in exceptionally high levels of delight and satisfaction and particularly low levels of frustration. Meanwhile, the anchoring effect will hold with membership discount rewards. In other words, under conditions of low cumulative satisfaction, a surprise reward (vs. membership discount rewards) will lead to significantly more favorable outcomes such as high levels of delight and satisfaction and low levels of frustration. Taken together, the following hypotheses are proposed: H1 . When cumulative satisfaction is low, a surprise reward (vs. a membership discount reward) leads to a higher level of customer delight. H2. When cumulative satisfaction is high, a surprise reward and a membership discount reward lead to similar levels of customer delight. H3. When cumulative satisfaction is low, a surprise reward (vs. a membership discount reward) leads to a lower level of customer frustration. H4. When cumulative satisfaction is high, a surprise reward and a membership discount reward lead to similar levels of customer frustration. H5. When cumulative satisfaction is low, a surprise reward (vs. a membership discount reward) leads to a higher level of customer satisfaction. H6. When cumulative satisfaction is high, a surprise reward and a membership discount reward lead to similar levels of customer satisfaction. 3. Methodology 3.1. Design This study employed a 2 (reward type: surprise reward vs. membership discount reward) by 2 (cumulative satisfaction: low vs. high) between-subjects experimental design. Participants were randomly assigned to one of the four experimental conditions: (a) high level of cumulative satisfaction and a surprise reward, (b) low level of cumulative satisfaction and a surprise rewards, (c) high level of cumulative satisfaction and a membership discount reward, and (d) low level of cumulative satisfaction and a membership discount reward. 3.2. Sampling Amazon Mechanical Turk was used to recruit adult consumers in the US to participate in this study. Amazon Mechanical Turk

29

(MTurk) is a commonly used online consumer panel for data collection in social sciences research, and the MTurk participants are highly similar to traditional samples (Buhrmester et al., 2011; Goodman et al., 2012; Paolacci et al., 2010). The research introduction clearly stated that the study involved restaurant loyalty reward programs. Following the recruitment message and the informed consent, participants were directed to a scenario. Participants were paid $.50 USD for their participation. 3.3. Stimuli The participants were asked to imagine that they were members of a frequent diner club at a moderately priced sit-down restaurant. To manipulate high vs. low levels of cumulative service satisfaction, participants were asked to imagine that they had experienced generally either positive or negative service experiences at the restaurant in the past. This manipulation is in line with Johnson et al. (1995) conceptualization of cumulative satisfaction as the depiction of consumers’ total consumption experience with a service. To manipulate the reward type, participants were further asked to imagine that, during tonight’s dining experience, they received either a surprise reward (“You’re the random reward winner today for a free dessert!”) or a membership discount reward (“You get 10% off with your frequent diner club card”; please see Appendix for the study stimuli). After reading the scenario, participants completed a survey. 3.4. Measurement Delight and frustration were measured with 7-point bipolar scales (1 = wouldn’t feel like this at all, 7 = would feel like this very much). Adapted from Finn’s (2005) three item scale, measures for delight include two items: elated and delighted (r = 0.81, p < 0.001). As for frustration, the measures added the item “angry” in addition to Valenzuela et al.’s (2010) single item scale for frustration – “frustrated” (r = 0.85, p < 0.001). Following Wu et al. (2014), the current research measured satisfaction with a four-item semantic differential scale adapted from Westbrook and Oliver (1981). Specifically, participants were asked to indicate how they felt after reading about a dining experience by choosing between each of the four sets of semantic differential measures: dissatisfied–satisfied, unhappy–happy, poor job–good job and poor choice–good choice (Cronbach’s ˛ = 0.97). As previous research indicated that involvement in the product category influences consumers’ evaluations (Finn, 2005; Russell-Bennett et al., 2007), this variable was measured in this study as a covariate. Consumer involvement in the product category was measured using 7-point Likert scales (1 = very strongly disagree, 7 = very strongly agree) for three items adapted from Suh and Youjae (2006). While Suh and Youjae’s (2006) scale touched on the aspects of importance, meaningfulness, necessity, value, interestingness, want and relatedness, the current measurement scale focused on the aspects of importance, meaningfulness and necessity: I depend upon this service category a great deal, This service category means a lot to me, and Compared to most products/services I buy, purchasing a meal in this type of restaurant is a really important purchase (Cronbach’s so = 0.90). Manipulation checks were placed at the end of the questionnaire. To check the manipulations for surprise rewards, participants were asked to what extent they felt surprised by the reward (1 = not surprised at all, 7 = very surprised). To check the manipulations for cumulative satisfaction, participants were asked to answer the following question: “Reflecting on your past experiences with this restaurant, how pleased or displeased do you feel? (1 = displeased, 7 = pleased).” Scenario realism was measured with two items: How realistic are the scenarios? On a 7-point bipolar scale (1 = not realistic at all, 7 = very realistic) and How easy is it for you to imagine

30

L. Wu et al. / International Journal of Hospitality Management 50 (2015) 27–35

Table 1 Summary of major measures.

6.0

Items

Reliability

5.5

Delight

1. Elated 2. Delighted

r = 0.81 (p = 0.000)

5.0

Frustration

1. Frustrated 2. Angry

r = 0.85 (p = 0.000)

4.5

Satisfaction

1. Dissatisfied–satisfied 2. Unhappy–happy 3. Poor job–good job 4. Poor choice–good choice

Cronbach’s ˛ = 0.97

Involvement in the product category

1. I depend upon this service category a great deal 2. This service category means a lot to me 3. Compared to most products/services I buy, purchasing a meal in this type of restaurant is a really important purchase

Delight

Variable

4.0 3.5

Cronbach’s ˛ = 0.90

3.0 2.5 2.0

Discount Reward

Surprise Reward

Low Cumulative Service Satisfaction High Cumulative Service Satisfaction

yourself in the scenarios? On a 7-point bipolar scale (1 = very difficult, 7 = very easy, r = 0.62, p < 0.001). Demographic information (e.g., gender, age, household income, etc.) was also collected from the participants. The major measures of the current research are summarized in Table 1. 4. Results 4.1. Demographics Based on the 2 × 2 between-subject design of this research and the rule of thumb of 30 participants per cell (Kerlinger and Lee, 2000), 120 responses were collected and a total of 111 complete and usable responses were received. The average age of participants was 42, ranging from 18 to 76 years. The gender split was 40% male and 60% female. Approximately 41% of the participants had a college degree and approximately 30% reported a household income over $60,000. Sixty-seven percent of the respondents were Caucasian. The participants reported dining out an average of 5 times during the past month. 4.2. Manipulation check Results of manipulation check showed that participants in the free dessert condition were significantly more surprised than their counterparts in the 10% discount condition (Msurprise = 5.73 and Mdiscount = 4.47, F(1,110) = 11.87, p < 0.01, partial 2 = 0.106). As expected, participants in the high cumulative satisfaction condition were significantly more pleased by the restaurant than their counterparts in the low cumulative satisfaction condition (Mhighcumsat = 6.37 and Mlowcumsat = 3.02, F(1,110) = 116.53, p < 0.001, partial 2 = 0.517). Scenarios were perceived as realistic (M = 5.13) and easy to comprehend (M = 5.40), and no significant differences existed among the experimental conditions. 4.3. Hypotheses testing To test the proposed hypotheses, a series of 2 (reward type: surprise reward vs. membership discount reward) by 2 (cumulative satisfaction: low vs. high) full-factorial analysis of covariance (ANCOVA) were performed separately on the three dependent

Fig. 1. Interaction plot for delight.

variables: delight, frustration and satisfaction. Following previous research in similar domains (Lacey et al., 2007; Mattila et al., 2013), participants’ gender, age, product involvement and frequency of dining out were controlled as covariates in the analysis. The cell means for delight, frustration and customer satisfaction are shown in Table 2. Delight. ANCOVA results for delight indicate that the main effects of reward type (Msurprise = 4.91, Mdiscount = 3.74; F(1,100) = 5.73, p < 0.05, partial 2 = 0.054) and cumulative satisfaction (Mhighcumsat = 5.66, Mlowcumsat = 3.07; F(1,100) = 55.27, p < 0.001, partial 2 = 0.356) are both significant (see Table 3). As predicted, the interaction effect of reward type and cumulative satisfaction is significant (F(1,100) = 4.55, p < 0.05, partial 2 = 0.043). Consistent with previous literature (Finn, 2005), the covariate effects of gender (F(1,100) = 11.14, p < 0.01, partial 2 = 0.100) and product involvement (F(1,100) = 4.96, p < 0.05, partial 2 = 0.047) are both significant. The interaction effect is displayed in Fig. 1. Given a low level of cumulative satisfaction, a surprise reward (vs. a discount reward) led to a significantly higher level of delight (Msurprise = 3.75, Mdiscount = 2.53; F(1,100) = 10.50, p < 0.01, partial 2 = 0.095; H1 supported). Conversely, in the high cumulative satisfaction condition, the two reward types results in similar levels of delight (Msurprise = 5.72, Mdiscount = 5.55; F(1,100) = 0.04, p = 0.843, partial 2 = 0.000; H2 supported). Frustration. Results of the ANCOVA test on frustration, shown in Table 4, indicate that the main effects of reward type (Msurprise = 2.41, Mdiscount = 3.62; F(1,100) = 6.18, p < 0.05, partial 2 = 0.058) and cumulative satisfaction (Mhighcumsat = 1.58, Mlowcumsat = 4.35; F(1,100) = 71.65, p < 0.001, partial 2 = 0.417) are both significant. The interaction effect of cumulative satisfaction and reward type is also significant (F(1,100) = 4.49, p < 0.05, partial 2 = 0.043). None of the covariate effects are significant. As shown in Fig. 2, when the level of cumulative satisfaction is low, a surprise reward (vs. a discount reward) resulted in a significantly lower level of frustration (Msurprise = 3.65, Mdiscount = 4.92; F(1,100) = 10.87, p < 0.01, partial 2 = 0.098; H3 supported). In the high cumulative satisfaction condition, the level of frustration was similar regardless of the reward type (Msurprise = 1.53,

L. Wu et al. / International Journal of Hospitality Management 50 (2015) 27–35

31

Table 2 Means and standard deviations of the results. Experimental manipulation

Outcomes (N = 111) Delight Mean (SD)

Frustration Mean (SD)

Satisfaction Mean (SD)

High cumulative service satisfaction

Surprise reward Discount reward

5.72 (1.44) 5.55 (1.34)

1.53 (1.15) 1.68 (1.23)

6.43 (0.88) 6.10 (1.25)

Low cumulative service satisfaction

Surprise reward Discount reward

3.75 (1.68) 2.53 (1.70)

3.65 (1.57) 4.92 (1.80)

3.48 (1.64) 2.67 (1.67)

Table 3 ANCOVA table for delight. Source

Type III SS

Covariates Age Gender Involvement in the product category Previous times of dining out Test effects Reward type Cumulative service satisfaction Interaction effect Error Total

DF

MS

Partial 2

1 1 1 1

3.780 24.125 10.739 0.007

1.746 11.142 4.960 0.003

0.189 0.001 0.028 0.955

0.017 0.100 0.047 0.000

12.407 119.677 9.840 216.522 2509.250

1 1 1 100 108

12.407 119.677 9.840 216.522 2509.250

5.730 55.272 4.545

0.019 0.000 0.035

0.054 0.356 0.043

Mdiscount = 1.68; F(1,100) = 0.074, p = 0.786, partial 2 = 0.001; H4 supported). Satisfaction. Results for satisfaction suggest that the main effects of reward type (Msurprise = 5.21, Mdiscount = 4.04; F(1,100) = 5.99, p < 0.05, partial 2 = 0.056) and cumulative satisfaction (Mhighcumsat = 6.31, Mlowcumsat = 3.03; F(1,100) = 118.02, p < 0.001, partial 2 = 0.541) are both significant (see Table 5). Consistent with previous research (Finn, 2005; Russell-Bennett et al., 2007), the covariate effects of gender (F(1,100) = 4.11, p < 0.05, partial 2 = 0.039) and product involvement (F(1,100) = 9.12, p < 0.01, partial 2 = 0.084) are both significant. The interaction effect of reward type and cumulative service satisfaction is not significant (F(1,100) = 1.54, p = 0.218, partial 2 = 0.015). However, a closer examination of the contrast effects finds supportive evidence for the theoretical predictions. The results indicate that a surprise reward (vs. a discount reward) led to a significantly higher level of customer satisfaction only when cumulative satisfaction was low (Msurprise = 3.48, Mdiscount = 2.67; F(1,100) = 6.98, p < 0.05, partial 2 = 0.065; H5 supported). On the other hand, in the high cumulative satisfaction condition, the two reward types resulted in similar levels of satisfaction with the current experience (Msurprise = 6.43, Mdiscount = 6.10; F(1,100) = 0.73, p = 0.394, partial 2 = 0.007; H6 supported).

5.0 4.5

Frustration

p-value

3.780 24.125 10.739 0.007

5.5

4.0 3.5 3.0 2.5 2.0 1.5 1.0

Discount Reward

F

Surprise Reward

Low Cumulative Service Satisfaction High Cumulative Service Satisfaction

Fig. 2. Interaction plot for frustration.

Table 4 ANCOVA table for frustration. Source Covariates Age Gender Involvement in the product category Previous times of dining out Test effects Reward type Cumulative service satisfaction Interaction effect Error Total

Type III SS

DF

MS

F

p-value

Partial 2

0.308 7.187 5.358 0.189

1 1 1 1

0.308 7.187 5.358 0.189

0.146 3.414 2.545 0.090

0.703 0.068 0.114 0.765

0.001 0.033 0.025 0.001

13.012 150.829 9.450 210.513 1402.250

1 1 1 100 108

13.012 150.829 9.450 2.105

6.181 71.648 4.489

0.015 0.000 0.037

0.058 0.417 0.043

32

L. Wu et al. / International Journal of Hospitality Management 50 (2015) 27–35

Table 5 ANCOVA table for satisfaction. Source Covariates Age Gender Involvement in the product category Previous times of dining out Test effects Reward Type Cumulative service satisfaction Interaction effect Error Total

Type III SS

DF

MS

F

p-value

Partial 2

3.502 7.178 15.933 0.142

1 1 1 1

3.502 7.178 15.933 0.142

2.004 4.107 9.116 0.081

0.160 0.045 0.003 0.776

0.020 0.039 0.084 0.001

10.463 206.290 2.685 174.787 2851.760

1 1 1 100 108

10.463 206.290 2.685

5.986 118.023 1.536

0.016 0.000 0.218

0.056 0.541 0.015

In sum, these results suggest that, when cumulative satisfaction is low, compared with a discount reward, a surprise reward leads to a higher level of customer delight (H1 supported), a lower level of customer frustration (H3 supported) and a higher level of customer satisfaction (H5 supported). Conversely, reward type failed to influence consumer responses when cumulative satisfaction was high (H2 , H4 and H6 supported). Altogether, all of the proposed hypotheses are supported.

5. Discussion and theoretical contributions Despite their prevalence and popularity, the effectiveness of loyalty reward programs remains controversial. Against a backdrop of scholars questioning whether “customer loyalty programs really work” (Dorotic et al., 2011; Dowling and Uncles, 1997; Leenheer et al., 2007; Meyer-Waarden and Benavent, 2006, 2009; Shugan, 2005), marketers strive to improve the effectiveness of loyalty reward programs by optimizing program designs and by diversifying rewards offerings (Dowling and Uncles, 1997; Kivetz and Simonson, 2003; Leenheer et al., 2007; Liu and Yang, 2009; Nunes and Dréze, 2006; Wagner et al., 2009). To contribute to this stream of literature, the current research examined the impact of yet another type of reward: surprise rewards. This study further examined the impact of cumulative satisfaction on customers’ reactions to surprise rewards (vs. discount rewards). Although some previous research has explored the relationship between emotions and loyalty behaviors (e.g., Barsky and Nash, 2002), a review of research found that the role of discrete emotions (e.g., delight, anger, etc.) has been largely ignored in the context of loyalty rewards programs. More importantly, to date, no research yet looked into the association of a specific type of discrete emotions and reward type in the hospitality marketing literature. To the best of the authors’ knowledge, the current study is the first to examine such a topic in the hospitality marketing literature. The results of the current research also make a novel contribution to the literature on loyalty rewards program effectiveness. The findings of the current research revealed that, compared with discount rewards, surprise rewards can enhance customer delight and satisfaction and lower customer frustration when given a low level of cumulative satisfaction. This result is consistent with the notion that randomness of incentives could suppress habit formation and break the contextual consistency, thus providing an opportunity for consumers to form new perceptions (Ji and Wood, 2007; Wood and Neal, 2009). On the other hand, along with previous research, the current study confirmed the guiding impact of cumulative customer satisfaction on customers’ evaluations of an on-going service experience (Johnson et al., 1995; Johnson and Fornell, 1991). Specifically, the results of this study showed that, when given a high level of cumulative satisfaction, surprise rewards are not any

better than discount rewards in enhancing customer evaluations for an on-going service experience. It is also important to recognize that, although the surprise reward did improve customer evaluations in low cumulative satisfactions conditions, the ratings were still relatively low and below the midpoint on a seven point scale. As such, it should be inferred that surprise rewards are not sufficient to completely eradicate the effects of previous poor service deliveries. It is also important to note that though the interaction effects were statistically significant, the main effects of cumulative satisfaction were still much stronger than any of the other effects. Such findings indicate that while surprise rewards may, to some extent, increase delight and satisfaction and decrease frustration, cumulative satisfaction is still the decisive factor for consumers’ evaluations for an on-going service experience. These results further confirmed the theoretical predictions of Tversky and Kahneman’s (1974) framework that, though adjustment processes do occur from time to time, once established, an anchoring effect generates a sticky impact on customer evaluations (Han et al., 2011; Mattila, 2003). Focused on customers’ on-site responses to surprise rewards, this research also provides some important implications for long-term marketing outcomes such as customer loyalty. The findings of this research revealed that surprise rewards can effectively reverse the negative impact of low cumulative satisfaction on customers’ evaluations of the current experience. Surprise rewards might thus enhance customer loyalty despite past bad service experiences. Since previous research shows that on-site service satisfaction contributes to long-term loyalty intention (Han et al., 2008), the effect of surprise rewards (vs. discount rewards) might positively influence customers’ long term loyalty intentions. In addition, as the memory of a delightful experience may last longer than the memory of merely a satisfying experience (Rust and Oliver, 2000), the positive experience of a surprise reward may last longer in a customer’s memory, thus yielding long-term behavioral benefits. Further, recent marketing research showed that interesting consumption experiences get more positive word of mouth in online consumer forums (Berger and Schwartz, 2011). Since service experiences with surprise rewards are generally considered more novel and interesting than those with discount rewards, surprise rewards may create yet another positive marketing outcome for hospitality firms. 6. Managerial implications Most loyalty reward programs continue to be based on “a one-size-fits-all approach” (Meyer-Waarden and Benavent, 2012) that relies on membership-based monetary rewards such as product discounts to encourage repeat purchases. Based on the

L. Wu et al. / International Journal of Hospitality Management 50 (2015) 27–35

findings of this study, it is recommended that hospitality managers should implement another reward option for their loyal customers – surprise rewards. While this practice has been adopted by some insightful managers, there is lack of empirical evidence that elucidates the positive outcome of surprise rewards on consumer experiences. The current research reveals that for customers with low cumulative satisfaction, surprise rewards are more effective than discount rewards in creating feelings of delight and enhanced satisfaction with the current consumption experience. In other words, surprise rewards might buffer the negative impact of past bad experiences. These findings are particularly relevant to businesses suffering from relatively low satisfaction ratings. For such units, implementing surprise rewards might offer a chance to stimulate business, enhance customer satisfaction and rebuild market reputation. Recent technological advancements have made it easy to implement surprise type rewards. For instance, some restaurant loyalty rewards apps allow customers to win random prizes such as free or discounted menu items when they use smart phones to scan a QR code at the restaurant entrance or at a dining table (Jargon, 2013). Hospitality managers are recommended to emphasize the surprise component with such random app rewards with the help of delightful animation and esthetic designs. There are also other ways in which loyalty rewards can be provided in a surprising way. Panera Bread, for example, designed a loyalty program named MyPanera, which operates solely through surprise and delight. MyPanera customers are often surprised at the register with unexpected and delightful rewards such as free bakery items, and such a reward approach has helped the firm generate quite some positive buzz on the internet (Bruell, 2012). In addition, Expedia offered $100 coupon to a random group of its Elite Plus Program members. Although only 10 percent of those receiving the coupon claimed it in the end, such a delightful reward increased the entire group’s transaction with Expedia by nearly 10 percent (Whitehurst et al., 2014). In addition, the findings suggest that hospitality managers should consider the customer’s cumulative satisfaction when implementing surprise rewards. As compared with those customers who hold high levels of cumulative satisfaction, customers with low levels of cumulative satisfaction are more likely to be influenced by surprise rewards when evaluating a current service experience. While such findings may suggest that managers implement surprise rewards to please those customers who are dissatisfied with the business in the past, it is important to bring up an important issue for managers to be cautious about – fairness perception. Specifically, it should be clarified that this research does not recommend that managers segment customer groups and offer surprise rewards only to the dissatisfied customers. If satisfied customers ever become aware of the existence of such managerial actions, a sense of unfairness could emerge and drive the satisfied customers away. While the results demonstrate the benefits of surprise rewards, it is important to note that this research is not questioning the effectiveness of discount rewards. A discount reward might be more effective than a surprise reward in certain contexts. For example, for some chain restaurants, offering a surprise reward might conflict with the company’s belief in standardized service. In such circumstances, a surprise reward might confuse loyal customers. Further, consumers from different cultural backgrounds might react to the same surprise reward differently. In certain cultures, a surprise reward may be perceived more negatively (Valenzuela et al., 2010). In sum, hospitality firms are recommended to consider employing a combination of expected discount rewards and unexpected surprise rewards in the design of their loyalty programs, after a strategic assessment of the pros and cons of both rewards types.

33

7. Limitations and future research As with any study, this research has several limitations. First, the scope of this study was limited to customers’ on-site responses to a reward offer. As one of the major aims of loyalty programs is to retain valuable customers, it is critical for businesses to understand how, over time, reward type influences consumers’ affective and behavioral evaluations such as intention to return or recommend to others. Second, discount rewards can be just as “surprising” depending on how they are presented. For example, presenting discount rewards with exciting design cues (e.g., visually stimulating colors and pictures) and service cues (e.g., announcing the reward using a very positive tone of voice) may add a sense of surprise to the customer experience as well. Future research should explore this possibility. Third, consistent with previous literature (Kim and Mattila, 2010), the current study operationalized the surprise reward as a free dessert. Such a surprise reward might only work for guests who like desserts. Further, it is important to note that the female and male genders were not equally represented in the sampled participants of this study. Considering the significant covariate effects of gender in the test models, it would be important for future research to examine if there is a gender-based difference in consumer responses to emotional stimuli such as surprise rewards. Last but not least, the scenario-based experiment method failed to capture consumers’ actual loyalty behaviors and the findings were limited to one context (i.e., restaurant). To that end, field studies are needed to directly observe the joint impact of reward type and cumulative satisfaction on customers’ onsite loyalty behaviors. It would be quite interesting to examine how surprise rewards operate at different tier levels and in other industry segments such as hotel or casino where loyalty programs are more common. Appendix. Research stimuli A.1. Cumulative service satisfaction A.1.1. High level of cumulative service satisfaction Assume that you are a customer of PotterHouse Restaurant, which is rated favorably by food critics in terms of its food quality and reasonable prices. It serves fresh seafood, which is popular with you and your family. During your past few visits, they were able to seat you quickly, even when you had not made a prior reservation. The service has been very good, and they serve a variety of fresh seafood, which has always been delicious. This evening, when you got there, you were seated right away. You had a pleasant and delicious dinner, and your server was prompt and friendly. Now that you have finished your entrée and your dishes have been cleared, your waiter returns to your table. A.1.2. Low level of cumulative service satisfaction Assume that you are a customer of PotterHouse Restaurant, which is rated favorably by food critics in terms of its food quality and reasonable prices. It serves fresh seafood, which is popular with you and your family. On your past few visits, you were kept waiting for about 15 min upon arrival, even after you had made a reservation. You have had rather poor service before at this restaurant, and the food quality has been pretty inconsistent. This evening, you had to remind the server twice before he served you the main course, even though you had long finished the soup. The food was mediocre, and your server was rather inattentive. Now that you have finished your entrée and your dishes have been cleared, your waiter returns to your table.

34

L. Wu et al. / International Journal of Hospitality Management 50 (2015) 27–35

A.2. Reward type A.2.1. Surprise reward The server smiles, and says, “You’re the random reward winner today! Every day we give out a free dessert to one of our frequent diner club members, and today, it looks like it’s you.” A.2.2. Discount reward The server hands you the check, smiles, and says, “Ok, you get 10% off with your frequent diner club card. I’ll take care of this whenever you’re ready.” References Barsky, J., Nash, L., 2002. Evoking emotion: Affective keys to hotel loyalty. Cornell Hotel Restaur. Adm. 43 (1), 39–46. Berezan, O., Raab, C., Tanford, S., Kim, Y.S., 2015. Evaluating loyalty constructs among hotel reward program members using eWOM. J. Hosp. Tour. Res. 39 (2), 198–224. Berger, J., Schwartz, E.M., 2011. What drives immediate and ongoing word of mouth? J. Mark. Res. 48 (5), 869–880. Bohner, G., Ruder, M., Erb, H.P., 2002. When expertise backfires: contrast and assimilation effects in persuasion. Br. J. Soc. Psychol. 41 (4), 495–519. Bolton, R.N., Kannan, P.K., Bramlett, M.D., 2000. Implications of loyalty program membership and service experiences for customer retention and value. J. Acad. Mark. Sci. 28 (1), 95–108. Bruell, A., 2012. Surprise! Here’s a Ticket to a Movie Premiere, on Best Buy, Retrieved February 13, 2012, from http://adage.com/article/news/latest-trendreward-consumers-delight/232718/ Buhrmester, M., Kwang, T., Gosling, S.D., 2011. Amazon’s Mechanical Turk: A new source of cheap, yet high-quality data? Perspect. Psychol. Sci. 6 (1), 3–5. Chernev, A., 2011. Semantic anchoring in sequential evaluations of vices and virtues. J. Consum. Res. 37 (5), 761–774. Davis, H.L., Hoch, S.J., Ragsdale, E.E., 1986. An anchoring and adjustment model of spousal predictions. J. Consum. Res. 13 (1), 25–37. Dorotic, M., Fok, D., Verhoef, P.C., Bijmolt, T.H.A., 2011. Do vendors benefit from promotions in a multi-vendor loyalty program? Mark. Lett. 22 (4), 1–16. Dowling, G.R., Uncles, M., 1997. Do customer loyalty programs really work? Sloan Manag. Rev. 38 (4), 71–82. Ferguson, R., Hlavinka, K., 2007. The COLLOQUY loyalty marketing census: sizing up the US loyalty marketing industry. J. Consum. Mark. 24 (5), 313–321. Finn, A., 2005. Reassessing the foundations of customer delight. J. Serv. Res. 8 (2), 103–116. Füller, J., Matzler, K., 2008. Customer delight and market segmentation: an application of the three-factor theory of customer satisfaction on life style groups. Tour. Manag. 29 (1), 116–126. Füller, J., Matzler, K., Faullant, R., 2006. Asymmetric effects in customer satisfaction. Ann. Tour. Res. 33 (4), 1159–1163. Goodman, J.K., Cryder, C.E., Cheema, A., 2012. Data collection in a flat world: the strengths and weaknesses of Mechanical Turk samples. J. Behav. Decis. Mak. 26 (3), 213–224. Han, H., Kim, Y., Kim, E.K., 2011. Cognitive, affective, conative, and action loyalty: testing the impact of inertia. Int. J. Hosp. Manag. 30 (4), 1008–1019. Han, X., Kwortnik, R., Wang, C., 2008. Service loyalty: an integrative model and examination across service contexts. J. Serv. Res. 11 (1), 22–42. Henderson, C.M., Beck, J.T., Palmatier, R.W., 2011. Review of the theoretical underpinnings of loyalty programs. J. Consum. Psychol. 21 (3), 256–276. Hu, H.S., Huang, C.T., Chen, P.T., 2010. Do reward programs truly build loyalty for lodging industry? Int. J. Hosp. Manag. 29 (1), 128–135. Jang, D., Mattila, A.S., 2005. An examination of restaurant loyalty programs: what kinds of rewards do customers prefer? Int. J. Contemp. Hosp. Manag. 17 (5), 402–408. Jargon, J., 2013. On the menu: loyalty apps. Wall Street J. (US edition), B5, July 11. Ji, M.F., Wood, W., 2007. Purchase and consumption habits: not necessarily what you intend. J. Consum. Psychol. 17 (4), 261–276. Johnson, M., Anderson, E.W., Fornell, C., 1995. Rational and adaptive performance expectations in a customer satisfaction framework. J. Consum. Res. 21 (4), 695–707. Johnson, M.D., Fornell, C., 1991. A framework for comparing customer satisfaction across individuals and product categories. J. Econ. Psychol. 12 (2), 267–286. Keh, H.T., Lee, Y.H., 2006. Do reward programs build loyalty for services? The moderating effect of satisfaction on type and timing of rewards. J. Retail. 82 (2), 127–136. Kerlinger, F.N., Lee, H.B., 2000. Foundations of Behavioral Research, 4th ed. Wadsworth, New York. Keiningham, T.L., Vavra, T.G., 2001. The Customer Delight Principle: Exceeding Customers’ Expectations for Bottom-Line Success. McGraw-Hill, New York. Kim, M.G., Mattila, A.S., 2010. The impact of mood states and surprise cues on satisfaction. Int. J. Hosp. Manag. 29 (3), 432–436.

Kivetz, R., Simonson, I., 2003. The idiosyncratic fit heuristic: effort advantage as a determinant of consumer response to loyalty programs. J. Mark. Res. 40 (4), 454–467. Lacey, R., Suh, J., Morgan, R.M., 2007. Differential effects of preferential treatment levels on relational outcomes. J. Serv. Res. 9 (3), 241–256. Leenheer, J., Van Heerde, H.J., Bijmolt, T.H.A., Smidts, A., 2007. Do loyalty programs really enhance behavioral loyalty? An empirical analysis accounting for self-selecting members. Int. J. Res. Mark. 24 (1), 31–47. Liu, Y., Yang, R., 2009. Competing loyalty programs: impact of market saturation, market share, and category expandability. J. Mark. 73 (1), 93–108. Loureiro, S.M.C., Kastenholz, E., 2011. Corporate reputation, satisfaction, delight, and loyalty towards rural lodging units in Portugal. Int. J. Hosp. Manag. 30 (3), 575–583. Mattila, A.S., 2003. The impact of cognitive inertia on post-consumption evaluation processes. J. Acad. Mark. Sci. 31 (3), 287–299. Mattila, A.S., 2006. How affective commitment boosts guest loyalty (and promotes frequent-guest programs). Cornell Hotel Restaur. Adm. Q. 47 (2), 174–181. Mattila, A.S., Hanks, L., Zhang, L., 2013. Existential guilt and preferential treatment: the case of an airline upgrade. J. Travel Res. 52 (5), 591–599. McFerran, B., Dahl, D.W., Fitzsimons, G.J., Morales, A.C., 2010. I’ll have what she’s having: effects of social influence and body type on the food choices of others. J. Consum. Res. 36 (6), 915–929. Meyer-Waarden, L., Benavent, C., 2006. The impact of loyalty programmes on repeat purchase behaviour. J. Mark. Manag. 22 (1–2), 61–88. Meyer-Waarden, L., Benavent, C., 2009. Grocery retail loyalty program effects: self-selection or purchase behavior change? J. Acad. Mark. Sci. 37 (3), 345–358. Meyer-Waarden, L., Benavent, C., 2012. Rewards that reward. Wall Street J. (June 19). Nunes, J.C., Dréze, X., 2006. The endowed progress effect: how artificial advancement increases effort. J. Consum. Res. 32 (4), 504–512. Oliver, R.L., Rust, R.T., Varki, R.T., 1997. Customer delight: foundations, findings, and managerial insight. J. Retail. 73 (3), 311–336. Palmatier, R.W., Jarvis, C.B., Bechkoff, J.R., Kardes, F.R., 2009. The role of customer gratitude in relationship marketing. J. Mark. 73 (5), 1–18. Paolacci, G., Chandler, J., Ipeirotis, P., 2010. Running experiments on Amazon Mechanical Turk. Judgm. Decis. Mak. 5 (5), 411–419. Plutchik, R., 1980. Emotion, A Psychoevolutionary Synthesis. Harper & Row, New York. Plutchik, R., 2003. Emotions and Life: Perspectives from Psychology, Biology, and Evolution. American Psychological Association, Washington. Rust, R.T., Oliver, R.L., 2000. Should we delight the customer? J. Acad. Mark. Sci. 28 (1), 86–94. Russell-Bennett, R., McColl-Kennedy, J.R., Coote, L.V., 2007. Involvement, satisfaction, and brand loyalty in a small business services setting. J. Bus. Res. 60 (12), 1253–1260. Sharp, B., Sharp, A., 1997. Loyalty programs and their impact on repeat-purchase loyalty patterns. Int. J. Res. Mark. 14 (5), 473–486. Shoemaker, S., Lewis, R.C., 1999. Customer loyalty: the future of hospitality marketing. Int. J. Hosp. Manag. 18 (4), 345–370. Shugan, S.M., 2005. Editorial: Brand loyalty programs: are they shams? Mark. Sci. 24 (2), 185–193. Stauss, B., Schmidt, M., Schoeler, A., 2005. Customer frustration in loyalty programs. Int. J. Serv. Ind. Manag. 16 (3), 229–252. St. James, Y., Taylor, S., 2004. Delight-as-magic: refining the conceptual domain of customer delight. Adv. Consum. Res. 31 (1), 753–758. Suh, J.C., Youjae, Y., 2006. When brand attitudes affect the customer satisfaction-loyalty relation: the moderating role of product involvement. J. Consum. Psychol. 16 (2), 145–155. Tanford, S., 2013. The impact of tier level on attitudinal and behavioral loyalty of hotel reward program members. Int. J. Hosp. Manag. 34, 285–294. Tversky, A., Kahneman, D., 1974. Judgment under uncertainty: heuristics and biases. Science 185 (4157), 1124–1131. Valenzuela, A., Mellers, B., Strebel, J., 2010. Pleasurable surprises: a cross-cultural study of consumer responses to unexpected incentives. J. Consum. Res. 36 (5), 792–805. Vanhamme, J., de Bont, C.J.P.M., 2008. “Surprise gift” purchases: customer insights from the small electrical appliances market. J. Retail. 84 (3), 354–369. Verhoef, P.C., 2003. Understanding the effect of customer relationship management efforts on customer retention and customer share development. J. Mark. 67 (4), 30–45. Wagner, T., Hennig-Thurau, T., Rudolph, T., 2009. Does customer demotion jeopardize loyalty? J. Mark. 73 (3), 69–85. Wansink, B., Kent, R.J., Hoch, S.J., 1998. An anchoring and adjustment model of purchase quantity decisions. J. Mark. Res. 35 (1), 71–81. Westbrook, R.A., Oliver, R.L., 1981. Developing better measures of consumer satisfaction: some preliminary results. Adv. Consum. Res. 8 (1), 94–99. Whitehurst, C., Writer, S., Colloquy, S., 2014. Shock and Awesome: Why Surprise Rewards Work, Retrieved April, 9, 2014, from http://www.retailwire.com/ news-article/17446/shock-and-awesome-why-surprise-rewards-work

L. Wu et al. / International Journal of Hospitality Management 50 (2015) 27–35 Wirtz, J., Mattila, A.S., Tan, R.L.P., 2000. The moderating role of target-arousal on the impact of affect on satisfaction – an examination in the context of service experiences. J. Retail. 76 (3), 347–365. Wood, W., Neal, D.T., 2009. The habitual consumer. J. Consum. Psychol. 19 (4), 579–592. Wu, L.L., Mattila, A.S., Han, J.R., 2014. Territoriality revisited: other customer’s perspective. Int. J. Hosp. Manag. 38, 48–56.

Yadav, M.S., 1994. How buyers evaluate product bundles: a model of anchoring and adjustment. J. Consum. Res. 21 (2), 342–353. Yoo, M., Bai, B., 2013. Customer loyalty marketing research: a comparative approach between hospitality and business journals. Int. J. Hosp. Manag. 33, 166–177. Zajonc, R.B., 1984. On the primacy of affect. Am. Psychol. 39 (2), 117–123.

35