Abstract
Preservice tips are becoming increasingly common in the marketplace (e.g., online food delivery, quick-service restaurants). While prior research has investigated how the practice of preservice tipping is perceived by customers, how preservice tipping impacts the perceptions and behaviors of employees remains unexplored. Does tipping early actually elicit better service? Through a series of four studies, our research compares the effectiveness of tips—a financial incentive, with compliments—a nonfinancial incentive. The results indicate that early tips and compliments are both effective in obtaining better service, but the relative effectiveness of a tip versus a compliment depends on the service context. In closed service contexts—which involve a continuous, relatively short interaction—tips are superior. For example, when getting a drink at a bar, buying a sandwich at a quick-service restaurant, or dropping off a car for valet parking, tipping early should lead to better customer service. In contrast, in open service contexts—which involve multiple interactions over a more extended period and provide an opportunity for a social connection—compliments become more effective. The results have practical implications for customers wishing to enhance their service experiences and for managers in motivating their employees.
Managing customers’ service experiences is of paramount importance to an organization’s success. While humans have been replaced by technology in a variety of service contexts (Van Doorn et al. 2017), services requiring more complicated behaviors (e.g., having dinner at a restaurant, getting drinks at a bar, or having a hotel room cleaned) typically still rely on employees (Wirtz et al. 2018). In these contexts, customer value is largely derived via interactions with service employees (Gao, Melero-Polo, and Sese 2020). For example, customer satisfaction with a service encounter can be influenced positively or negatively by employees’ affective state (Jha et al. 2017).
Research in customer cocreation suggests that customers can play an important role in dictating their service experience (Kelleher et al. 2019). Drawing from this prior work, we investigate the relative effect of two actions customers can take early in their service experience to influence the service they receive: tips and compliments. Although tips have traditionally been conceptualized as rewards given after a service experience (Lynn and Grassman 1990), preservice tips are becoming increasingly common in a variety of online food delivery apps and quick-service restaurants (e.g., coffee shops, delis, food trucks, smoothie shops), in part due to automated payment systems (e.g., Square), which request tips when placing an order (Warren, Hanson, and Yuan 2020). Tipping before the end of a service is also an accepted (albeit optional) practice in many traditional services such as valets, hotel housekeeping, golf courses, and bars. While soliciting tips before service can be perceived negatively by customers (Warren, Hanson, and Yuan 2020), it is unclear how employees perceive the practice and, importantly, whether tipping early in the service interaction actually improves service.
The current research examines the effectiveness of offering a tip before the service is complete relative to offering a nonfinancial incentive—namely, a compliment on the employee’s work—in heightening service employees’ customer orientation during the remainder of the service experience. While sometimes conceptualized at the organizational level, a customer orientation manifests at the individual employee level through service employee efforts (Brady and Cronin 2001; Stock 2016). A service employee’s customer orientation is comprised of their level of motivation and their ability to serve customers (Hennig-Thurau and Thurau 2003).
Through a series of four studies, we show that both compliments and tips can be effective in enhancing a service employee’s customer orientation, but that the relative effectiveness of one incentive over the other depends on the extent to which the service context is “open” versus “closed.” Open service contexts are those in which a service task is completed over time across multiple stages, such as a meal at a restaurant, a weeklong stay at a hotel, or a home renovation (Buehler, Peetz, and Griffin 2010). Closed service contexts, in contrast, involve tasks that can be completed within a single, continuous session and tend to feature only a brief encounter with the service provider (e.g., a drink at a bar, a tire replacement, a car wash). Our studies demonstrate that tips are superior incentives in closed contexts, while compliments become more effective in open service contexts. We also identify the psychological mechanisms that underlie the relationship between tips (vs. compliments) and an employee’s customer orientation.
Our findings make a number of contributions to the literature. While the service cocreation literature has primarily focused on how customers can assist employees in creating value (e.g., Kelleher et al. 2019), our research takes the opposite perspective by demonstrating how customers can motivate employees to create additional value. Our findings related to facilitating customer orientation also broaden the scope of value that customers can cocreate to include employee motivation, effort, and friendliness. By investigating the effect of offering a tip before the end of a service experience, we also advance the literature on employee tips. Whereas prior work has demonstrated how tips can be used to influence future experiences or to reward prior ones (e.g., Kwortnik, Lynn, and Ross 2009), we show how customers can proactively manage service employees’ behavior within a single service encounter.
Our work also contributes to the literature on buyer-monitoring (Kwortnik, Lynn, and Ross 2009) by introducing compliments as a new form of buyer-monitoring. We show that compliments have a different source of effectiveness than tips with regard to influencing service employees. Compliments are effective in contexts wherein it is likely that the service provider and customer will develop an ongoing relationship. However, tips are more effective when the service encounter is short and finite. These findings contribute to the literature related to the discrepancy between money and the need for others (Vohs, Meade, and Goode 2006), as service employees’ preferences for money versus relationship development appear to be context dependent. Our mediation findings additionally demonstrate that, unlike compliments given by employees (Main, Dahl, and Darke 2007), compliments given by customers early in a service relationship positively influence perceptions of the customer’s character.
Our findings provide insight into how and why frontline employees may choose to offer customers enhanced service (Brady, Voorhees, and Brusco 2012). Our moderation findings demonstrate that the degree to which a context is perceived as open or closed can be manipulated to increase the relative effectiveness of compliments versus tips. These results have implications for practitioners, as they can provide tools for increasing or decreasing employees’ susceptibility to these different incentives based on how their interactions with customers are structured.
Our research is applicable to service contexts where it is possible to tip or compliment before the end of a service experience, such as bars, food delivery apps, golf courses, quick-service restaurants, valet parking, gas stations, and cruises. Of course, while many service contexts are comprised of multiple interactions and thus provide the opportunity to tip or compliment before the end of the experience, it is sometimes not appropriate to do so. For example, tipping before the end of a service experience is not customary while eating at a restaurant. Tipping is also not customary in some geographical regions (e.g., Australia, Japan, parts of Europe) and in many professional service contexts (e.g., dental, financial). Our research is necessarily constrained to regions and contexts where tipping before the end of the experience is possible and appropriate.
The article proceeds as follows. We begin by exploring the cocreation literature and how tips and compliments can be used to cocreate value in a service experience. Next, we discuss the theoretical foundations of our central arguments, namely, that a service context’s degree of openness moderates the relative effectiveness of tips versus compliments. We then test our hypotheses across four experimental studies. We conclude with a discussion of the implications of our findings, the limitations of our designs, and future research directions.
Literature Review and Hypotheses Development
Value Cocreation
Contemporary marketing thought emphasizes the importance of involving customers in value creation (Payne, Storbacka, and Frow 2008). Our research is concerned with the ability of customers to create value within service contexts (i.e., value cocreation), defined as “participation by the customer and optionally other actors—such as other customers, service personnel, and other organizations—in practices through which value emerges” (Kelleher et al. 2019, p. 122). Value in this context is the service employee’s customer orientation, since it results in increased motivation and service behavior that is beneficial to the customer.
Many service contexts involve personalized interactions between service employees and customers, which provide the potential for customers to cocreate value. Customers can engage in value cocreation in different ways, including “influencing behavior,” defined as “customer contributions of resources such as knowledge, experience, and time to affect other actors’ perceptions, preferences, or knowledge,” and “mobilizing behavior,” defined as “customer contributions of resources such as relationships and time to mobilize other stakeholders’ actions toward the focal firm” (Jaakkola and Alexander 2014, p. 255). We focus on a form of cocreation wherein customers contribute resources that influence service employee perceptions and in doing so incentivize a customer orientation (Payne, Storbacka, and Frow 2008). This is consistent with literature on customer orientation which demonstrates that its foundation is in employee perceptions and motivation (Kelley 1992).
Considerable evidence supports the value of engaging customers in cocreation within service contexts (e.g., Auh et al. 2007; Hoyer et al. 2010). The perceived value a customer derives from a service is largely based on how well the service employee’s behavior aligns with their needs throughout the experience (Voorhees et al. 2017). Following cocreation theory, we suggest that customers can incentivize service employees financially and/or socially to align with their needs. Hence, we examine two primary incentives that customers can use to cocreate value within service contexts: tipping (financial incentive) and compliments (social incentive).
Tips and Compliments as Incentives
Financial and social incentives can be used to increase service employees’ customer orientation and thus cocreate value. We suggest that tips and compliments are two incentives that are commonly used by customers within service contexts. In cocreation terms, tips and compliments are both influencing and mobilizing behaviors in that they influence service employee perceptions of the customer and in doing so enhance their customer orientation (Jaakkola and Alexander 2014).
Tipping has been primarily examined as an outcome of service (e.g., Lynn and Gregor 2001). A small body of work has looked at the role of tipping in motivating service employees (Lynn and Withiam 2008). However, this line of research has had a relatively macro-level focus (Lynn, Zinkhan, and Harris 1993), such as how tipping policies influence motivation or general perceptions of tips as motivating (Lynn, Jabbour, and Kim 2012). Only one prior article examines the effects of tipping prior to the completion of a service encounter (i.e., Warren, Hanson, and Yuan 2020), and this work focuses on customer perceptions of the practice. The current research extends this work by examining how preservice tipping impacts employees. A brief summary of the research examining tips as an incentive appears in Table 1.
Tipping is an important customer practice that is used to improve service quality in many service industries (Lynn and McCall 2000). Kwortnik, Lynn, and Ross (2009) differentiate between no-tipping policies, wherein tips are built into the final price, and pro-tipping policies, wherein tips are voluntary. While the authors showed a positive correlation between voluntary tipping policies and service performance in some contexts (e.g., leisure cruises), they call for research to identify other incentives that can influence service quality.
In the cases noted above, the tip is presumed to be given at the end of the service interaction as a reward for good service. Certainly, tipping at the end of the service is the norm in many service environments; however, there are numerous contexts in which customers might tip at different stages of the service experience. For example, a hotel guest can tip housekeeping after each night of their stay or upon departure. Similarly, bar patrons and golfers can tip after each round of service or when the final bill arrives. In fact, it is becoming increasingly common for tips to be solicited before the end of a service interaction, most notably with food delivery services and quick-service restaurants (Warren, Hanson, and Yuan 2020). Tipping prior to or during the service experience may be an effective incentive as it may motivate the employee to provide good service during the rest of the interaction.
We also explore compliments as an incentive since prior research has demonstrated that compliments can influence employee motivation and behavior (e.g., Nasr et al. 2014; Nasr, Burton, and Gruber 2018; see Table 1 for a brief summary). Social praise and recognition of a job well done can be highly motivating for employees (Magnus 1981). Moreover, unlike tips, which tend to emphasize the transactional nature of the customer-service provider interaction (Mills and Clark 1982), compliments can foster liking toward the giver (Erickson and Eckrich 2001), which should motivate enhanced service toward that person (Vonk 2002). We extend these findings by explicating the process through which compliments incentivize service employees and by comparing the effectiveness of compliments to tips.
Research Investigating the Effects of Tips and Compliments.
Although there are numerous contexts where tipping prior to the end of the service interaction is an acceptable practice, it is our assumption that customers will be inclined to do so. To test this assumption, and to demonstrate the prevalence of such customer behavior, we conducted a pilot study that gauged whether customers have tipped before the end of a service interaction, and, if so, how likely they are to do so. We focused on tipping in particular, as there are fewer norms around the timing of compliments. We selected the contexts of hotels and bars, which are the contexts we utilize in our experimental studies.
Pilot Study
We asked customers about their tipping preferences, with a focus on two service contexts that may entail multiple interactions over a longer period of time, thus lending themselves to tipping before the end: getting drinks at a bar and staying at a hotel. Participants (N = 121 MTurk workers, M age = 35.83, SD = 11.16, 61.2% male) completed a questionnaire regarding their prior tipping behavior at bars and hotels. First, we presented participants with three different tipping patterns while getting drinks at the bar and asked them whether they had tipped in that manner before (yes/no). All tipping patterns consisted of three different rounds of drinks: the first involved providing the full tip after the first drink (e.g., $9–$0–$0); the second involved dividing the tip evenly after each drink (e.g., $3–$3–$3); and the third provided the entire tip at the end (e.g., $0–$0–$9). The results provided support for the commonality of tipping before the end of a service interaction, with nearly one third of participants (28.9%) having given the full tip after the first drink (e.g., $9–$0–$0) and almost two thirds (60.3%) having tipped evenly after each drink.
Next, we asked the same question about tipping hotel cleaning staff, with tipping patterns presented over three nights of a hotel stay. Again, the results supported the notion that customers do tip hotel staff before the last day of a stay, with 21.5% of participants reporting having provided the full tip on the first night (e.g., $9–$0–$0), and almost half (43.8%) reporting having tipped after each night of a stay (e.g., $3–$3–$3).
Having established the commonality of tipping before the end of service, we turn to our theoretical argument regarding the relative effectiveness of tips versus compliments. Given that we are seeking to influence employee motivation and behavior, it is important to recognize that the values and goals among service staff may differ, and the relative effectiveness of compliments versus tips will depend on these values and goals. One major source of difference in service employee values—and, in turn, what will motivate them—is the service context.
The Moderating Role of Service Context Openness
In order to understand how customers can effectively influence service staff, it is important to consider the nature of the service context itself, as the values of service staff can vary in part based on the context. One critical distinction is the degree to which a service context is “open” or “closed.” Prior research in psychology has highlighted the degree of openness as a way to conceptualize tasks (Buehler, Peetz, and Griffin 2010). “Closed” tasks, such as proofreading a document (Pezzo, Litman, and Pezzo 2006) or writing a simple summary (Koole and van Spijker 2000), can be completed relatively quickly in one session. Open tasks are spread over a longer time and multiple sessions and include tasks such as completing tax returns (Buehler, Griffin, and MacDonald 1997) and Christmas shopping (Buehler and Griffin 2003). Although openness or closedness is more properly characterized as a continuum rather than discrete, most tasks can be conceptualized as being more closed or more open.
Service contexts can similarly be construed based on their duration, as well as the number of stages at which the service provider interacts with the customer (Bolton 1998). A closed service context is finished in a single, continuous session (e.g., a single drink at a bar, a car wash), whereas an open service context involves more interactions and time spent with the customer (e.g., a weeklong hotel stay, a home renovation). Importantly, since open contexts involve several interactions and are relatively longer, they allow for greater relationship building between the service provider and the customer.
In addition to the duration and number of interactions, the nature of the interactions differs between open and closed contexts. In an open service context, such as a weeklong stay at a hotel, a greater amount of information can be shared between the parties involved (i.e., employee and customer), and they can develop knowledge of and concern for each other (Goodwin 1996). As a result, service contexts that are more open enhance the importance of social connectedness, which is defined as the “emotional distance or connectedness between the self and other people” (Lee and Robbins 1995, p. 239). As a result of these features, relationship development is fostered in an open service context, and the parties involved value each other and their social/emotional connection, while the relative importance of financial compensation is diminished (Blau 1964).
On the opposite end of the continuum, due to the limited interaction, closed service contexts, such as a single drink in a bar, create more of an “economic” exchange, which minimizes relationship development and focuses on the exchange of economic benefits (Lee and Robbins 1995). Closed contexts tend not to provide an opportunity to learn about and connect with one another and thus reduce the emphasis on the socio/emotional needs of the other party (Mills and Clark 1982). Rather, the parties involved develop more of a concern for financial compensation and equity, placing less value on developing a relationship (Homans 1958).
With regard to the relative influence of tips and compliments on service employees in value cocreation, we propose that the open versus closed nature of the service context is of critical importance. In closed service contexts, which are primarily focused on the exchange of goods and money, tips should be more effective than compliments because financial resources are of primary value for the service employees, and there are few interaction opportunities. The early introduction of money into a service relationship can enhance the perception that it is more of an economic exchange, which by definition decreases the relative importance of personal connection and increases the value of financial equity (Belk 2010).
In relatively open service contexts, which enable relationship development and increase the value of social benefits, compliments should be more effective in heightening an employee’s customer orientation. The importance and centrality of relationship development in interactions between service staff and customers should dictate reciprocation behavior (Aggarwal 2004). Compliments can enhance the emotional connection between two people (Erickson and Eckrich 2001), which is of primary value in open service contexts. Since the exchange of money reduces the personal nature of the context, tipping before the end of service may not be conducive to increasing customer orientation. This is supported by research demonstrating that obtaining money leads to less of a need for personal relationships (Vohs, Mead, and Goode 2006) and creates social distance between people (e.g., Zhou, Vohs, and Baumeister 2009).
To summarize, we predict that a service context’s degree of openness or closedness will moderate the effectiveness of tips versus compliments. In closed (open) contexts, tips (compliments) will be superior; however, as a context becomes more open (closed), the relative effectiveness of compliments (tips) should increase. Stated formally:
We test our hypotheses in two contexts (i.e., hotel and bar) across four studies. In Studies 1 and 2, we demonstrate that, in the relatively closed service context of getting drinks at a bar, tips are superior to compliments in fostering a customer orientation. In Study 3, we demonstrate that, in the open service context of a hotel stay, complimenting employees early in the service experience is superior to tipping. In Study 4, we hold the bar context fixed but manipulate the context’s degree of openness via the number of exchanges between the service provider and the customer. We find that, although tips are superior to compliments, making the service context more open by increasing the amount of potential interactions serves to increase the relative effectiveness of compliments.
Study 1
The goal of Study 1 was to test Hypothesis 1a, which predicts that, in a closed context, early tips will be more effective than compliments in fostering a customer orientation. An employee who has a strong customer orientation should be motivated to provide good service and should be more likely to enhance the offerings they provide to the customer. Thus, across our studies, we operationalize an employee’s customer orientation in different ways by assessing the actual service they provide, the service they would expect to provide, and the motivation that they feel. In Study 1, customer orientation is operationalized via how nice the service provider is perceived to be by the customer, and whether the customer receives free products.
It is important to note that a service “experience” can include multiple interactions between the service provider and customer yet still be considered part of the same overall service experience. For example, a seven-night stay at a hotel, one night at the bar, or dinner at a restaurant would each be considered a single service experience made up of multiple interactions. Our manipulations are implemented after initial service behaviors, which should influence the service staff’s subsequent behavior at later points in the service experience. This timing is important, as it establishes the compliments and tips as a legitimate response to some level of service, while also allowing the service provider to enhance their customer orientation.
Participants and Design
Participants for Study 1 were recruited by a marketing research firm in a major Canadian city. Participants (N = 80 customers, M age = 36.18, 45.0% male) were asked to participate in a secret shopper study wherein they would visit a local bar to order drinks and evaluate the service they received. We employed a two-cell between-participants design in which we manipulated whether the bartender was tipped or complimented. Specifically, after receiving their first drink, the secret shoppers either tipped the bartender or complimented them on the quality of the beverage.
Procedure and Measures
Participants were given the cover story that the research was for a secret shopper study aimed at evaluating a variety of bars across the city. In reality, Study 1 was a field study comparing the effectiveness of tips and compliments, with participants being randomly assigned to use one of our two customer incentives (i.e., tip versus compliment) during their interaction. Participants went to a local bar assigned to them by the marketing research firm; each participant went to a different bar in the city, for a total of 80 different bars across participants. The establishments were a mix of restaurants, lounges, and hotel bars.
We provided participants with a script and a set of instructions (see Online Appendix for details). When participants arrived at their assigned location, they took a seat at the bar and ordered a club soda with a few splashes of orange juice. Those in the tip condition were instructed to take a few sips and ask to clear the tab. They then paid for the first drink and tipped $2. When they finished their drink, they asked for a second drink with a slight modification. After the second drink, they paid and tipped another $2. Those in the compliment condition did the exact same thing, except they complimented the bartender on the quality of the first drink after they tried it and left the tab open until they were done their second drink, at which point they paid and tipped $4. This approach was selected to ensure that the bartenders received the same tip amount across all conditions. Before being sent to the bar to complete the study, participants were given a link to a questionnaire and were instructed to complete it on their phone as the service interaction progressed, as some measures were time-sensitive.
Customer Orientation
Our focal dependent variable was the bartender’s customer orientation. We theorized that, given the relatively closed context of the bar, tips would be more effective than compliments at enhancing the bartender’s customer orientation. We assessed customer orientation in a number of ways, including both behavioral and perceptual measures based on Kwortnik, Lynn, and Ross (2009). For behavioral measures, we asked participants whether the bartender offered them a free drink (yes/no), how long it took them to bring their second drink (in minutes and seconds), how often they joked with them (1 = not frequently, 7 = very frequently), how often they smiled at them (1 = not frequently, 7 = very frequently), whether the bartender recommended other items for them to have (yes/no), and whether the bartender invited them to come back (yes/no).
With regard to perceptual measures, we asked participants to evaluate how nice and friendly they perceived the bartender to be toward them, both ranging from 1 = not at all nice/friendly to 7 = very nice/friendly. We also assessed other measures for exploratory purposes in this and subsequent studies; however, these measures are not reported here. Finally, we assessed demographic variables, including gender (of the customer and the bartender), age, and previous work history in a bar setting (yes/no).
Results
Customer Orientation
Not surprisingly, initial analysis of the data revealed the importance of the gender of the customer-bartender dyad, so we coded for whether there was a match or mismatch of gender (male/male, female/female = 1, female/male = 0) and included it as a covariate in all analyses. One participant did not enter the gender of the bartender and was thus excluded from the analyses.
To determine whether tips were more effective than compliments in fostering a service employee’s customer orientation, we compared the two customer incentives on each dependent measure using ANCOVA. The results partially supported Hypothesis 1a. In particular, bartenders were directionally but not statistically significantly nicer to those who tipped early than they were to those who complimented early, M tip = 5.85, SD = 1.33, M compliment = 5.28, SD = 1.43; F(1, 76) = 3.89, p = .052. There was also a directional, but not significant, effect on perceptions of the bartenders’ friendliness, M tip = 5.64, SD = 1.46, M compliment = 5.13, SD = 1.71; F(1, 76) = 2.40, p = .126, and on their tendency to offer the customer a free drink, M tip = .89, SD = .31, M compliment = .80, SD = .41; F(1, 76) = 2.09, p = .153. The results for other service measures were also directionally consistent but did not approach statistical significance (ps > .1).
Discussion
Relative to compliments, tipping elicited greater customer orientation. Tipping before the end of service led to a directional, although not statistically significant, increase in how nice bartender seemed. Importantly, the results come from the field wherein customers influenced working bartenders. Given the field nature of Study 1 and the mystery shopper cover story, it must be noted that our data reflect the customer’s perspective not the bartender’s. In Study 2, we support these findings by examining the perspective of the bartenders. Moreover, Study 2 examines the mechanism underlying the effectiveness of tips.
The Mediating Role of Generosity in a Closed Context
A closed service context implies more of an economic exchange relationship that is predicated on receiving equity—which is most commonly financial in nature—and less about developing personal relationships (Zhang and Epley 2009). Tipping can be a useful tool for maintaining equitable relationships with servers in closed service contexts (Lynn and Grassman 1990). Customers may be motivated to tip by feelings of fairness, generosity (Azar 2005), or even guilt about the inequality between servers and customers (Shamir 1984), rather than as an attempt to develop a favorable relationship with the service provider, which may be more common in open contexts (Conlin, Lynn, and O’Donoghue 2003).
From the employee’s perspective, service interactions in closed contexts are similarly based on economic value and the balance between how much is given and received. A tip may be perceived as generous, as the service interaction is only partially complete. Given that early generosity alters the balance of what is given and received, an early tip is likely to result in a higher level of customer orientation. In other words, service is determined by the customer’s perceived generosity: The more generous a contribution is relative to what is expected, the more value will be reciprocated to make the interaction equitable.
Compliments can also be perceived as being generous and as a contribution to the service relationship (Vonk 2002). However, compliments are primarily used in contexts wherein social connection is a valued part of the experience, namely, in more open service contexts. In a closed context, wherein economic incentives hold higher value, tips should be perceived as more generous and therefore elicit greater customer orientation.
To summarize, we suggest that, given the value placed on financial compensation in a closed service context, tips will have a stronger association with value (Zhou, Vohs, and Baumeister 2009). This perceived value will make the tipper seem more generous to service staff, which will in turn elicit more a greater level of customer orientation in an effort to balance the equitable relationship. Stated formally:
Study 2
The main purposes of Study 2 were to replicate the findings of Study 1 from the perspective of the bartender and to test Hypothesis 2. Study 2 also added new behavioral measures of customer orientation, which could only be assessed by asking bartenders themselves. Specifically, we asked bartenders how much effort they would put in to serving the customer, and how likely they would be to give that customer a free drink.
Participants and Design
The study used a scenario methodology with participants with current or prior bartending experience. All participants were recruited through Qualtrics’ panel service (N = 211 current or former bartenders). We included an attention check by asking participants to rate their agreement with the statement, “I often eat cement,” scored from 1 = not at all, to 7 = very much so. We removed participants (29) who did not answer “1,” leaving 172 participants in the analysis (45.9% Male, M age = 43.08). Participants were asked to imagine themselves at their bartending job and were given a hypothetical customer interaction. They were randomly assigned to either a tip or a compliment condition.
Procedure and Measures
In each scenario, a customer sits at the bar, orders a drink, pays for it, and then asks for another drink, which marks the ends of the service experience. In the tip condition, the customer gets their first drink, which costs $7 and then gives a $3 tip; in the compliment condition, the customer takes a sip of the first drink and says, “Nicely done, it tastes great,” and then pays for it with no mention of the tip (see Online Appendix for details). After reading the scenario, participants completed the dependent measures, manipulation checks, and demographic questions.
Manipulation Check
In order to check the manipulation, we asked participants whether the customer in the scenario tipped or complimented them (yes/no). We also sought to provide additional support for the ecological validity of our manipulations by asking participants whether they had ever experienced those customer behaviors in their jobs as bartenders (yes/no).
Customer Orientation
Customer orientation was measured by combining two items based on Kwortnik, Lynn, and Ross (2009), which captured both motivation (“how much effort would you have put into serving that customer afterward”; 1 = minimum effort, 7 = maximum effort), and something more tangible (“how likely would you be to give that customer a free drink at some point”; 1 = not at all likely, 7 = very likely; r = .129, p = .092).
Generosity
In an effort to limit common method variance, we assessed our mediator with an open-ended response. We asked participants to “please describe your thoughts about the customer and their behavior in the scenario.” Three independent coders (2 female and 1 male) rated the degree to which the customer was described as being generous (α = .945) from 1 (not at all) to 4 (very much so). Discrepancies were resolved through discussion with the primary author. There were 13 participants who did not provide their perceptions of the customer or who wrote something unrelated. These participants do not have a generosity score.
Results
Manipulation Check
Both manipulations were successful. A chi-square test of independence between the conditions was significant (χ2 = 68.63, N = 172, p < .001) wherein those in the tip condition were more likely to say that they were tipped (95.5%) than were those in the compliment condition (35.7%). The same analysis of the compliment condition was significant (χ2 = 27.92, N = 172, p < .001). Those in the compliment condition were more likely to say that they were complimented (92.9%) than were those in the tip condition (58%). We also wanted to demonstrate that our manipulations actually happen. We ran the same analysis as above on whether the participants had experienced our compliment and tip manipulations during their time as a bartender. The results revealed that compliments and tips were highly and equally common (compliments 90.5%; tips 85.2%).
Customer Orientation
Given the broad range of experience in our sample of bartenders (1 year to 50 years), we wanted to control for any influence this may have had on the results. To do so, we controlled for years of experience in all subsequent analyses. We compared the two conditions using a univariate ANCOVA on customer orientation. As expected, bartenders who were tipped had directionally more of a customer orientation (M tip = 5.10) than those who were complimented, M compliment = 4.66; F(1, 168) = 3.84, p = .052. 1
Generosity
We ran the same analysis on generosity to determine whether the bartender perceived customers who tipped as more generous than those who complimented. As expected, the univariate ANCOVA revealed that customers who tipped were perceived as more generous (M tip = 3.10) than those who complimented, M compliment = 2.51; F(1, 156) = 3.84, p < .001.
Mediation
Having demonstrated that tips increase both the perceived generosity of the customer and the bartender’s customer orientation, we tested whether generosity mediated this relationship. We used Hayes’ (2017) PROCESS Model 4 with 5,000 bootstrap resamples, with the incentive (tip vs. compliment) as the independent variable, generosity as the mediator, and customer orientation as the dependent variable. The results of the analysis revealed that generosity mediated the relationship between tipping and customer orientation (β = .1890, SE = .0962, 95% CI [.0221, .3965]). In particular, tipping increased the perceived generosity of the customer (β = .5817, SE = .1283, 95% CI [.3283, .8351]), which in turn increased the bartender’s customer orientation (β = .3249, SE = .1437, 95% CI [.0411, .6088]).
Discussion
Study 2 provides additional support for the notion that tips are more effective than compliments in fostering customer orientation within a relatively closed service context. We also build upon Study 1 by providing directional support for Hypothesis 1a from the bartender’s perspective. Finally, Study 2 demonstrates that the effectiveness of tips in a closed setting is driven by perceptions of generosity, which facilitates reciprocation. While Studies 1 and 2 support the superiority of tips over compliments, Study 3 examines these effects in a more open context, wherein compliments should be more effective. In Study 3, we also investigate the mechanism driving the superiority of compliments.
The Mediating Role of Likability in an Open Context
As opposed to closed service contexts, where parties involved are concerned about economic value, in open service contexts, the development of a relationship takes on greater importance. Therefore, we expect the mediator driving the success of compliments in open contexts to differ from the mediator that drives the effectiveness of tips in closed contexts. In open contexts, we expect compliments to be more effective than tips because they make the customer seem more likable (Gordon 1996) and help to develop a relationship between the customer and service employee.
Likability plays a fundamental role in human interaction, particularly in influencing what individuals will do for others. People are more compelled to do what is necessary to develop relationships with individuals who are well-liked (Cialdini and Goldstein 2004). In our research, providing great service is one way that employees can foster a relationship with customers. This is also supported by research on flattery, which suggests that one outcome of being flattered is a motivation to foster a relationship with the flatterer (Vonk 2002).
We suggest that, in open service contexts, customers who use compliments will be perceived as more likable than those who use tips, which will in turn heighten the employee’s customer orientation. Research has found that although money is retrospectively seen as a reward, from a prospective focus, it can also be seen as bribery (Torfason, Flynn, and Kupor 2013). Thus, tipping before the end of service in an open context, where multiple stages of the experience are yet to occur, could be perceived as more of a bribe than a sign of appreciation. However, complimenting someone based on the service given up to that point would not be subject to such criticism. Thus, we expect that relative to tips, early compliments in an open service context should lead servers to perceive customers as more likable and to be more inclined to provide them with enhanced service (Cialdini and Goldstein 2004). More formally:
Study 3
Participants and Design
All participants (N = 176 undergraduates, 63.4% Male, M age = 20.22) were asked to imagine themselves as a hotel room service staff member. They were randomly assigned to one of two conditions in which they imagined that they received either a tip or compliment.
Procedure and Measures
In the tip condition, participants were asked to imagine walking into a room and seeing that the guest left a $5 tip for them on the table. In the compliment condition, they were asked to imagine bumping into the guest in the hall and the guest saying, “Nice work taking care of my room, it looks great!” (see Online Appendix for details). After reading the scenario, participants completed the dependent measures, manipulation check, and demographic questions.
Manipulation Check
We asked participants whether the customer in the scenario complimented or tipped them (yes/no).
Customer Orientation
The participant’s felt level of customer orientation was measured by asking, “how much effort would you have put into serving the consumer in the scenario?” with the following anchors: 1 = minimum, 7 = maximum.
Mediator
Likability was measured using three items (α = .944): “to what degree was the consumer welcoming/likable/pleasant” (1 = not at all, 7 = very much so).
Results
Manipulation Check
Both manipulations were successful. A chi-square test of independence between the two conditions and whether the customer tipped was significant (χ2 = 73.40, N = 175, p < .001). Those in the tip condition were more likely to say that they were tipped (78.2%) than were those in the compliment condition (13.6%). The same analysis of the compliment condition was significant (χ2 = 37.94, N = 175, p < .001). Those in the compliment condition were more likely to say that they were complimented (85.2%) than were those in the tip condition (40.2%).
Customer Orientation
An independent sample t test between the two conditions was directionally consistent as it indicated that those who were complimented would have put more effort into serving the customer (M = 5.82) than those who were tipped, M = 5.49, t(172) = −1.94, p = .055.
Mediator
An independent sample t-test between the two conditions on likability was significant. Those who complimented the staff member were perceived as being more likable (M = 5.91) than those who tipped, M = 5.40, t(173) = 2.71, p = .007.
Mediation Analysis
We tested likability as the mediator of the relationship between compliments and customer orientation using Hayes’ (2017) Model 4. The indirect path from incentive type to customer orientation as mediated by likability was significant (β = .3202, SE = .1128, 95% CI [.1067, .5523]). In particular, compliments increased the likability of the customer (β = .6516, SE = .2377, 95% CI [0.1805, 1.1227]), which in turn mediated an increase in the employee’s customer orientation (β = .2615, SE = .1513, 95% CI [.0446, .6297]).
Discussion
In Study 3, we found support for Hypothesis 1b, namely, that compliments are more effective incentives when the service context was open. Moreover, the findings supported Hypothesis 3 by showing that relative to tips, compliments increased the likability of the customer, which in turn increased the amount of effort that staff were willing to put into serving them. While the effectiveness of compliments and tips varied across closed and open contexts in Studies 1 through 3, the service context, by definition, varied. In Study 4, we manipulate a single service context to be more closed versus open to demonstrate that compliments will increase in effectiveness relative to tips as a service context comes to be perceived as more open.
Study 4
The main goal of Study 4 was to test Hypotheses 1a and 1b within a single context of a bar. We chose the bar context because it is one in which the service experience could involve single or multiple interactions with a service employee. To gain an initial sense of the relative preference for tips versus compliments in open and closed service contexts, we first conducted a survey of bartenders and servers. We reasoned that bartenders’ and servers’ service experiences would align well with closed and open contexts, respectively, and would therefore enable a test of the preference for compliments versus tips in open and closed contexts. Our theorizing suggests that those who work in the more closed context (i.e., bartenders) will have a stronger preference for tips, whereas those who work in the more open context (i.e., servers) will have a stronger preference for compliments.
One hundred university students with experience as bartenders and servers (18–36 years old, M age = 22.25, 33.3% male, 3.73 average years of experience) completed the survey. They were first asked to indicate their current role at their job using the options, bartender/server/other. 2 Participants were then given the following passage: “Please let us know which of the following you would rather have a customer do early on in your experience with them. As you think about your answer, remember to assume you are going to get a cash tip at the end of the service encounter regardless, and that the total amount of money you receive will be identical in both cases.” The participants were then given a semantic differential scale with tip scored as 1 and compliment scored as 7, and they were told that the further they chose in one direction, the stronger their preference for that option.
We ran an independent samples t test between the bartenders and servers on their preference for tips versus compliments during the service interaction. The results were significant such that servers had a stronger preference for compliments over tips (M servers = 5.62, SD = 1.89) relative to bartenders, M bartenders = 4.38, SD = 2.51, t(98) = 2.62, p = .01. This result supports our contention that compliments are relatively preferred in open contexts. Of course, there are other differences between bartenders and servers that may explain this result. Therefore, in the main study, we focused only on bartenders.
In the main study, we manipulated the openness of the service context via the degree of expected interaction between customers and servers. Specifically, we sought to make the bar context, which is traditionally a closed context, more open to determine whether this would increase the relative effectiveness of compliments. As we have demonstrated, tips before the end of service should be superior to compliments within the relatively closed context of a bar (Studies 1 and 2); however, the aim of Study 4 was to manipulate the openness of the bar context to test whether compliments would increase in effectiveness as the context became more open. We used the scenario of going to a bar during a weeklong stay at a resort to create a more open context. Given that the bar context is typically a relatively closed context, we expect that tips would still be superior overall.
Participants and Design
This was a 2 (incentive: tip vs. compliment) × 2 (degree of openness: low vs. high) between-participants design wherein participants were asked to imagine themselves as a bartender (N = 201 MTurk workers, 55.2% male, M age = 33.67).
Procedure and Measures
In each scenario, the customer orders one drink and pays for it (see Online Appendix for full details). In the tip condition, the customer gives a $3 tip on the first drink, which costs $7; in the compliment condition, the customer takes a sip of the first drink and says, “Nicely done, it tastes great,” and then pays for it with no mention of the tip. In the long-duration condition, participants are told that the customer will be staying at the resort for a week and will be coming to that bar for all of their drinks. In the short-duration condition, participants are told that the customer is passing by for that drink only. Following the description, participants completed the dependent measures, manipulation checks, and demographic questions.
Manipulation Check
The manipulation check was the same as in the previous studies.
Customer Orientation
We used the likelihood of serving the customer first as our measure of customer orientation. We also measured the amount of effort expended using the same measure as in Studies 2 and 3.
Results
Manipulation Check
A chi-square test of independence between the conditions and whether the customer tipped was significant (χ2 = 185.39, N = 201, p < .001). Those in the tip condition were more likely to say that they were tipped (99.0%) than those in the compliment condition (3.0%). The same analysis of the compliment condition was significant (χ2 = 178.00, N = 201, p < .001). Those in the compliment condition were more likely to say that they were complimented (99.0%) than those in the tip condition (5.0%).
Customer Orientation
A 2 (incentive) × 2 (degree of openness) ANOVA on the likelihood of serving that customer first revealed a significant main effect of openness, F(1, 197) = 4.85, p = .029, a significant main effect of incentive, F(1, 197) = 27.06, p < .001, and a nonsignificant interaction, F(1, 197) = 2.96, p = .087). Given our hypotheses, we conducted planned contrasts despite the lack of a statistically significant interaction. Consistent with Studies 1 and 2, tips were significantly superior (M tip = 5.55) to compliments (M compliment = 4.24, p < .001) in increasing the likelihood that the bartender would serve the tipping customer first. Tips were also significantly superior (M tip = 5.64) to compliments in the more open service context (M compliment = 4.98, p = .015). As predicted, compliments were significantly more effective in the open condition (M = 4.98) than in the closed (M = 4.24, p = .006). That is, compliments gained effectiveness as participants expected more interaction and relationship development with the customer. The same analysis on the amount of effort revealed only a significant main effect of incentive type, F(1, 197) = 7.12, p = .008: Bartenders who were tipped would have given significantly more effort in serving the customer (M tip = 5.99) than those who were complimented (M compliment = 5.50).
Discussion
Study 4 provides controlled support for the previous findings, namely, that the relative effectiveness of tips and compliments is moderated by the openness or closedness of the service context. In the typical bar condition (closed context), tips are superior to compliments in increasing the service employee’s customer orientation. Although tips remained more effective when the closed context was manipulated to be more open, this increase in openness led to compliments becoming significantly more effective than they were in the closed context.
General Discussion
Across four experiments, we investigated how customers can cocreate value in their own service experiences by incentivizing service employees. We compared the effectiveness of tips and compliments in increasing service employees’ customer orientation, both in terms of their motivation and their service behaviors. As our findings demonstrate, the service context (i.e., the degree of openness) matters because it moderates the relative effectiveness of tips and compliments. Tips are superior to compliments in closed service contexts, which are considered as being focused more on an economic exchange rather than the development of a relationship with the customer. However, compliments gain effectiveness as the context becomes more open and the amount of interaction increases. We also demonstrated the process through which both incentives enhance an employee’s customer orientation. In closed contexts, tips led the service employee to perceive the customer as more generous (Study 2); in open contexts, compliments led the service employee to perceive the customer as more likable (Study 3).
Theoretical Implications
The studies presented herein collectively contribute to the customer service literature. First, our exploration of the customer’s perspective demonstrates the role that they play in determining the outcomes of their service experiences via their influence on service employees. Our findings draw from the cocreation literature, which highlights the important role of customers in creating positive service experiences. While the literature has focused on the role of customers in assisting employees in creating value, our findings demonstrate that customers have the ability to create value by influencing employees during service interactions.
By comparing tips and compliments and showing their respective effectiveness within a service experience, our results identify new ways for customers to cocreate value. In demonstrating the importance of the degree of openness of the service context in determining the effectiveness of the two behaviors, our results also provide insights into why and when customers are able to cocreate value through tips versus compliments. When the context involves only brief interactions, tips are more effective than compliments in enhancing a service provider’s customer orientation. However, when the context provides an opportunity for interpersonal connection (i.e., is more open), compliments gain in effectiveness. These findings contribute to the literature related to the discrepancy between money and the need for others (Vohs, Meade, and Goode 2006), as service employees appear to prefer money or relationship development based on the context.
We also contribute to the service relationship literature by demonstrating why both tips and compliments can work in motivating service employees and obtaining enhanced service. Although a sales person’s use of flattery early in a service experience can activate persuasion knowledge and work against them (Campbell and Kirmani 2000), our research demonstrates that early compliments can have positive effects when used by customers. Compliments regarding quality of work make customers seem more likable to service staff, which in turn motivates them to provide better service. While tips can at times have negative connotations, for example, being construed as bribery (Torfason, Flynn, and Kupor 2013), they were perceived as an indication of generosity within the closed service context and enhanced service providers’ customer orientation. Together, these findings suggest that depending on the service context, the use of compliments or tips before the end of a service experience can reflect positively on the impression of the customer and enhance the level of service received as a result.
Practical Implications
Our research offers a number of practical implications. Most notably, our findings provide insights for customers regarding how to cocreate value by influencing service employees. If customers consider the nature of the service encounter—namely whether it is relatively open versus closed—they can take relatively simple steps to improve the service that receive. Our findings are particularly useful because often customers are uncertain about when and whether to tip, including whether a tip might be appreciated versus offensive (Lynn 2015).
This research applies to service contexts in which it is appropriate for customers to tip or compliment the service employee before the end of a service experience. While we demonstrate two such service contexts (i.e., hotel stay, drinks at the bar), there are many others where tipping or complimenting before the end of a service interaction is possible and even encouraged. Indeed, with the growth of online purchases and point-of-sale platforms such as Square and Clover, preservice tipping is now the norm for food delivery and quick-service restaurants (e.g., coffee shops, delis, food trucks, smoothie shops). Note that these are relatively closed contexts, which suggests that customers have the opportunity to enhance the service they receive through tipping. In more open service contexts like a round of golf or a cruise vacation, customers may more effectively enhance their service experience through compliments. Here, we suggest that managers reinvestigate the tipping policy and consider an after-only tipping option.
In deciding whether or not to give customers the option to tip early, we note that the practice should be considered not only in terms of the enhancements to customer service but also in terms of customers’ acceptance of the practice (Warren, Hanson, and Yuan 2020). Our investigation was limited to contexts in which early tipping was possible, but not required or requested. Explicitly requesting a tip prior to the completion of a service is controversial and can lead to customer backlash (Warren, Hanson, and Yuan 2020).
Understanding how customers can influence the customer orientation of service providers is also important for managers wishing to enhance employee motivation. The ability of customers to motivate employees is important given the potential struggles that managers can have in motivating employees and the troubles that some incentive systems can create (Carpenter and Dolifka 2017). Positive interactions with customers may not only enhance customer service but may also enhance employee job satisfaction (Isen and Reeve 2005).
Our results also suggest that managers may be able to influence the degree to which a service is perceived as open or closed. As shown in Study 4, managers can make a service context more “open” with contextual cues that imply that a relationship will develop between the customer and employee or by actually starting the development of a relationship. For example, managers could have employees ask customers for their names and then use them during the service experience to create a feeling of personal connection between the employee and customer.
Limitations and Future Research
As with any research, our work has limitations that can seed future inquiry. While we focused on compliments related to service quality (i.e., cleanliness of a room, taste of a drink), compliments can vary in nature. Future research could consider how the observed effects may differ depending on the target of the compliment. For example, customers could compliment servers on their appearance rather than their level of service, which might backfire. It would also be interesting for future research to explore whether the nature of customer generosity matters. While we demonstrate the effectiveness of a financial form of generosity (i.e., tips), generosity could also manifest in nonfinancial forms through behavior (e.g., helping), which is a more emotional form of generosity. The relative effectiveness of the various forms of generosity would likely depend on several factors, including the personality, skillset, and psychological state of the service employee, and, as our results demonstrate, the service context.
Research has demonstrated that influence practices can backfire when they are perceived as being overly generous, as such acts tend to trigger persuasion knowledge (Estelami and Demaeyer 2002). Undeserved special treatment can also backfire when there is an audience as it generates concerns that the individual will be perceived negatively by others (Jiang, Hoegg, and Dahl 2013). In line with the suggestions of Kamenica (2012), it would be interesting to explore factors that influence when a tip amount would be perceived to be too small or too large, and what the implications for the employee’s customer orientation would be.
Finally, it would be interesting to understand whether the nature of the customer-employee dyad might alter perceptions of the incentive, as demonstrated by the importance of the gender combinations in Study 1. The power dynamic between the customer and service employee may also matter, with power of the service employee over the customer (e.g., car mechanic, nightclub bouncer) increasing the likelihood of a compliment backfiring due to perceived ulterior motives. In conclusion, our findings advance our understanding of the role of customers in cocreating value in service experiences through the use of compliments and tips. Our findings provide the foundation for future research to explore when and how customers can cocreate value most effectively in their service experiences.
Supplemental Material
Supplemental Material, JSR_Executive_Summary_July_30 - Employee Reactions to Preservice Tips and Compliments
Supplemental Material, JSR_Executive_Summary_July_30 for Employee Reactions to Preservice Tips and Compliments by Raymond Lavoie, Kelley Main, JoAndrea Hoegg and Wenxia Guo in Journal of Service Research
Supplemental Material
Supplemental Material, JSR_Revised_Web_Appendix_Aug_26 - Employee Reactions to Preservice Tips and Compliments
Supplemental Material, JSR_Revised_Web_Appendix_Aug_26 for Employee Reactions to Preservice Tips and Compliments by Raymond Lavoie, Kelley Main, JoAndrea Hoegg and Wenxia Guo in Journal of Service Research
Footnotes
Acknowledgments
We would like to thank Darren Dahl and Yany Gregoire for feedback and insights on earlier versions of this article. We are grateful to funding for this research that was provided by the Social Sciences and Humanities Research Council of Canada to the first author as a doctoral fellowship and the second author in the form of an Insight Grant # 453-2014-0898.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: This study received funding from Social Sciences and Humanities Research Council of Canada (Grant # 453-2014-0898).
Supplemental Material
Supplemental material for this article is available online.
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
