The effect of customer incivility on proactive customer service performance mediated by emotional exhaustion and moderated by proactive personality

Abstract

The existing body of research offers limited insight into the detrimental effects of incivility phenomena on proactive customer service performance (PCSP) from the customer’s perspective. Drawing on the stressor–strain–outcome (SSO) framework and conservation of resources (COR) theory, this paper examines how customer incivility is associated with PCSP, with a particular focus on emotional exhaustion as a mediator and proactive personality as a moderator. Utilizing survey methodology, customer-contact employees in China were sampled with a time lag of three weeks in three waves. SPSS 27.0, the PROCESS macro (Model 59) and AMOS 27.0 were utilized to analyze the data. The results showed that customer incivility had a positive impact on emotional exhaustion and reduced PCSP. Additionally, the indirect effect of customer incivility on PCSP through emotional exhaustion was significant, indicating a significant mediating effect. The results also revealed that proactive personality significantly moderated the association between customer incivility and emotional exhaustion and the moderated mediation index was also significant. These findings indicate that proactive personality serves as a positive psychological resource, enhancing employees’ capacity to mitigate the adverse effects of customer incivility (stressor) and emotional exhaustion (strain) while enabling them to consistently engage in proactive service behaviors (outcome), thereby breaking the vicious path of the SSO framework.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-57984-3.

Introduction

Customer incivility, characterized by discourteous or disrespectful conduct from customers toward service personnel¹, represents a pervasive phenomenon among customer-contact employees^2,3. Such behavior is nearly unavoidable for those in customer-contact roles, with a majority encountering it during their tenure⁴. The ramifications are considerable, encompassing heightened levels of employee burnout, diminished job satisfaction, and elevated turnover intentions⁵. Given that customer-contact employees frequently interact with customers who may be demanding, dissatisfied, or frustrated, it’s understandable that they’re susceptible to incivility^6,7. Employees who encounter customer incivility tend to report higher emotional exhaustion, with lower service performance often observed in such contexts. Hence, it is important for service-oriented companies to pay attention to the issue of customer incivility and its effects on their employees. In addition, investigating factors that may buffer these detrimental outcomes is therefore of significant theoretical and practical importance.

While previous research has investigated how customer incivility, as a stressor, affects employee attitudes and behaviors (e.g.,^8–11), significant limitations persist within the extant literature. First, the impact of customer incivility on proactive customer service performance (PCSP) remains underexplored. To the best of the authors’ knowledge, few empirical studies have explicitly examined this relationship. Instead, research on PCSP has primarily concentrated on factors inside organizations that either support or impede PCSP. Positive factors inside organizations, such as organizational justice, leader humor and leader humility have been identified^12–14, along with negative factors like exploitative leadership, abusive supervision and leader aggressive humor^15–17. Nevertheless, negative factors outside organizations, particularly customer incivility, have not received commensurate scholarly attention, representing a significant lacuna in the current understanding of PCSP determinants.

To bridge the existing research gaps in PCSP, this study investigates the relationship between customer incivility and PCSP. Furthermore, it introduces emotional exhaustion as a mediator to delve into the underlying process between antecedent and outcome, aiming to comprehensively figure out how customer incivility impacts customer-contact employees’ PCSP. Drawing from conservation of resources (COR) theory¹⁸, customer incivility as a stressor, it continuously depletes customer-contact employees’ valuable resources. Consequently, the limited resources required for work leave employees susceptible to stress, anxiety, and depression, with a state of emotional exhaustion often accompanying these experiences¹⁹. Hobfoll²⁰ emphasizes that emotional exhaustion can stem from the threat of losing valuable resources, further hindering service employees from dedicating themselves to implementing in-role and extra-role performance. Therefore, it appears that emotional exhaustion acts as a mediator in the association between customer incivility and PCSP. These relationships also align with the stressor–strain–outcome (SSO) framework²¹, where customer incivility is viewed as a job stressor, emotional exhaustion as a job strain, and PCSP as a job outcome. Although previous studies have tried to provide detailed insights into how customer incivility related to employees’ PCSP, specifically, negative affectivity²² and work engagement² were used as mediators, however, emotional exhaustion has never been utilized as a mediator to reveal the critical underlying mechanism between customer incivility and PCSP.

Existing literature has consistently confirmed proactive personality as a crucial protective factor, equipping individuals to effectively manage stress^23,24. Prior studies have also demonstrated that proactive personality moderated the psychological states and behaviors of employees following negative events^25,26. Accordingly, proactive personality could serve as a significant moderator affecting the relationship between customer incivility and emotional exhaustion (or related performance). As discussed before, customer incivility may be positively associated with emotional exhaustion. However, the strength of this association varies among individuals due to differences in personality traits. To be specific, those with high proactive personality exhibit the ability to adapt to stressful situations, guiding their thoughts and behaviors effectively²⁷. Customer-contact employees with high proactive personality showcase greater proficiency in emotion regulation, allowing them to deal with organizational stress and mitigate any negative effects on how they feel emotionally²⁸ Consequently, having proactive personality might help to moderate how customer incivility impacts its outcomes in this study. Furthermore, in response to the SSO model’s call for potential moderating variables²⁹, proactive personality was integrated as a moderating variable. To the best of our knowledge, this is the first time the SSO model has incorporated a moderating variable, especially within the hospitality domain. With this context in mind, this study aimed to address above-mentioned research gaps by examining how customer incivility affects PCSP, the role of emotional exhaustion as a mediator and the role of proactive personality as a moderator.

This study addresses the following research questions: Firstly, does customer incivility have direct effect on PCSP? Secondly, does customer incivility have indirect effect on PCSP? Thirdly, do customer incivility and proactive personality have interactive effect on PCSP? The current research identifies three main contributions to existing literature. First, it enriches the understanding of PCSP by offering robust evidence that customer incivility can undermine such specific discretionary actions within the hospitality sector. Second, it tackles the gap in understanding the mechanism behind the customer incivility–PCSP link by recognizing emotional exhaustion as a potential mediator. Lastly, it demonstrates that employees with a high proactive personality experience fewer negative effects from customer incivility, thereby preventing a significant reduction in their PCSP, while also responding for the first time to the SSO model’s call for investigating potential moderating variables.

Literature review and hypotheses

Customer incivility and PCSP

Researchers highlighted that the service quality significantly related to customer satisfaction and loyalty^30,31. However, customer service in the hospitality industry is inherently uncertain and personalized³². Many effective ways of providing service can’t be fully planned out in advance through official job descriptions and specific operating procedures³³. Additionally, spontaneous and unsolicited service behaviors play a crucial role in enhancing customers’ experiences³³. Therefore, the initiative employees take in serving customers is pivotal for the successful delivery of services³⁴. Consequently, it is crucial for both researchers and managers to pay attention to hospitality customer-contact employees’ PCSP³⁵.

Customer incivility is defined as low-intensity deviant acts by individuals in customer roles, involving ambiguous harmful intent toward employees and a violation norms of mutual respect and civility³⁶. This construct broadly covers a range of disrespectful and discourteous verbal and nonverbal customer behaviors. By exposing employees to rude, demeaning, or disrespectful conduct, customer incivility fosters a hostile work environment. Consequently, extensive research has documented its adverse consequences for employees. For example, such mistreatment has been found to impair employee performance²², increase role stress³⁷, and reduce overall service quality within organizations³⁸. Additionally, it is known to contribute to lower job satisfaction³⁹ and increased intention to quit³⁹.

Based on COR theory, customer incivility may diminish hospitality employees’ PCSP. The theory postulates that when individuals confront a real or perceived loss of resources, they work to safeguard the resources they have²⁰. Therefore, it’s reasonable to suggest that customer-contact employees experiencing customer incivility opt for lower levels of PCSP as a means to preserve existing resources, in contrast to those who do not suffer such incivility. If customer-contact employees consistently lose resources without being able to replenish them or compensate in other ways, they might feel that they lack the necessary resources to fulfill their job requirements. Thus, the hypothesis proposed is as follows.

H1

Customer incivility is negatively related to PCSP.

The mediating role of emotional exhaustion

Emotional exhaustion is regarded as a work-related strain that reduces individual emotional resources as a result of excessive psychological demands⁴⁰. Such work-related strain has been found to impede organizational performance⁴⁰. The Koeske and Koeske’s²¹ SSO paradigm outlines the work-stress process, defining job stressors as challenging events in the working environment that employees perceive as difficult. There are various stressors in the hospitality sector, such as illegitimate tasks⁴¹, abusive supervision¹⁷, workplace ostracism³², sexual harassment in the workplace⁴² and leader aggressive humor¹⁵. However, there have been few studies that specifically examine how customer incivility, as a stressor in the workplace, impacts job strain within this industry.

According to Koeske and Koeske²¹, job strain refers to employees’ psychological reactions to job stressors, which often coincide with the depletion of their physiological and affective resources. The current research concentrates on emotional exhaustion, the initial stage of burnout syndrome²⁹ and a core dimension of burnout⁴³. Emotional exhaustion is a commonly addressed job strain in previous studies across various contexts, often accompanying negative consequences such as frustration, feeling depleted, and a general lack of emotional investment and care²⁹. Consistent with SSO model²¹, this study proposes a positive stressor–strain relationship, suggesting that the more customer incivility customer-contact employees experience, the higher their emotional exhaustion level will be. Thus, the following hypothesis is proposed:

H2

Customer incivility is positively related to emotional exhaustion.

In the SSO framework, the outcomes are defined as lasting behavioral or psychological effects that occur due to prolonged stress and strain. Our research framework illustrates that customer-contact employees’ emotional exhaustion negatively impacts their PCSP, aligning with the negative strain–outcome association in the SSO model. To be specific, when such employees experience emotional exhaustion following exposure to customer incivility, they may struggle to attain effective PCSP due to resource limitations. Additionally, customer-contact employees might react negatively by reducing PCSP to avoid further resource depletion. Those dealing with emotional exhaustion need to allocate their restricted time and energy towards calming themselves and stopping any further emotional decline, which tends to co-occur with reduced focus and engagement at work, and such patterns are often associated with PCSP. Moreover, based on COR model of stress and burnout, resource-deprived employees employ defensive tactics to prevent further loss of resources²⁵. These defensive strategies, often manifesting as decreased engagement and commitment to organizations³⁵, may include relinquishing PCSP.

H3

Emotional exhaustion is negatively related to PCSP.

Our research hypothesized an SSO framework with customer incivility as job stressor, emotional exhaustion as job strain, and PCSP as job outcome. Drawing on existing literature, when employing SSO framework, it is posited that customer-contact employees’ job stressor, represented by customer incivility, heightens their job strain (emotional exhaustion), subsequently reducing their performance outcome such as PCSP. Hence, the following hypothesis is constructed:

H4

Emotional exhaustion acts as a mediator between customer incivility and PCSP.

The moderating role of proactive personality

In response to the SSO model’s call for potential moderating variables²⁹, proactive personality was incorporated as a moderating variable. Proactive personality refers to a stable, dispositional trait that reflects an individual’s tendency to take initiative, identify opportunities, and act to bring about environmental change. People with a proactive personality are not constrained by situational forces; instead, they actively shape their surroundings⁴⁴. This trait is relatively enduring and is typically measured through personality scales.

Proactive personality characterizes individuals’ inclination to actively seek opportunities, display initiative, take decisive actions, and persist until achieving closure through instigating change⁴⁵. When faced with customer incivility, employees lacking proactivity not only incline to passively tolerate the current circumstances but also encounter challenges in embracing new chances to take charge⁴⁶. Consequently, they undergo greater psychological distress, signifying a heightened level of emotional exhaustion. Following this rationale, the relationship between customer incivility and emotional exhaustion is expected to be less pronounced for proactive individuals compared to non-proactive employees. Thus, proactive personality as a type of personal characteristic resource, is anticipated to alleviate the association between customer incivility and emotional exhaustion. Therefore, the following hypothesis is proposed:

H5

Proactive personality moderates the relationship between customer incivility and emotional exhaustion.

PCSP, as a behavioral construct, refers to the specific, discretionary actions that employees take to deliver exceptional service beyond formal job requirements⁴⁷. It is context-bound and observable⁴⁸. Proactive personality is a trait, which is stable over time and across situations. PCSP is a state or behavior, which can fluctuate daily based on workload, mood, empowerment, or leadership support⁴⁷.

Within the COR framework, how customer-contact employees respond to resource depletion due to strain, like emotional exhaustion, depends on individual differences such as personality traits, abilities, or skills^49,50. Specifically, proactive personality serves as a type of personal characteristic resource that helps replenish supportive resources. This, in turn, enables the effective use of positive coping strategies to counteract workplace strain⁵¹. For those with a proactive personality, the association between emotional exhaustion and PCSP is expected to be weaker compared to those without such a personality trait. This is because individuals with a proactive personality can utilize their coping resources to express their proactive nature actively²⁵. Consequently, they are less likely to seek retribution or reduce discretionary behaviors like PCSP to minimize resource losses. Conversely, for employees low in proactive personality, the association between emotional exhaustion and PCSP may be stronger in non-proactive employees compared to proactive ones. This is due to insufficient available resources for employees with low proactive personality to cope with workplace strain²³. According to COR theory, non-proactive individuals are more sensitive to resource losses, prompting them to react by conserving remaining resources through reducing PCSP⁵². Hence, the following hypotheses are posited:

H6

Proactive personality moderates the association between emotional exhaustion and PCSP.

H7

The indirect effect of customer incivility on PCSP via emotional exhaustion is conditional on proactive personality.

Proactive employees actively seek constructive opportunities to engage with their environment. In contrast, non-proactive employees tend to passively await opportunities and are less likely to recognize chances to act upon their surroundings²⁵. Prior research have shown that proactive employees are highly motivated from within to navigate challenging situations using positive coping methods^26,53. Consistent with this perspective, proactive employees are resistant to prolonged entanglement in customer incivility; instead, they proactively attempt to alter unfavorable situations⁵⁴. Moreover, they consistently pursue effective approaches to initiate significant improvements, avoiding further resource loss rather than retracting their PCSP as a type of retaliation against customer incivility. Thus, proactive personality may serve as a moderator in the direct association between customer incivility and PCSP. Therefore, the following hypotheses are put forward:

H8

Proactive personality moderates the direct association between customer incivility and PCSP.

Ultimately, a moderated mediation model was constructed, as depicted in Fig. 1.

Fig. 1.

The conceptual framework.

Methodology

Participants and data collection

A group of 337 customer-contact employees employed by eight five-star hotels in Beijing and Guangzhou in China was chosen using a convenience sampling method. The researcher employed a three-wave approach with a three-week time lag. Initially, permission was obtained from human resources department managers of the selected hotels. To ensure the validity of the responses, the participants were explained that their responses were strictly confidential and there was no correct or incorrect answer. Additionally, participants were asked to indicate their current positions in the filtering questions, to ensure that the final sample included only customer-contact employees. The survey comprised an introduction outlining the project’s objectives and methodology, along with a questionnaire covering respondents’ demographic details, customer incivility, emotional exhaustion and proactive personality. The questionnaires were distributed to 440 customer-contact employees and 389 completed surveys were collected in the first wave. In the second wave, 359 valid questionnaires returned. In the last wave, immediate supervisors of each respondent who had successfully completed the second wave rated their PCSP and 57 supervisors evaluated between six to ten employees each. A total of 337 valid responses were gathered from 57 supervisors, with an overall response rate of 76.6%. Each survey instrument was assigned a distinct code linked to the participant’s name, enabling accurate tracking of individuals across three waves of data gathering.

Instruments

The initial version of the survey was developed in English. To minimize translation inaccuracies and uphold measurement quality, a bilingual language expert performed translation and back-translation procedures following established cross-cultural adaptation guidelines⁵⁵. Before full administration, the questionnaire underwent a pilot test using a test–retest approach to evaluate its stability, while two domain experts reviewed its content validity to confirm the relevance of items to the research setting. The construct of customer incivility was evaluated using a 10-item instrument adapted from Wilson and Holmvall’s⁵⁶ scale, which demonstrated excellent internal consistency (Cronbach’s α = 0.900). An example statement is “How often have customers blamed you for a problem that you did not cause?” Proactive personality was measured using a six-item scale from Bateman and Crant⁴⁵, also showing strong reliability (Cronbach’s α = 0.891). A sample item is “I am always looking for better ways to do things.” Emotional exhaustion was gauged with eight items from Maslach and Jackson⁵⁷, also with strong reliability (Cronbach’s α = 0.921). A sample item was “I feel used up at the end of the workday.” PCSP was evaluated with seven items from Rank et al.⁴⁸, exhibiting high reliability (Cronbach’s α = 0.936). A sample item was “My staff member takes initiative to communicate client requirements to other service areas and collaborates in implementing solutions.” Excluding customer incivility, all constructs were assessed using five-point Likert scales, with responses ranging from strongly disagree (1) to strongly agree (5). Customer incivility’s response options anchored from 1 (never) to 5 (very often). Prior studies have found that several demographic characteristics of customer-contact employees (e.g., age, gender, education, and tenure) may impact their service performance, these variables were incorporated as controls.

Data analysis

For this study, descriptive analyses and bivariate correlations were computed with IBM SPSS Statistics, version 27.0. To evaluate the construct validity of the measurement framework, confirmatory factor analysis (CFA) was conducted utilizing AMOS, version 27.0, consistent with established validation principles⁵⁸. Hypothesis testing was carried out through the PROCESS macro (Model 59).

Results

Respondent profile

The traits of 337 customer-contact employees are demonstrated in Table 1. In term of gender, 62.6% were female. As for age, 38.9% were 18–25, 47.8% were 26–35 years, and 13.3% were between the ages of 36 and 45. With regard to educational background, 15.1% held middle school certificates, 29.1% had high school certificates, 39.5% possessed associate degrees and 16.3% held bachelor’s degrees or higher. Regarding years of working, 22.6% had been working at their hotels for less than a year, 45.1% for 1to 3 years, 27.0% for 4 to 6 years, and 5.3% for over 6 years. Moreover, the demographic profile of supervisors has also been presented in Table 1.

Table 1.

Demographic profiles of customer-contact employees (n = 337) and supervisors (n = 57).

Demographic variables	Customer-contact employees	Supervisors
Gender
Male	126 (37.4%)	26 (45.6%)
Female	211 (62.6%)	31 (54.4%)
Age
18–25	131 (38.9%)	7 (12.3%)
26–35	161 (47.8%)	34 (59.6%)
36 and above	45 (13.3%)	16 (28.1%)
Education background
Middle school	51 (15.1%)	0
High school	98 (29.1%)	11 (19.3%)
Associate degree	133 (39.5%)	32 (56.1%)
Bachelor and above	55 (16.3%)	14 (24.6%)
Years of working
1 year and below	76 (22.6%)	0
1–3 years	152 (45.1%)	8 (14.0%)
4–6 years	91 (27.0%)	21 (36.9%)
6 years above	18 (5.3%)	28 (49.1%)

Common method bias test

In this study, employees’ survey responses were guaranteed to be anonymous and the researchers emphasized to respondents that there was no correct or incorrect answer. To assess the potential presence of common method bias (CMB), the researchers applied three analytical approaches. First, a single-factor test based on Harman’s method was performed, revealing that the first factor accounted for only 16.31% of the cumulative variance, which was well below the 50% threshold and indicated no dominant factor in the unrotated structure⁵⁹. Second, the partial least squares (PLS) model was extended by incorporating a latent method factor, specifically a marker variable (social desirability) with four items⁶⁰. The modified PLS model showed that the average method variance attributed to the marker variable was just 0.5%, suggesting that CMB did not pose a significant issue. Third, a full collinearity assessment was conducted on the PLS model; all variance inflation factors fell below the recommended cutoff of 3.3⁶¹, further indicating the absence of serious CMB. Based on these combined results, it can be concluded that CMB was unlikely to meaningfully affect their findings.

Measurement model

The full measurement model presented an acceptable fit to the data (x²/df = 2.86, CFI = 0.912, TLI = 0.908, RMSEA = 0.067). Before testing the research hypotheses, CFA was conducted. The reported CFA results are based on corrected data after fixing several data-entry errors. The results in Table 2 indicated a good model that fits the data. Additionally, both Cronbach’s alpha and composite reliability (CR) values for all variables exceeded the recommended thresholds. Convergent validity was confirmed by observing average variance extracted (AVE) values greater than 0.5 for each construct, following criteria by Bagozzi and Yi⁶². Furthermore, strong evidence was established for discriminant validity by comparing the square roots of AVE with bivariate correlation coefficients, consistently finding the former to be higher, as shown in Table 3.⁶³.

Table 2.

Results of reliability and convergent validity of full measurement model.

Variables and items	Standardized loadings	Variances	CR	AVE	Cronbach α
Customer incivility			0.920	0.538	0.919
CI1	0.610	0.628
CI2	0.696	0.516
CI3	0.658	0.567
CI4	0.736	0.458
CI5	0.803	0.355
CI6	0.683	0.534
CI7	0.745	0.445
CI8	0.905	0.181
CI9	0.798	0.363
CI10	0.652	0.575
Emotional exhaustion			0.921	0.596	0.921
EE1	0.771	0.406
EE2	0.858	0.264
EE3	0.691	0.523
EE4	0.751	0.436
EE5	0.779	0.393
EE6	0.697	0.514
EE7	0.754	0.431
EE8	0.856	0.267
Proactive customer service performance			0.931	0.660	0.931
PC1	0.735	0.460
PC2	0.882	0.222
PC3	0.829	0.313
PC4	0.788	0.379
PC5	0.812	0.341
PC6	0.813	0.339
PC7	0.822	0.324
Proactive personality			0.897	0.592	0.897
PP1	0.744	0.446
PP2	0.802	0.357
PP3	0.812	0.341
PP4	0.766	0.428
PP5	0.715	0.489
PP6	0.783	0.387

CR = Composite reliability; AVE = Average variance extracted.

Table 3.

Mean, standard deviation, inter-correlations of variables and discriminant validity.

Variables	Mean	SD	1	2	3	4	5	6
1. CI	3.66	0.852	0.73
2. EE	3.53	0.928	0.55***	0.77
3. PCSP	3.59	1.064	− 0.57 ***	− 0.59***	0.81
4. PP	3.57	0.943	− 0.22***	− 0.29***	0.45***	0.77

N = 337; SD = standard deviation; PCSP = proactive customer service performance; CI = customer incivility;

PP = proactive personality; EE = emotional exhaustion; *** P < 0.001; the square root of AVE for each variable is.

shown with bold Italics.

Tests of mediation and moderated mediation

As shown in Table 4, customer incivility had negative impact on PCSP (β = − 0.32, p < 0.001), supporting H1. Customer incivility was positively linked to emotional exhaustion (β = 0.42, p < 0.001), supporting H2. Furthermore, emotional exhaustion negatively affected PCSP (β = − 0.37, p < 0.001), confirming H3. Moreover, the indirect effect of customer incivility on PCSP through emotional exhaustion was significant, with a 95% confidence interval of (− 0.21, − 0.09), indicating support for H4. In addition, proactive personality exerted a significant moderating effect on the association between customer incivility and emotional exhaustion (β = − 0.18, p < 0.01). To illustrate this effect visually, slopes were calculated at 1 standard deviation above and below the mean⁶⁴. Figure 2 visually illustrated the interaction pattern, which substantiated that proactive personality moderated the association between customer incivility and emotional exhaustion, supporting H5. Furthermore, proactive personality significantly moderated the link between emotional exhaustion and PCSP (β = 0.13, p < 0.01). To visualize this effect, slopes were calculated at 1 standard deviation above and below the mean⁶⁴, shown in Fig. 3. Therefore, H6 was confirmed. Moreover, the findings indicated that the conditional indirect effect was stronger in the low proactive personality condition (β = − 0.24, CI − 0.37, − 0.13) and weaker in the high condition (β = − 0.08, CI − 0.13, − 0.03), with a significant index of moderated mediation (index = 0.07, CI 0.03, 0.10), thereby supporting H7. Additionally, proactive personality significantly moderated the direct association between customer incivility and PCSP (β = 0.14, p < 0.001), as shown in Fig. 4, thus supporting H8.

Table 4.

Results of mediation and moderated mediation (PROCESS: model 59).

Variables	Emotional exhaustion (M)					PCSP(Y)
	B	SE	P	LLCI	ULCI	B	SE	P	LLCI	ULCI
Gender	0.17	0.09	> 0.05	− 0.01	0.35	− 0.01	0.09	> 0.05	− 0.18	0.17
Age	− 0.06	0.07	> 0.05	− 0.21	0.09	0.18	0.07	> 0.05	− 0.05	0.14
Education	− 0.06	0.05	> 0.05	− 0.16	0.04	0.04	0.05	> 0.05	− 0.03	0.22
Tenure	− 0.13	0.06	> 0.05	− 0.25	0.01	− 0.12	0.06	> 0.05	− 0.24	0.01
CI (X)	0.42	0.05	< 0.001	0.35	0.56	− 0.32	0.06	< 0.001	− 0.47	− 0.23
PP (W)	− 0.22	0.05	< 0.001	− 0.33	− 0.14	0.37	0.05	< 0.001	0.25	0.43
X*W	− 0.18	0.05	< 0.001	− 0.27	− 0.09	0.14	0.04	< 0.01	0.05	0.21
EE(M)						− 0.37	0.05	< 0.001	− 0.61	− 0.39
	R2 = 0.26					R2 = 0.48
Bootstrapping results for the indirect effect
EE	B	SE	LLCI	UICI
	− 0.15	0.03	− 0.21	− 0.09
Bootstrapping results for conditional indirect effect
PP	B	SE	LLCI	UICI
PP(low)	− 0.25	0.06	− 0.38	− 0.13
PP(high)	− 0.08	0.03	− 0.13	− 0.03
Index	0.08	0.02	0.03	0.10

X = independent variable. W = Moderator. M = Mediator. Y = Dependent variable. B = Unstandardized coefficient.

SE = Standard error. P = Probability. LLCI = Lower-level confidence interval. ULCI = Upper-level confidence interval.

Fig. 2.

The interactive effects of customer incivility and proactive personality on emotional exhaustion.

Fig. 3.

The interactive effects of emotional exhaustion and proactive personality on PCSP.

Fig. 4.

The interactive effects of customer incivility and proactive personality on PCSP.

Discussion

Drawing on the SSO framework and COR theory, this study examines how customer incivility relates to PCSP through the mediating role of emotional exhaustion and under the moderating effect of proactive personality. Unlike prior PCSP research that has concentrated largely on internal organizational factors, whether positive or negative, the present investigation focuses on an external negative stressor: customer incivility. Furthermore, it introduces emotional exhaustion as an explanatory mechanism to clarify the psychological process linking incivility to PCSP, a mediating relationship not yet established in the literature. Additionally, proactive personality is incorporated as a moderating variable, marking one of the first attempts to introduce a boundary condition into the SSO sequence and thus responding directly to the framework’s call for identifying potential moderators.

In order to further interpret these research findings, the Chinese cultural context deserves explicit attention. Chinese hotel employees are socialized to value interpersonal harmony (hexie) and protect face (mianzi). Customer incivility directly threatens these norms. Because service expectations in China are highly customer oriented (e.g., ‘the customer is always right’), employees often suppress their immediate negative reactions to maintain harmony. However, this suppression paradoxically accelerates emotional exhaustion, consistent with the emotional labor literature. Furthermore, proactive personality may be particularly beneficial in this cultural setting: proactive employees can reinterpret uncivil behaviors as less personal or subtly seek supervisory support without openly violating harmony or face concerns. Thus, the Chinese cultural context may help explain why the detrimental pathway from customer incivility to emotional exhaustion is especially meaningful, and why proactive personality appears to be a valuable personal resource in this study setting.

Theoretical implications

This study makes significant contributions to existing literature by investigating the impact of customer incivility on customer-contact employees’ PCSP. While Cheng et al.²² demonstrated that customer incivility’s negative effect on PCSP was mediated by negative affectivity, they did not directly explore the relationship between the two. Jang et al.’s² study is the only one to examine this direct relationship in the hospitality industry in South Korea. Our research fills this gap by examining this relationship in the context of the Chinese hospitality industry.

Second, recognizing the importance of customer-contact employees’ PCSP for hotels’ survival in the competitive hospitality sector⁶⁵, numerous researchers have focused on its antecedents. Accordingly, more and more positive or negative factors inside organizations have been examined as PCSP’s antecedents (e.g.,^12,15,66). However, few researchers have delved into the impact of factors outside organizations especially customer–related on employees’ PCSP. Therefore, this study fills this important gap by introducing customer incivility to the understanding of how customer-related stressor weakens employees’ PCSP.

Third, this study aimed to explore a crucial underlying mechanism by introducing emotional exhaustion as a mediator. This approach provided detailed insights into how customer incivility affects the PCSP of customer-contact employees. On one hand, according to COR theory, when hotel employees face customer incivility, this stressor gradually depletes their valuable resources¹⁸. Consequently, they must allocate more resources to deal with this stressor, leaving them with limited resources for work and causing stress, anxiety, and sometimes emotional exhaustion^34,67. This, in turn, further depletes the existing resources of hotel employees⁶⁸, making it challenging for them to engage in PCSP. Thus, the results aligned with COR theory. Additionally, from the perspective of SSO framework, the current study elucidates the overall work-stress process (job stressor → customer incivility; strain → emotional exhaustion; outcome → PCSP). The positive association between customer incivility and emotional exhaustion is related to customer-contact employees’ poor PCSP in hotels. Such a single SSO model of customer incivility was used to predict emotional exhaustion and PCSP, which is an approach largely overlooked in previous studies.

Fourth, this study explored the moderating function of proactive personality in the relationships among customer incivility, emotional exhaustion, and PCSP. The inclusion of this construct responds to the recognized need to identify potential moderators within the SSO model²⁹. Employees with a less proactive disposition tend not only to endure customer incivility more passively but also demonstrate a reduced capacity to mobilize resources and pursue new opportunities⁴⁶. Consequently, they are likely to experience more intense psychological strain, manifesting as elevated emotional exhaustion. Accordingly, it was hypothesized that the positive association between customer incivility and emotional exhaustion would be weaker for employees with a proactive personality than for their less proactive counterparts. The empirical results supported this proposition, confirming the theorized buffering role of proactive personality.

Practical implications

First, managers and supervisors can support customer-contact employees in enhancing their capacity to respond to customer incivility. While some employees may naturally possess stronger conflict resolution skills, organizations can further develop this competency through structured initiatives. For instance, training programs that integrate role-playing exercises, perspective-taking activities, and curated instructional videos can effectively strengthen practical dispute management abilities. Additional preparatory measures may include creating dedicated forums for open discussion, where staff can exchange experiences, emotional responses, and coping strategies, or establishing a formal mentoring system. In such a program, employees facing customer incivility can be paired with experienced advisors who provide guidance and share evidence-based insights. Collectively, these approaches help equip customer-contact employees to better navigate and mitigate instances of incivility.

Second, the current results reveal that customer incivility (stress) is positively related to customer-contact employees’ emotional exhaustion (strain), and this, in turn, corresponds with a decline in their PCSP. Organizations are encouraged to establish an on-site psychological counseling center where employees can access support for managing stress and strain as needed. Past studies indicate that the efficacy of recovery and coping techniques plays a crucial role in assisting employees in overcoming workplace stress and strain, thereby positively impacting their performance^65,66. Such a support center can serve as a sustained resource, offering employees guidance from trained professionals on proactive techniques to regulate stress and emotional strain, thus preventing adverse effects on job performance.

Finally, the findings demonstrated that proactive personality functioned as a protective factor, mitigating the adverse impact of customer incivility on emotional exhaustion. This implies that hotel management would benefit from integrating considerations of proactive personality into both recruitment and talent development strategies. Specifically, where legally permissible, hiring managers could incorporate validated assessments to evaluate candidates’ levels of proactive personality during the selection process, potentially prioritizing those with stronger proactive traits⁶⁹. Furthermore, management can cultivate a more proactive orientation among existing employees through targeted training interventions. Substantial empirical evidence supports that such developmental initiatives not only alleviate workplace stress but also enhance emotional regulation competencies and improve PCSP^68,70.

Limitations and future research

Despite its valuable insights, this study has several limitations. Firstly, the study utilized a time-lagged design to collect data. While such designs can provide evidence of temporal precedence, they do not establish definitive causal relationships due to the inability to rule out alternative explanations such as unmeasured confounding or reverse causality. Therefore, causal claims should be interpreted with caution. Future research may benefit from adopting longitudinal or experimental designs that span a more extended period to more rigorously test causal hypotheses⁷¹. Secondly, reliance on self-reported measures renders this study susceptible to CMB, potentially magnifying the relationships between variables. While potential bias was not significant in this study, it is recommended to collect data from multiple sources in future studies to minimize this concern⁶⁰. Thirdly, data were collected exclusively from hotel employees currently occupying customer-facing roles. This sampling approach necessarily excluded individuals who may have experienced high levels of customer incivility and subsequently left their positions or been transferred to non-customer-service departments, as accessing their perceptions posed considerable methodological challenges. Fourthly, the study elucidates the adverse effect of customer incivility on frontline employees’ PCSP, identifying emotional exhaustion as a key explanatory mechanism. Building on this, future research could investigate other sequential mediators within the stressor-strain-outcome framework, such as burnout, psychological distress, or job dissatisfaction, to further unpack the association between incivility and PCSP. Fifthly, while emotional exhaustion was the theoretically specified mediator, future research could examine a more comprehensive model that simultaneously tests competing mechanisms (e.g., anger, moral disengagement, surface acting) to determine their relative explanatory power. It must be acknowledged that these variables were not statistically controlled for, which represents a limitation of the current study. Finally, as the study was conducted within the Chinese cultural context, the generalizability of the findings to Western settings may be limited⁷². Cultural norms emphasizing interpersonal harmony in China likely shape how hotel employees perceive and respond to customer incivility, suggesting the need for cross-cultural replications to determine the boundary conditions of the proposed model.

Conclusions

This moderated mediation model represents a significant advancement in the literature, extending prior investigations into customer incivility and PCSP by formally incorporating emotional exhaustion as a mediator and proactive personality as a moderator. It constitutes a novel effort to empirically verify both the direct effect of customer incivility on PCSP and its indirect effect through emotional exhaustion, while further demonstrating the moderating role of proactive personality within this pathway. The results establish that customer incivility reduces PCSP, a relationship that is partially mediated by emotional exhaustion and contingent upon levels of proactive personality. Importantly, the findings underscore that a proactive personality not only buffers the adverse links between customer incivility, emotional exhaustion, and PCSP, but also weakens the indirect mediating effect of emotional exhaustion in the association between incivility and performance.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Qi Wang contributed to conceptualization, writing, review and editing of the manuscript. Shiliang Sun was responsible for preparation of the original draft and supervision. Zhen Yan contributed to writing, reviewing, and editing the manuscript. Yue Niu conducted methodology development, and contributed to conceptualization and proofreading.

Data availability

The data reported in this study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

This manuscript is not under review elsewhere, and the results have not been published previously or accepted for publication. All methods were carried out in accordance with relevant guidelines and regulations. The study strictly adhered to the principles of the Helsinki Declaration and obtained approval from the Ethics Committee of Qingdao Vocational and Technical College of Hotel Management.

Consent to participate

Participants were informed about the purpose of the study, the voluntary nature of their participation, and their right to withdraw at any time. They provided informed consent before completing the questionnaire.