It takes a village: examining how and when brand-specific transformational leadership affects employees in internal brand management

Abstract

The hospitality work environment presents many unique challenges for employees and organizations, such as the intertwined and collective work nature, implicit job expectations, and a shrinking labor market. The demand for highly skilled employees who are sufficient to deliver the unique brand values to customers is on the rise for building successful service brands through employees. The need to retain talented employees is further intensified by the “great resignation” movement in the USA along with the COVID-19 pandemic. This study demonstrates the positive role of brand-specific transformational leadership in promoting employees’ highly engaged brand building behavior (investment-of-self) and their resistance to outside competing job offers. More importantly, drawing upon cognitive dissonance theory, this study shows that the impact of brand-specific transformational leadership is mediated by employees’ sense of brand community internally, as well as moderated by perceived brand promise accuracy. These results supported the essential role of achieving employees’ cognitive consonance in brand communication both internally and externally. These results are supported by 203 US hospitality employee responses from multiple data collections purposefully designed with temporal and cognitive distance.

Introduction

The importance of a hospitality business building a strong brand is well established. However, sustaining a strong and competitive hospitality brand goes beyond creating, designing, and communicating the brand promise to customers. This is because the customers’ actual brand experiences are heavily influenced by hospitality employees’ attitudes and behavior during service encounters (Xiong and King 2019). For example, Southwest Airlines employees often embody the “Don’t take yourself too seriously” brand value by interacting with customers in a humorous manner, which contributes to the airline’s success (Sampson 2019). If employee performance is inconsistent with customers’ previously held brand expectations established through external brand communication (e.g., the brand promise), customers are less likely to develop brand satisfaction or loyalty (Murillo and King 2019). Therefore, internal brand management (IBM) has become an important topic in both academic research and industry practices for hospitality businesses (Buil et al. 2016). IBM focuses on aligning employees’ attitudes and behavior with specific brand values so that they are able to deliver the promised brand experience to customers.

Considering the intangibility and variability of service products, it is often difficult to prescribe a set of standard procedures to satisfy every customer request and brand expectation. Nonetheless, hospitality employees from different departments are expected to create consistent and coherent brand experiences that are aligned with an organization’s promoted brand to meet customer expectations (Xiong and King 2020). For example, a customer’s overall brand experience at an amusement park is developed through his/her interactions with ticketing agents, ride operators, equipment rental agents, food and beverage sellers, safety staff, etc. In addition, as service employees’ day-to-day job performance often requires interactions with coworkers, the social environment within the organization also plays a significant role in affecting employees’ decision-making. Therefore, when seeking to deliver such brand-aligned experiences, researchers have argued for the importance of leadership when ensuring employees are motivated and proficient at creating and adapting their own set of knowledge and skills, as well as curating an internal community in delivering the brand promise to customers (Barros-Arrieta and García-Cali 2021; Buil et al. 2019). However, there is a lack of understanding of the type of leadership that can effectively promote employees’ brand supporting attitudes and behavior in the internal branding context.

The role of leadership is expected to be even more critical for hospitality organizations in the USA with the COVID-19 pandemic. Described as “the great resignation,” industry experts and academic scholars pointed to the phenomenon where employees are increasingly leaving their current jobs for external opportunities (Leonhardt 2021; Rosenbaum and Ramirez 2022). This trend hurts low-paying in-person service jobs the most, such as jobs in leisure and hospitality industries (U.S. Bureau of Labor Statistics 2021). For example, in November 2021, almost a million employees in leisure and hospitality industries in the USA quit and most of them (92%) were in accommodation and food services. Employees today have more power to choose organizations that work better for their needs and wants. In addition to seeking better pay and work-life balance, reports show that employees also want better organizational cultures and practices that treat them more as human beings, a sense of community with social and emotional support, as well as development opportunities (De Smet et al. 2021). Thus, in order to avoid the “big regret” of losing employees, service organizations need to rethink their internal marketing practices.

Previous literature has highlighted transformational leadership (TFL) as an effective leadership style that binds followers together in reaching a higher collective goal as well as creates a sense of community for followers (e.g., Powell 2020; Uen et al. 2012). A transformational leader emphasizes, to his or her employees, the collective outcome of their job behavior and the value of belonging to something larger than themselves (Brown and Treviño, 2006). Eberly et al. (2017) further suggested TFL decreases followers’ turnover intentions in extreme contexts, such as natural disasters. In previous internal branding studies, TFL is also shown to contribute to employees’ brand-supporting behavior, both in-role (e.g., standard job responsibilities) and extra-role (e.g., helping coworkers) (Buil et al. 2019; Uen et al. 2012). Although it is expected that general TFL contribute to employee brand-aligned performance, there is a paucity of research that articulates the impact of brand-specific TFL on employees as well as the mechanisms of such an impact. As it is critical to highlight the relevance and meaningfulness of brand values to employees when they are expected to deliver brand-aligned experiences to customers (Xiong et al. 2013), this study adopts the brand-specific TFL term proposed in Morhart et al. (2009) to highlight the intentional emphasis of brand values in TFL. Compared to the definition of TFL that focuses on the leadership style that transforms followers in general (e.g., Pieterse et al. 2010), brand-specific TFL describes “a leader’s approach to motivating his or her followers to act on behalf of the corporate brand by appealing to their values and personal convictions” (Morhart et al. 2009, p. 123).

In short, this study focuses on how and when brand-specific TFL affects employees’ brand-supporting behaviors in internal branding. Considering the growing IBM literature (e.g., Piehler et al. 2019; Xiong and King 2018) that emphasizes employees’ social learning behaviors (i.e., learning through observing others in the organization), as well as the collective work nature of providing a holistic brand experience to customers, this study draws upon communities of practice literature (Wenger and Snyder 2000) and identifies the central roles of an internal employee-based brand community in IBM. Specifically, based on previous IBM research, this study proposes that brand-specific TLF is positively associated with employees’ investment-of-self in brand building and employee competitive resistance to outside job offers, both of which are strong indicators of employees’ brand-supporting behavior. More importantly, based on the communities of practice literature, it is proposed that this effect is mediated by employee sense of brand community. This study further draws upon cognitive dissonance theory (Festinger 1957) and identifies the moderating role of employee perceived brand promise accuracy between brand-specific TFL and employee sense of brand community. As such, these findings expand current IBM research through identifying the mechanism of brand-specific TFL in promoting employees’ brand-supporting behaviors. Practical insights with respect to enhancing the effectiveness of TFL for hospitality organizations as well as strengthening employees’ sense of belonging and shared passion through building a sense of brand community are provided accordingly.

Literature review

The effect of brand-specific TFL on employees in internal branding

TFL and brand-specific TFL have been discussed in several internal branding studies (e.g., Buil et al. 2019; Chiang et al. 2020) and are consistently identified as positive drivers of employees’ brand-aligned performance for a few reasons. First, there are four aspects of transformational leadership, including idealized influence (e.g., manifest charisma and instill pride in followers/subordinates), inspiration (e.g., communicate purpose and share the organizational vision), intellectual stimulation (e.g., help followers gain new perspectives and set high performance standards), and individualized consideration (e.g., provide individual attention and coaching) (Avolio et al. 1999; Bass 1985). As a result of these characteristics, TFL is highly recommended in organizational environments where employees are encouraged and expected to engage creatively and develop innovative solutions at work (e.g., Jaiswal and Dhar 2015). This is consistent with the internal branding management context where brand knowledge in terms of the brand identity, values, and the skills necessary to bring out the brand uniqueness during service encounters is highly tacit (King and Grace 2009; Xiong et al. 2013). As brand-specific TFL further emphasizes employees’ internalization of brand values through empowering, modeling, and coaching (Morhart et al. 2009), employees are more likely to draw upon not just the organization-provided brand information, but also their own experiences and intuition to create brand-consistent experiences for customers.

Second, as employees are expected to go above and beyond their prescribed job roles to deliver an authentic brand experience for customers in the internal branding context, researchers have argued for the importance of employees developing a strong internal motivation to deliver extra-role performance (Xiong and King 2015; Xiong and King 2018). Previous research suggests that employees need to perceive the meaningfulness and relevance of the brand as well as the importance of their role in affecting the overall brand success in order to be motivated to deliver extra-role brand-aligned performance (King and Grace 2006; Xiong and King 2015). Transformational leaders are inspirational, able to articulate compelling and relevant brand visions for employees and can promote a higher level of self-involvement from employees. This enables employees to rise above their self-interest and deliver higher-than-expected performance (Bass 1985; Brown and Treviño, 2006; Pieterse et al. 2010). Therefore, brand-specific TFL is likely to drive employees to go above and beyond the prescribed job duties and become brand ambassadors (Morhart et al. 2009).

Third, the creation of a holistic, brand-consistent experience requires seamless collaboration among service employees across different departments within the organization. Transformational leadership is an effective leadership style in fostering a teamwork-oriented environment that promotes the value of belonging to something larger than oneself or one’s department while emphasizing organization-wide outcomes to employees (Bass 1985; Brown and Treviño, 2006). Transformational leaders can further satisfy employees’ inherent needs for relatedness and role identity in building self-concept (Morhart et al. 2009). Thus, brand-specific TFL is likely to foster a positive work environment that gears toward providing a coherent brand experience to customers.

Although previous literature supports brand-specific TFL’s roles in promoting employees in-role and extra-role performance (e.g., Morhart et al. 2009), the focus was on job performance or general work engagement (e.g., Buil et al. 2019). There is a lack of consideration of effective indicators that lead to employees’ subsequent brand-aligned performance. Therefore, this study expands on current brand-specific TFL studies by examining its impact on more immediate outcomes of brand-specific TFL. Given these recommended research directions and drawing on recent research advancement in internal branding, this study focuses on brand-specific TFL’s role in affecting employees’ investment-of-self in brand building and competitive resistance to employment opportunities in other organizations. The following sections provide the theoretical foundation for each proposed link.

Employee investment-of-self in brand building

Rooted in the psychological ownership literature (e.g., Brown et al. 2014; Pierce et al. 2001, 2003), employee investment-of-self in internal brand management describes the extent to which an employee invests his/her personal time, skills, ideas, and energies into delivering the brand promise to customers. Xiong et al. (2019) demonstrated that when employees invest their personal resources into brand building, they are likely to develop psychological ownership of the brand and elicit extra-role behaviors even when such behaviors are considered risky to their job status (e.g., voicing concerns and providing constructive feedback to management). Compared to employees’ organization/brand-based identification that suggests employees develop self-concepts by “attaching” themselves to the organization/brand, employees’ psychological ownership of the brand reflects employees actively taking the brand into their domain (Xiong et al. 2019), which contributes to their strong sense of care and responsibility toward the brand naturally (Chang et al. 2012; Xiong et al. 2019). Although investment-of-self has been recognized as the most powerful means to the development of psychological ownership (Pierce et al. 2001) as well as extra-role behaviors in internal brand management (Xiong et al. 2019) for employees, it is unclear how brand-specific TFL may affect this positive employee outcome.

According to Morhart et al. (2009), brand-specific transformational leaders act as role models by living the brand. They inspire and motivate employees to internalize the vision, mission, and values of the brand (i.e., “own the brand”) and become brand champions. Transformational leaders are also proficient at encouraging employees’ creativity and innovative solutions (Jaiswal and Dhar 2015), which is a crucial skill for employees to create consistent and coherent brand experiences during service encounters. As followers of such leaders, employees are more likely to invest their personal resources (e.g., ideas, time, energy, and creativity) into delivering the brand promise. Hypothesis 1 is presented below:

H1

Brand-specific transformational leadership is positively associated with employee investment-of-self in brand building.

Employee competitive resistance

In recent years, tourism and hospitality organizations have faced many challenges in finding and retaining high-value employees due to a shrinking workforce (Pearlman and Schaffer 2013) and unfavorable work conditions such as irregular work hours and work stress (Xiong and King 2019). As a result, tourism and hospitality organizations often suffer from an increase in turnover rate, decreased service quality and customer satisfaction, as well as an increase in hiring costs (Coy 2006). These challenges are even more pressing in the internal branding context as employees are further expected to devote their judgment, experiences, and intuition in delivering brand-aligned experiences for customers (Xiong and King 2015). Considering that high employee turnover can be detrimental and costly for organizations (Park and Shaw 2013), it is critical to examine how to retain employees when they are presented with competing external job opportunities. Following the conceptualization of customer competitive resistance, which is defined as a customer’s intention to disregard competing offers from other providers (Asatryan and Oh 2008; Reynolds and Arnold 2000), this study proposes the term of employee competitive resistance to describe the employee intention to stay with the brand when facing outside competing job offers. This term is distinct from employee retention or employee turnover intention which are often used in previous literature (e.g., Tse et al. 2013) to describe the general employee intention to stay with the organization in the short and long term. Employees may choose to stay in an organization due to the lack of outside offers while employee competitive resistance further considers the “pull-to-leave” force from outside of the organization. Therefore, it may represent a higher level of employee commitment to the organization. However, the extent to which TFL may affect employee competitive resistance lacks examination.

The relationship between TFL and employee retention is supported in previous studies. Based on the social exchange theory, Tse et al. (2013) suggested that when employees perceive mutual trust, loyalty, and positive emotions from TFL, they are more likely to feel “obligated” to reciprocate by staying with the organization. In addition, TFL also fosters positive social exchanges, instills a sense of community among employees, and helps them view their jobs from a community member’s perspective (Chun et al. 2016; Herman et al. 2013; Morhart et al. 2009). Thus, TFL’s role in retaining followers can remain strong even in extreme contexts (e.g., military combat, natural disasters) as TFL increases the bond between followers and their roles as well as others in the organization (Eberly et al. 2017). This binding force is likely to be stronger under brand-specific TFL as employees are encouraged to adopt the brand values into their self-concepts (Morhart et al. 2009), which forms a common ground for developing a strong and unique employee brand bond as well as social relationships among employees. In addition, embodying the organization’s brand is considered a potential employer brand attractiveness attribute (Lievens and Highhouse 2003; Ronda et al. 2018). Thus, employees are likely to be more resistant to outside employment opportunities under brand-specific TFL.

In addition, as employees invest more of their own effort, time, energy, and ideas into building the brand, they are likely to develop a sense of psychological ownership toward the brand (i.e., this brand is mine) (Xiong et al. 2019). Based on psychological ownership studies (Brown et al. 2014; Pierce et al. 2001), with a higher level of investment-of-self in brand building, employees are likely to perceive a higher value of the brand and develop stronger intentions to protect and enhance the brand performance. Subsequently, they are more likely to value that brand more than competing brands and develop resistance against outside job offers. Hypothesis 2 and 3 are presented as follows.

H2

Brand-specific transformational leadership is positively associated with employee competitive resistance.

H3

Employee investment-of-self has a positive impact on employee competitive resistance.

The mediating role of employee sense of brand community

With the identified important roles of TFL, researchers have advocated for a deeper theoretical understanding of TFL by examining potential mediators and moderators of its impact on employee behaviors and attitudes (Avolio et al. 2004). In the context of internal branding, hospitality employees are expected to not only provide quality services, but also to internalize the brand values and perform in a brand-aligned manner to meet customers’ brand expectations (Xiong and King 2019). In this process, leaders and managers play an important role in transmitting the organizational expectations to employees. However, the effect of leadership on employee attitudes and behavior may not be straightforward. According to Moreland et al. (2001), compared to organizational socialization between management and employees, work group socialization can have a stronger influence on employees. This argument is consistent with the two-step flow process in communication theory (Katz and Paul 1955; Watts and Dodds 2007). That is, compared to direct media influence, individuals are more likely to be influenced by intermediaries between the media and people around them (Katz and Paul 1955). Considering frontline employees are likely to interact with their coworkers more often than their leaders and managers, it is likely that brand-specific TFL’s influence on employees is likely to be both direct and indirect through this community of practice.

According to communities of practice literature (Wenger 2011; Wenger and Snyder 2000), when individuals share concerns or passion for something they do, they are likely to form a community of practice and engage in collective learning in order to achieve a better outcome. As such, they are more likely to experience a sense of belonging to the community of practice. In the context of internal branding, Xiong and King (2020) suggest that when employees are expected to internalize tacit brand values into their daily performance, they are likely to build an internal brand community and rely on this community to learn and exchange their expertise and passion with coworkers in brand building (Xiong and King 2020). For example, coworkers can provide important complementary job knowledge as well as implicit insider information to fellow employees (Xiong and King 2018), which may provide a more realistic picture of the organizational expectations (Miles and Mangold 2004). That is, this internal employee community can help employees to make sense of the implicit brand values and decrease potential perceived inconsistency between formal brand knowledge dissemination from management and informal brand knowledge and skills exemplified by coworkers. As such, employees are more likely to internalize brand values into their brand delivery to customers. Therefore, it is proposed that the impact of brand-specific TFL on employees is also indirect (mediated) through an internal employee brand community.

When employees develop a strong sense of brand community internally, they are likely to develop social and emotional connections with coworkers, gain important insider information about the organization, and clarify implicit organizational expectations of their roles (Xiong and King 2020). As such, employees are also likely to perceive the brand as “our brand” and elicit a higher level of investment-of-self in brand building. Along the same logic, employees are also less likely to withdraw from the organization when they develop strong bonds with coworkers and perceive they are valuable members of an internal community (Liaw et al. 2010; Tse et al. 2013). Thus, they are also likely to develop stronger competitive resistance to outside employment offers. The mediating role of employee sense of brand community is hypothesized as follows:

H4

The relationship between brand-specific transformational leadership and employees’ investment-of-self is mediated by employee’s sense of brand community.

H5

The relationship between brand-specific transformational leadership and employees’ competitive resistance is mediated by employee’s sense of brand community.

The moderating role of perceived brand promise accuracy

Further, considering TFL influences employees through communication, role modeling, coaching, it is further important to consider factors that may affect this process, such as how employees are affected by external brand communication targeting customers (Piehler et al. 2019). One particular factor is the extent to which employees perceive the externally communicated brand promise to be accurate and consistent with their brand perception. As shown in advertising studies (e.g., Celsi and Gilly 2010), employee attitudes and behavior are also affected by advertising targeting external consumers (i.e., spillover effect). It is not uncommon for employees to perceive discrepancy between internal and external communication (Celsi and Gilly 2010). In such a case, employees are likely to experience cognitive dissonance due to conflicting cognitions of the brand. As suggested in Festinger (1957), when individuals experience cognitive dissonance, they are likely to feel discomfort. This unpleasant state motivates them to seek ways to decrease the experienced dissonance. Such dissonance reduction practices can include changing cognitions, adding/subtracting cognitions, or adjusting the importance of cognitions (Hinojosa, et al. 2017). Thus, based on cognitive dissonance theory (Festinger 1957), when employees perceive the externally communicated brand promise to be inaccurate or inconsistent with their brand perceptions, they are likely to experience cognitive dissonance which may propel subsequent dissonance reduction behaviors. For instance, employees may alter their original cognition and think the organization/management is not true to its words and therefore discount their trust toward and confidence in the leadership and organization (Miles and Mangold 2004). This can be detrimental for TFL (Braun et al. 2013). Employees are also likely to adjust the importance of the cognitions by discounting the importance of incorporating brand values into their work performance, which dampens the effect of brand-specific TFL. Along the same logic, when employees perceive a high level of brand promise accuracy, they are more likely to experience cognitive consonance and value congruence with the organization (Celsi and Gilly 2010), which should promote employee information exchange, communication effectiveness, and a friendly organizational environment (Xiong and King 2015). Therefore, employee perceived brand promise accuracy may moderate the impact of brand-specific TFL. H6 is presented as follows:

H6

The relationship between brand-specific transformational leadership and employees’ sense of brand community is positively moderated by employees’ perceived brand promise accuracy.

In summary, building upon transformational leadership research and internal branding literature, this study proposes that brand-specific TFL’s impact on employees’ investment-of-self into brand building and their competitive resistance to outside offers is mediated by employees’ sense of an internal brand community and moderated by perceived brand promise accuracy. Figure 1 presents the proposed relationships. The following sections provide a detailed description of how these relationships are assessed with empirical data collected from US-based hospitality employees in multiple periods.

Fig. 1.

Proposed relationships

Method

Measures

All measurement items were adopted from existing literature with demonstrated reliability and validity. Brand-specific transformational leadership was measured by eight items reflecting leadership characteristics (i.e., intellectual stimulation, inspirational motivation, idealized influence in terms of both attributes and behaviors, and individual consideration) in an internal branding context from Morhart et al. (2009). The Cronbach’s alpha for this measurement was 0.96. Employees’ sense of brand community was measured using the 8-item sense of brand community scale from Xiong and King (2020) developed based on McMillan and Chavis (1986). The Cronbach’s alpha for this construct was 0.95 in this study. Employee competitive resistance was measured based on the adaption of the customer competitive resistance in Reynolds and Arnold (2000) and Asatryan and Oh (2008). An example item is “I will continue working in my organization even if there are attractive job offers from its competitors.” The Cronbach’s alpha was 0.84. Employee investment-of-self was measured using the corresponding measurement items from Xiong and King (2018) and yielded a Cronbach’s alpha of 0.89. Lastly, perceived brand promise accuracy was measured with four items regarding marketing message accuracy in Seriki et al. (2016) with adaptation to the internal branding context. An example item is “Our brand marketing materials provide accurate information.” The Cronbach’s alpha for this measure was 0.86. All items were retained with no further adjustment (Table 1).

Table 1.

Measurement details

Constructs and items	Mean	s.d	Standardized loading	t value***
Brand-specific transformational leadership
My manager…
1. Seeks different perspectives when interpreting our organization’s brand values	4.96	1.721	0.753	–
2. Suggests a brand promoter’s perspective of looking at how to do my job	5.22	1.605	0.863	13.230
3. Articulates a compelling vision of our organization’s brand	5.21	1.675	0.840	12.817
4. Expresses confidence that brand-related goals will be achieved	5.56	1.463	0.841	12.841
5. Instills pride in me for being associated with our organization’s brand	5.35	1.615	0.899	13.909
6. Lives our organization’s brand in ways that build my respect	5.22	1.695	0.869	13.357
7. Spends time teaching and coaching me in brand-related issues	5.06	1.762	0.884	13.637
8. Helps me to develop my strengths with regard to becoming a good representative of our brand	5.35	1.647	0.919	14.292
Employee sense of brand community
1. I can get what I need working in this brand community	5.24	1.513	0.887	–
2. This brand community helps me fulfill my needs	5.12	1.602	0.903	19.582
3. I feel like a member of the community that works for this brand	5.36	1.545	0.912	20.091
4. I belong in this brand community	5.34	1.548	0.880	18.435
5. I have a say about what goes on in this brand community	4.48	1.945	0.694	11.987
6. People in this brand community are good at influencing each other	5.30	1.426	0.785	14.676
7. I feel connected to this brand community	5.33	1.509	0.895	19.168
8. I have a good bond with others in this brand community	5.50	1.411	0.716	12.535
Employee competitive resistance
1. I disregard job offers from other brands	4.20	1.744	0.768	–
2. I will continue working in my organization even if there are attractive job offers from its competitors	4.38	1.846	0.856	11.866
3. I would rather work for a different brand than my current organization’s brand. (Reverse-coded)	4.87	1.722	0.820	11.516
Employee investment-of-self
1. I have invested a major part of “myself” into delivering the brand promise to customers	5.40	1.503	0.823	–
2. I have invested many of my ideas into building the brand	4.78	1.736	0.720	11.178
3. I have invested a number of my talents into this delivering the brand promise to customers	5.68	1.323	0.673	10.259
4. I have invested a significant amount of my life into building the brand	4.84	1.706	0.833	13.610
5. In general, I have invested a lot in building the brand	5.04	1.562	0.867	14.340
Employee perceived brand promise accuracy
1. Our brand marketing materials provide accurate information	5.76	1.320	0.835	–
2. I feel I can deliver what our brand ads promise	5.89	1.192	0.742	11.730
3. Our brand marketing materials exaggerate how great our service is (reverse-coded)	5.27	1.658	0.701	14.107
4. Our brand marketing materials give customers a pretty good idea what they can expect from us	5.71	1.294	0.849	10.880

s.d. refers to standardized deviation; – means paths set to 1

***p < 0.001

To limit common method bias resulting from self-reported data in cross-sectional surveys, following suggestions in Lindell and Whitney (2001), an a priori marker variable (intention to purchase organic food) that is not theoretically associated with the focal constructs in the questionnaire was included. The inclusion of this marker variable enabled researchers to estimate the extent of common method bias in later analysis. All items were measured using a 7-point Likert scale, with 1 representing the lowest rating of agreement and 7 representing the highest rating of agreement. Table 1 provides the details of the measures with corresponding factor loading results revealed in later analyses.

Data collection

Participants were recruited from US-based hospitality employees with facilitation from a survey research company (Qualtrics) before the COVID-19 pandemic. In addition to reaching target populations in an efficient and effective manner, Qualtrics also provided an online survey system where several systematic measures were applied to disqualify abnormal responses (e.g., conflicting responses, and responses completed in far less than average completing time). Anonymity and confidentiality of respondents were ensured as the researchers received no identifying information from the survey platform. Considering survey insights are self-reported, in order to aid objectivity of responses and limit respondents’ perceived saliency among questions in the same survey, special attention was paid to the introduction of temporal and cognitive distance among closely related construct questions (MacKenzie and Podsakoff 2012; Podsakoff et al. 2003). The antecedent variables and outcome variables were deliberately separated in two rounds of surveys to the same individual respondent. The first survey only contained questions regarding outcome variables and demographic information. The second survey only contained questions regarding the antecedent variables. As the survey platform assigned a unique identifier number to each panel member, the responses from the same respondent that completed both surveys were matched. In the first round, the survey platform randomly selected 1500 current US-based hospitality employees from large panels and the first round of surveys were distributed via subjects’ registered emails (hidden from researchers). In this round, four hundred fifty-seven valid responses (i.e., responses with no missing values) were received (response rate = 30.5%). After a four-week period (Crossley et al. 2007), a second round of data collection was conducted by sending questions regarding antecedent variables to the 457 respondents who provided completed responses in the first round. One hundred and eighteen valid survey responses were received in this round (25.8% of the initial surveys sent out). The responses from the second round were matched with responses from the first round based on the unique identifier information of each respondent. As a result, 118 full responses were collected. Each full response contains the responses to questions in both surveys from the same individual respondent.

This four-week lag that separated data collections of antecedent variables and outcome variables can reduce the potential for common method bias, as respondents are less likely to be influenced by the saliency among questions if they are placed in different surveys (Crossley et al. 2007; MacKenzie and Podsakoff 2012). Following Crossley et al. (2007) and Karatepe (2012), a four-week lag was decided to limit respondents’ memory decay. However, a drawback to using this method is that it often suffers from lower response rate (Crossley et al. 2007). In order to warrant a robust examination of the proposed structural relationships among the focal constructs while ensuring sample quality, a third round of data collection was conducted by obtaining a new sampling pool of US-based hospitality employees (about 800). This sample excluded individuals who were involved in the first two rounds of data collection. Due to research time and budget constraints, all questions were included in this survey and distributed to this new sampling pool of potential respondents. In seeking to decrease common method bias in this situation, several remedies were adopted following Podsakoff et al. (2003). Specifically, a random sequence was applied in arranging the questions to counterbalance the order of the measurement of the antecedent and outcome variables. Different anchor points (e.g., evaluation of leaders, agreements with statements, and perceived accuracy of statements) were also adopted in different questions to reduce the perceived commonalities in scale endpoints. Consistent with previous rounds of data collection, attention filter questions, marker variable questions, and completion time threshold were also included. As a result, 85 valid responses were collected in this round. Combined with the 118 previously full responses, a final dataset of 203 responses was used for this study.

Although the sample size of 203 may be viewed as relatively small for structural analyses, Iacobucci (2010) and Hair et al. (2006) suggested that sample size consideration should also be based on model characteristics and model complexity. When the variables are reliable and effects are strong, a small sample size (e.g., 150) can be sufficient for a proper solution (Iacobucci 2010; Xiong and King 2019). In addition, considering the rigid procedures adopted in enhancing data quality (e.g., marker variable, temporal distance among multiple data collections), it is argued that the sample size of 203 was sufficient for this study.

Sample profile

Of the 203 respondents, about three-fourths (74.4%) identified themselves as hotel employees while the remainder (25.6%) were restaurant employees. Approximately three-fourths (73.9%) of the respondents were female. About four in ten respondents (41.4%) were aged 25–34, with just under one-fourth (23.6%) being 35–44 years old. Less than one-third of respondents have worked in their organization for 1–3 years (31.0%), and for 3–5 years (29.6%), respectively. Approximately three-fourths (75.9%) of the respondents held full-time positions in their organization. Two-thirds (67.5%) of the respondents held entry-level and supervisor positions. Two-thirds (66.2%) of the respondents identified their organizations to be at a 4–5 star level. Just under three-fourths (72.4%) of the organization brands were identified as chain brands as opposed to independent brands. Brand training was suggested to be fairly common among the respondents. Over one-fourth (28.8%) of respondents reported that their organizations host brand-related training monthly and approximately one-fifth of respondents (21.7%) received such training every 2–3 months. Only 4.4 percent of respondents indicated they never received such training.

Data analyses

Preliminary analyses were conducted to assess data consistency and the extent of common method bias. First, a series of t-tests and Chi-square tests were conducted to compare responses to the same questions from different rounds of data collection. No significant differences were identified, demonstrating consistency across data collection events. Second, potential nonresponse bias was assessed through a wave analysis (Rogelberg and Stanton 2007). Specifically, the first 30 responses and last 30 responses in each data collection were compared through independent sample t-tests. No significant differences were found, demonstrating limited nonresponse bias. Third, as mentioned before, a theoretically unrelated marker variable was included to account for the potential common method bias (Lindell and Whitney 2001). All relationships remained significant after including the marker variable in the measurement model check (Table 2). The correlations between the marker variable and the variables of interest in this study were also assessed. The correlation values ranged from 0.187 to 0.279. Subsequently, following the recommendations in Lindell and Whitney (2001) as well as Coelho et al. (2018), the smallest correlation between the marker variable and the variables of interest (0.187) was regarded as a proxy estimate of the common method bias and was partialled out. The adjusted correlations lend further support to the measurement validity of this study. Lastly, the researchers checked the potential multivariate normality issues by assessing the skewness and kurtosis of each item’s distribution. The skewness and kurtosis values met the normality recommendations by Byrne et al. (2013). Thus, in order to examine the underlying structural relationships among the focal constructs, a two-step structural equation model (SEM) analysis (Anderson and Gerbing 1988) using SPSS AMOS 24 using maximum likelihood estimation was conducted.

Table 2.

Correlation, composite reliability (CR), and average variance extracted (AVE)

	CR	AVE	1	2	3	4	5
1. Brand-specific TFL	0.958	0.739	0.860
2. Sense of brand community	0.949	0.702	0.637	0.838
3. Investment-of-self	0.890	0.619	0.292	0.353	0.787
4. Competitive resistance	0.856	0.665	0.352	0.529	0.305	0.815
5. Perceived accuracy	0.864	0.615	0.636	0.750	0.236	0.455	0.784

The square root of each construct’s AVE value is shown in bold and italic in diagonal space. Values below the diagonal are the correlations among constructs

In the first step measurement model check, a good model fit was achieved with χ² = 611.615 (p < 0.001, df = 339); χ²/df = 1.80, CFI = 0.95, TLI = 0.94; RMSEA = 0.06, SRMR = 0.05). As shown in Table 1, all standardized factor loading values were statistically significant (p < 0.0001) and were above 0.65. The Cronbach’s α values were 0.96, 0.95, 84, 0.89, and 0.86, respectively, in the order shown, supporting good measurement reliability. The average variance extracted (AVE) values for all constructs were above 0.6 (Table 2), suggesting the majority of variance in each construct was explained by the measurement items. Composite reliability (CR) values for all constructs were above 0.85. Taken together, convergent validity was supported (Anderson and Gerbing 1988; Fornell and Larcker 1981; Hair et al. 2006). With respect to discriminant validity, as shown in Table 2, the square roots of respective AVE value were larger than respective adjusted cross-correlations among the variables of interest, lending support for discriminant validity (Fornell and Larcker 1981). The heterotrait–monotrait (HTMT) ratios for each variable of interest were also calculated to further assess discriminant validity following Henseler et al. (2015). Since all of the HTMT ratios (ranged from 0.396 to 0.8) were below the conservative cutoff value of 0.85 (Henseler et al. 2015), they provided complementary support for discriminant validity in this study.

With the established validity of the measurement model, the hypotheses were tested through SEM via SPSS AMOS 24 using maximum likelihood estimation. Two relevant control variables (tenure and position level) were included in the following analyses as they may affect employees’ experiences with the organizational leadership (Bernerth and Aguinis 2016). As shown in Table 3 (direct effect model), the independent variable (i.e., brand-specific TFL) has a significant impact employee investment-of-self (β = 0.43, p < 0.001) and competitive resistance (β = 0.36, p < 0.001) directly. Employee investment-of-self also has a positive impact on competitive resistance (β = 0.27, p < 0.001). Thus, H1, H2 and H3 were supported. In addition, the control variable employee tenure showed a significant impact on employee investment-of-self (β = 0.23, p < 0.001). It is noted that this direct effect model test (e.g., tests for H1 and H2) are not necessary for the mediation tests for H4 and H5 statistically (Zhao et al. 2010). However, they are necessary for testing H1 and H2, both of which are proposed based on previous literature.

Table 3.

Path estimates in direct effect model and indirect effect model for mediation tests

Path	Direct effect model	Indirect effect model
TFL → Investment-of-self	0.43(5.568)***	n.s
TFL → Competitive resistance	0.36(4.29)***	n.s
Investment-of-self → Competitive resistance	0.27(3.13)***	0.178(2.317)***
TFL → Brand community	–	0.705(9.640)***
Brand community → Investment-of-self	–	0.347(3.461)***
Brand community → Competitive resistance	–	0.50(4.909)***
R2
Investment-of-self	0.28	0.241
Competitive resistance	0.29	0.408
Brand community	–	0.496
Fit indices
χ2	236.233	572.422
Df	127	287
χ2/df	1.86	1.995
CFI	0.96	0.936
TLI	0.95	0.927
RMSEA	0.065	0.070

TFL refers to brand-specific transformational leadership; investment is short for investment-of-self; brand community is short for employees’ sense of brand community; each value in the box shows the standardized path estimate

***p < 0.001; *p < 0.05; t value is placed in parentheses; – this path is not included in the model; n.s. means not significant

H4 and H5 propose that the relationships between brand-specific TFL and the two outcome variables are mediated through employee sense of brand community. Following MacKinnon et al. (2004), employee sense of brand community was included in the structural analysis and the mediated (indirect) effects were identified through a bootstrapping procedure (2000 samples) with a 95% confidence interval and bias-corrected percentile method. The results showed that brand-specific TFL has a significant standardized indirect effect on employee investment-of-self (95% CI [0.18, 0.48]), p < 0.05), as well as on employee competitive resistance (95% CI [0.33, 0.56]), p < 0.05). Under this analysis, the strength of the positive relationships between brand-specific TFL and employee investment-of-self as well as competitive resistance dropped to an insignificant level (Indirect effect model in Table 3), suggesting an indirect-only mediation effect described in Zhao et al. (2010). Thus, H4 and H5 were supported.

With respect to the hypothesized moderating role of perceived brand promise accuracy (H6) between brand-specific TFL and employee sense of brand community, the procedures recommended by Aiken et al. (1991) in testing interaction effects were followed. This procedure is widely adopted organizational behavior research (e.g., Liu and Cho 2018). Specifically, brand-specific TFL and perceived brand promise accuracy were mean-centered. An interaction term was created by multiplying the two mean-centered terms. The interaction term (brand-specific TFL × perceived brand promise accuracy), brand-specific TFL and perceived brand promise accuracy were then included in the structural model as three antecedent variables linking to employee sense of brand community. Other paths stayed the same. The bootstrapping technique (2000 samples) with a 95% confidence interval and bias-corrected percentile method was applied to obtain more details of the moderation effect. This interaction model also demonstrated good model fit (χ² = 737.598 (p < 0.001, df = 414); χ²/df = 1.782, CFI = 0.936, TLI = 0.928; RMSEA = 0.06). The interaction term contributed to sense of brand community (β = 0.105, p < 0.05). Almost seventy percent (69.8%) of variance of employee sense of brand community was explained. The significant relationships between sense of brand community to the two outcome variables remained strong and positive (β = 0.45 for invest-of-self, and β = 0.54 for competitive resistance). The moderation effects was further evaluated based on the low and high levels of perceived brand promise accuracy (Table 4). It is shown that at a high level of perceived brand promise accuracy (i.e., the value is one standard deviation above the mean), brand-specific TFL has a strong and positive impact on employee sense of brand community (β = 0.429, p < 0.05). That is, when employees perceive the externally communicated brand promise to be highly consistent with their brand perceptions, they are more likely to experience a sense of community as a result of brand-specific TFL. However, when perceived brand promise accuracy is at a low level (i.e., the value is one standard deviation below the mean), the impact of brand-specific TFL on employee sense of brand community becomes insignificant (β = 0.186, p = 0.157). Therefore, H6 is supported.

Table 4.

The effects of brand-specific TFL on employee brand community at high and low levels of perceived brand promise accuracy

	β	Stand error	p	BootLLCI	BootULCI
− 1SD	0.186	0.136	0.157	− 0.069	0.463
Mean	0.308	0.108	0.005	0.095	0.529
+ 1SD	0.429	0.134	0.003	0.17	0.694

− 1SD refers to one standard deviation below the mean value of perceived brand promise accuracy (i.e., the low level). Mean refers to the mean value of perceived brand promise accuracy. + 1SD refers to one standard deviation above the mean value of perceived brand promise accuracy (i.e., the high level)

BootLLCI bootstrap lower-level confidence interval, BootULCI bootstrap upper-level confidence interval

Discussion and theoretical implications

This study draws from cognitive dissonance theory to consider how employees process brand knowledge dissemination from formal management/leadership, informal community of coworkers, and the brand promise communicated to customers. Lower perceived cognitive dissonance helps employees to make sense of the implicit organizational expectations and engage in brand building. Specifically, this study examined how and when brand-specific TFL affects employee investment-of-self in brand building and competitive resistance to outside offers. This study found that brand-specific TFL plays a positive and significant role in promoting both. The findings also demonstrated brand-specific TFL effects through employee sense of an internal brand community and is moderated by perceived brand promise accuracy based on cognitive dissonance theory.

This study is consistent with previous studies that demonstrated the positive roles of TFL in community building (Chun et al. 2016) as well as positive outcomes of a strong sense of brand community among employees (Xiong and King 2020). This study contributes to internal branding literature by examining how and when brand-specific TFL works in hospitality organizations where seamless collaboration among service employees are required to deliver the brand promise to customers. The strong mediation effect of employee sense of brand community is in line with Morhart et al. (2009), Chun et al. (2016) and Tse et al. (2013) that suggested TFL’s impact can be mediated by social exchanges among employees, which contributed to employees’ role identification in brand building and relatedness to brand community. This study further provides a cognitive dissonance perspective to explain how brand-specific TFL affects employees’ investment-of-self and competitive resistance through employee sense of brand community. In addition, this study supports the spillover effects of external marketing communication (e.g., advertising) (Celsi and Gilly 2010) on employees, which is rarely discussed in internal brand management. Along with the cognitive dissonance arguments in supporting the role of an internal brand community, this study further shows that brand-specific TFL’s effect on employees is stronger when employees perceive the brand promise to be accurate, reducing potential cognitive dissonance between internal and external brand communication. Further, it also indicates that the relationship between brand-specific TFL and employee sense of brand community holds even when employees perceive the brand promise to be inconsistent with their brand perceptions, further supporting the positive roles of TFL in deterring external pull-to-leave forces for employees (Mitchell et al. 2001).

This study also contributes to both TFL and internal branding literature by expanding brand-specific TFL’s roles in encouraging employee retention, especially when they are enticed with outside offers. Xiong and King (2020) examined employee sense of community in an internal branding context and suggested that it contributes to employees’ brand belief and brand-aligned performance. This study adds to the understanding of employee sense of brand community by showing the significant antecedent role of brand-specific TFL. In addition, previous studies commonly adopt employee retention or organizational commitment under TFL (e.g., Tse et al. 2013) with limited consideration of contextual factors such as competitive job offers. With employees increasingly seeking meaning and purpose within their employment post-pandemic, this finding is highly timely for cultivating and retaining talented employees in service branding through leadership. Also, it is well-established that employees are expected to go beyond their prescribed roles to deliver brand-aligned experiences in internal branding (Xiong and King 2015). This expectation may require a higher level of employee devotion and perceived ownership of brand success (Xiong and King 2018). This study built upon this notion and incorporated employee investment-of-self from psychological ownership studies (e.g., Brown et al. 2014; Pierce et al. 2001) as a highly relevant outcome of internal branding, which is propelled by brand-specific TFL and employee sense of community.

Practical implications

With the demonstrated effects of brand-specific TFL and its boundary conditions in internal branding, the study presents important managerial implications. First, consistent with Buil et al. (2019) and Morhart et al. (2009), it is believed that brand-specific TFL is a desired leadership style for hospitality organizations that seek to build and sustain a successful brand through employees’ brand-aligned performance. As noted in Morhart et al. (2009), managers are likely to achieve better outcomes in internal branding when they promote a unified brand vision, live the brand values, coach employees to develop their interpretation of brand-aligned performance, and provide individualized support to employees. Thus, it is recommended that brand-specific TFL should be infused into recruiting, training, evaluating, and promoting processes of managers in hospitality organizations. In addition, the ability to foster a supportive internal brand community, enhance employees’ willingness to stay with the organization in a competitive labor market, and promote employees’ self-devotion to brand building are suggested as effective measures to evaluate brand-specific transformational leaders’ effectiveness. The adoption of these evaluation criteria can further incentivize managers to exhibit brand-specific TFL.

Second, given the strong mediation effect of employee sense of brand community, it is suggested that hospitality organizations pay special attention to the informal internal community in the organization as it may alter the effects of brand-specific TFL. Employee experiences in an internal community affect how they perceive their leaders in organizations. Therefore, cultivating and sustaining positive and close relationships with internal community leaders is critical for TFL effectiveness. In addition, it is critical for management to foster a supportive social environment where employees can experience a sense of belonging and obtain necessary skills and knowledge in order to excel at their roles. Brand-centered events, celebrations, and gatherings that strengthen employees’ social connections and exchanges are encouraged. As a result, employees are more likely to receive social and emotional support that is highly relevant to mitigate the demanding hospitality work environment. As a community, they are also more likely to develop a shared vision, a strong bond with other community members, as well as embody the brand values advocated by their leaders. This bonding force also aids the retention of employees.

Further, the brand promise as shown in external brand communication is a controllable factor by organizations. Thus, it is critical to consider employees as an audience of external brand communications and carefully monitor their perceived accuracy of the brand promise. Although it is intuitive to not over-promise in advertisement to avoid customer disappointment, it is also important to depict a consistent look and feel of the brand image communicated to employees internally. In addition, as many external brand communication/promotion materials conveying the brand promise often feature employees, such communication should show reasonable expectations for employees. As such, employees are less likely to experience cognitive dissonance and more likely to deliver brand-aligned performance that is consistent with customer brand expectations (which is based on the brand promise).

Limitations and future research

There are a few limitations in this study that may be addressed in future research. First, considering the self-reported nature of this study, future studies are encouraged to collect data from multiple sources including manager reviews, employee insights, and/or customer evaluations on different constructs. In addition, as in many studies that examine social environments (e.g., a brand community) with ongoing exchanges among members, it would be beneficial to obtain longitudinal data to capture the continuous reciprocation among employees to gain more insights on the full dynamics the effects of brand-specific TFL. It is also interesting to examine whether the impact of brand-specific TFL differs based on an organization’s departments. As noted in Morhart et al. (2009), employees in sales department may be less influenced by TFL because of the goal-oriented mentality in generating sales. A second limitation of this study is generalizability in other cultural contexts. The data in this study were collected from a broad selection of US-based hospitality employees, however, the cultural characteristics of US organizations can be very different from those in other countries. For instance, Tse et al. (2013) suggested that the leader-member exchange mechanism might be more salient in an eastern cultural context with a higher power distance. Thus, variables that reflect culture characteristics should be included in future research as additional moderators that may affect transformational leadership’s effectiveness. Further, although employee investment-of-self in brand building is expected to propel their extra-role brand-aligned performance, this relationship is not assessed in this study. Future studies are encouraged to establish a nomological network of other typical internal branding outcomes (e.g., employee work engagement and brand identification) under brand-specific transformational leadership. Although this study applied multiple methods to control for common method bias such as the technique of including a theoretically unrelated marker variable (Lindell and Whitney 2001), it is important to note that partialling out the proxy marker variable method bias is inadequate to control for other sources of common method biases, such as social desirability and implicit theories (Podsakoff et al. 2003). The aforementioned suggestions such as longitudinal studies as well as cross-culture studies might be valuable remedies. Nevertheless, future studies are recommended to continue this effort to control for common method bias when data are self-reported. Last but not the least, the influence of external brand communication outside of organizations’ control on employees is largely unexplored in literature and needs future research attention. Although the general idea of achieving employees’ cognitive consonance should still hold based on this study, the extent to which such external brand communication from third-party sources (e.g., online travel agencies and news report) may further affect employees’ brand attitudes (e.g., brand belief and trust) is unclear and holds both theoretical and practical significance in IBM.

Lina Xiong

is an Assistant Professor in Department of Human Dimensions of Natural Resources, Colorado State University. She has a Ph.D. in Business Administration with concentration in tourism and hospitality management. Her research focuses on service brand management and internal branding in service organizations. She publishes in top academic journals in hospitality and marketing. She emphasizes a sense of purpose and wellbeing in her research and life.

Declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.