While increasing the organizational identification of employees has been described as the ultimate goal of internal marketing and internal branding, one of its most common practices is to communicate organizational values both internally and externally. However, very little is known about the relative effectiveness of different types of organizational brand communication. Drawing from construal level theory, the current research investigates whether the degree of construal fit, defined as the extent to which the construal of communication matches an employee’s construal of headquarters, determines the relative impact of organizational brand communication. A series of studies consisting of two cross-industrial multilevel field studies and a lab experiment provide evidence that organizational brand communication with low-level (high-level) construal is more effective to target employees with low (high) social distance to headquarters. These findings suggest that construal fit qualifies the effectiveness of organizational communication.
Increasing the organizational identification and commitment of employees has been described as the ultimate goal of internal marketing and employer branding (e.g., Fernandez-Lores et al., 2016; Knox and Freeman, 2006; Wieseke et al., 2009). However, only 30% of employees identify with the organization they work for (David, 2013). This finding is alarming given that employees’ organizational identification, defined as the “perceived oneness with an organization and the experience of the organization’s successes or failures as one’s own” (Mael and Ashforth, 1992, p. 103), not only predicts employees’ attitudes and behaviors in terms of their in-role performance, organizational citizenship behavior, customer orientation, and performance (e.g., Bell and Menguc, 2002; Lee et al., 2015; Wieseke et al., 2007, 2009), but even motivates customers to spend more money in retail stores (Lichtenstein et al., 2010).
The internal marketing and internal branding literatures emphasize the importance of communicating organizational values and brand values to increase employees’ organizational identification (e.g., Álvarez-González et al., 2017; Berry and Parasuraman, 1992; Buil et al., 2016; Cardador and Pratt, 2006). Indeed, besides leadership (e.g., Wallace et al., 2013; Wieseke et al., 2009), organizational communication is acknowledged as “the most common internal marketing application” (Ahmed and Rafiq, 2003, p. 1183), and as an important driver of organizational identification (Cheney and Christensen, 1999). To date, research focused either on the effects of internal communication (e.g., Malhotra and Ackfeldt, 2016; Smidts et al., 2001; Wiesenfeld et al., 1999) or external communication (i.e., advertising; Hughes, 2013; Gilly and Wolfinbarger, 1998; Wolfinbarger and Gilly, 1991) on employees. However, both communication types occur at the same time, and may simultaneously influence employees. In consequence, there is the need to “better understand the relative impact of internal and external communication on employee attitudes and behaviors” (Baker et al., 2014, p. 654). This is an important gap because many marketers have to justify the allocation of their communication budgets, and need decision criteria on how to balance internal and external communication (Forrester Research, 2014).
The present study addresses this gap by investigating how employees’ organizational identification is simultaneously affected by both internal and external brand communication. Drawing from construal level theory (Trope and Liberman, 2010), it is argued that social distance to headquarters determines an employee’s mental representation of headquarters and thus also the construal level of headquarters. Moreover, the principles of construal level theory suggest that internal brand communication will be more likely associated with low-level construal (i.e., a concrete representation of an object) while external brand communication will be more likely associated with high-level construal (i.e., an abstract representation of an object). Furthermore, it is postulated that the degree of construal fit, defined as the extent to which an employee’s construal of headquarters1 matches the construal of organizational brand communication, determines the relative impact of internal versus external communication.
The example of Caterpillar illustrates the different levels of construal used in organizational brand communication. Caterpillar targets employees with a description of how concrete behaviors represent the organizational brand value “sustainability”. This internal communication further includes detailed behavior guidelines, for example that employees should “focus on improving the quality and efficiency of our operations while reducing our environmental impact” (Caterpillar, 2015, p. 33). On the contrary, Caterpillar communicates the same organizational brand value with high-level construal in external communication, and targets the public with a description of why sustainability is important with the claim “today’s work, tomorrow’s world” (Caterpillar, 2015).
A series of preliminary studies empirically examines the association between organizational brand communication and construal level. Subsequently, two cross-industrial multilevel field studies with 1102 employees and a lab experiment with 219 participants test for the effect of construal fit between organizational brand communication and social distance to headquarters on employees’ organizational identification. All studies consistently support the importance of construal fit, and the combination multiple data sets and complementary methods allows to establish causal relationships and valid results.
Therefore, the current research advances the branding literature in several ways. The present studies are the first to offer a theoretical integration and empirical investigation of the combined effects of internal versus external brand communication on employees’ organizational identification. While doing so, the findings add to the emerging literature on the use of brand communication to ensure that employees are identified and aligned with the organization (e.g., Baker et al., 2014; Hughes, 2013). Moreover, this study uncovers an association between the internal versus external focus of organizational brand communication and its level of construal. Uncovering this association is meaningful to understand the relative impact of internal and external brand communication because their construal level qualifies the effectiveness of communication on different employees. The present studies reveal that organizational brand communication is only likely to exert the desired effects on organizational identification when the construal level of the communication matches the social distance between an employee and headquarters. These findings provide practical recommendations on how to frame brand messages to effectively increase employees’ organizational identification.
Theoretical background and hypothesesOrganizational brand communication and organizational identificationThe marketing literature highlights that both internal and external brand-related communication positively influence organizational identification and related constructs (e.g., Baker et al., 2014; Malhotra and Ackfeldt, 2016; Wolfinbarger and Gilly, 1991). Following Hughes (2013) and Smidts et al. (2001), internal (external) brand communication is defined as the formal, non-personalized internal (external) brand communication executed by headquarters. Internal communication informs organizational members about organizational brand values and activities (e.g., Ahmed and Rafiq, 2003; Berry and Parasuraman, 1992). Such detailed information should reduce uncertainties about the organization, help to comprehend, define, and reassure the organizational purpose, and pint out the salient characteristics that distinguish the own organization from other organizations (e.g., Dutton et al., 1994). Internal communication may further enhance an employee’s intrinsic motivation to belong to a certain organization (Smidts et al., 2001), increase favorable perceptions of an organization (Ashforth and Mael, 1989; Malhotra and Ackfeldt, 2016), and retain talented staff (Knox and Freeman, 2006).
In contrast, external communication primarily targets an external audience such as customers. However, several marketing researchers have argued that employees are an important second audience for external communication (e.g., Celsi and Gilly, 2010; Hughes, 2013; Wentzel et al., 2010; Wolfinbarger and Gilly, 1991). Christensen (1997) even claimed that external communication may be more important for internal than external audiences. In line with these notions, the communication literature suggests that employees are receptive to external communication and react in terms of interpreting and evaluating the messages they receive via advertising (Drumwright, 1996). In particular, external communication may increase an employee’s knowledge about the identity of the organization (Gilly and Wolfinbarger, 1998), and further help to differentiate an organization in competitive labor markets in order to successfully attract talented staff (an important goal of employer branding; Knox and Freeman, 2006).
Given the high plausibility and the previous support in extant empirical research, the effects of internal and external brand-related communication on organizational identification are used as a baseline when investigating the moderating effect of construal fit. In the following, existing research on the association between organizational communication and construal level theory is summarized to argue how construal fit will determine the relative impact of internal and external brand communication on employees’ organizational identification.
A construal level perspective on internal and external communicationConstrual level theory differentiates between high-level and low-level construal (Trope and Liberman, 2010). A high-level construal implies that an individual mentally represents a certain object abstractly. By doing so, the individual focuses on the bigger picture and central features that capture the core of the object. Such high-level, abstract, context-independent representations emphasize superordinate goals and primary features (e.g., “why” goals). In contrast, a low-level construal implies that an individual mentally represents a certain object concretely, and focuses on the peripheral, secondary features that are less essential to the overall gist of the object. Such low-level, concrete, context-dependent representations emphasize subordinate goals and secondary features (e.g., “how” goals). A large body of research indicates that the level of construal has wide-ranging effects on multiple outcomes (e.g., Liberman and Trope, 2008; Trope and Liberman, 2010; Tsai and McGill, 2011; Zhu et al., 2017).
Previous construal level research differentiates between high-level and low-level construal within organizational communication in different ways, including vision communication versus goal setting (Berson et al., 2015) and abstract versus concrete calls for action (Berson and Halevy, 2014). The principles of construal level theory further suggest that the construal level of internal and external communication will likely be different, as illustrated with the example of Caterpillar Inc. External communication will be more abstract (as opposed to internal communication) because (1) it takes a more distal perspective, (2) it is directed towards a more heterogeneous audience, (3) it addresses a more distal target group, and (4) it portrays an organization in a more positive light.
First, individuals switch between abstract and concrete representations as they move from distal to proximal perspectives. Berson et al. (2015) used the example of an executive at Google who might construe a vision as organizing the world’s information when talking to external stakeholders, yet delineate concrete ways (e.g., upgrade equipment, attract outstanding software engineers) and a clear timeline when talking to employees about how to implement such ideas. The external communication example involves using a high construal articulation with a distant audience while the internal communication example involves using a low construal articulation with a more proximate audience. This switch from high to low construal is functional because it makes communication more fluent and more readily comprehensible to the respective audiences (Berson et al., 2015).
Second, Joshi and Wakslak (2014) found that communicators use more abstract messages with high-level construal when they encounter a heterogeneous audience and more concrete messages with low-level construal when they encounter a homogeneous audience. Internal communication is primarily directed towards employees which represent a rather homogenous target group in comparison to the external audience because all employees share the same formal membership to the organization (Cardador and Pratt, 2006). Therefore, organizations are expected to use more concrete messages when they target their employees.
Third, Amit et al. (2013) found that a communicator’s preference for using less (vs. more) abstraction is increasingly higher when communicating with proximal (vs. distal) others. Since internal communication addresses a more proximal target group—members of the organization—as compared to external communication, which addresses a more distal target group of nonmembers, organizations are expected to focus on low-level, concrete construal in internal communication, and on high-level, abstract construal in external communication.
Fourth, Douglas and Sutton (2003) found that communicators who portray themselves in a positive light tend to use more abstract statements. External communication such as advertising is primarily an attempt to stimulate consumers and other publics to engage in actions that are favorable to the organization (Gilly and Wolfinbarger, 1998). Not surprisingly, most organizations communicate as positive as possible to consumers and other publics whereas they adopt a more realistic stance toward their employees (Wentzel et al., 2010). This suggests that external communication is more abstract with a higher level of construal compared to internal communication in order to portray the organization in a more positive way.
Taken together, prior research on construal level theory suggests that external brand communication will be more abstract with a higher level of construal compared to internal brand communication, which will be more concrete with a lower level of construal.
Construal fit and organizational identificationConstrual level theory highlights the relationship between social distance and the extent to which an individual’s mental representation of an object is abstract or concrete (Trope and Liberman, 2010). Social distance is a sub dimension of psychological distance, which refers to “the extent of divergence from direct experience of me, here and now along the dimensions of time, space, social perspective, or hypotheticality” (Liberman and Trope, 2014, p. 365). We focus on the social dimension of distance because social distance is central to social identity theory (Ashforth and Mael, 1989), because Joshi and Wakslak (2014) as well as Joshi et al. (2016) established a clear association between social distance and the construal level of communication, and because previous research in intra-organizational settings used social distance when investigating construal fit between different “actors” and communication aspects (e.g., Berson and Halevy, 2014; Cole et al., 2009; Herhausen et al., 2017).
Based on the general definition from construal level theory, social distance to headquarters is defined as an employee’s social divergence from headquarters, evoked by the amount of direct interaction with headquarters (Stephan et al., 2011)2. The central notion of construal level theory is that the more distant an object is from an individual, the higher its construal level and the more abstract its mental representation while the closer an object is, the lower its construal level and the more concrete its mental representation (e.g., Liberman et al., 2007; Soderberg et al., 2014).
Following this notion, employees’ social distance to headquarters affects their representation of headquarters. If employees’ construal of headquarters is high, employees will mentally represent headquarters more abstract and schematic; whereas if employees construal of headquarters is low, employees will mentally represent headquarters more concrete and detailed (e.g., Trope and Liberman, 2010). Thus, employees who have no direct or infrequent interaction with headquarters due to their working environment, for example sales and service employees or blue-collar workers, will experience high social distance to headquarters (Wieseke et al., 2012). On the contrary, employees with direct and regular headquarter interactions, for example managers or administrative staff, will experience low social distance to headquarters.
Construal level theory further suggests that construal fit determines the effectiveness of organizational communication. The reason is that construal fit increases psychological engagement (Berson and Halevy, 2014), enhances perceived credibility of information (Hansen and Wanke, 2010), intensifies emotional reactions to messages (Lee et al., 2009), and makes communication from a specific source more effective (Herhausen et al., 2017). The combined action of these mechanisms should enhance the effectiveness of organizational communication.
Thus, it is proposed that employees’ organizational identification is influenced by the construal fit between their mental representation of headquarters and the construal level of organizational communication. Employees with low social distance to headquarters have a more concrete mental representation of headquarters and thus, might be more receptive to concrete, low-level construal. In such case, there would be a construal fit between internal communication with low-level construal and low social distance to headquarters. In contrast, employees with high social distance to headquarters have a more abstract mental representation of headquarters thus, might be more receptive to abstract, high-level construal. In such case, there would be a construal fit between external communication with high-level construal and high social distance to headquarters.
H1:Organizational identification of employees with low social distance to headquarters is more strongly influenced by internal brand communication with concrete, low-level construal than by external brand communication with abstract, high-level construal.
H2:Organizational identification of employees with high social distance to headquarters is more strongly influenced by external brand communication with abstract, high-level construal than by internal brand communication with concrete, low-level construal.
MethodologyThe predictions are tested in several studies. A series of three preliminary studies examines the association between the internal versus external focus of organizational brand communication and its level of construal. As expected, companies use concrete, low-level construal in internal brand communication and abstract, high level construal in external brand communication. Two multilevel field studies comprising the responses of 1102 employees of an industrial services firm and a retail company and one experimental lab studies with 219 participants test the proposed effects of construal fit. Study 1 demonstrates the proposed relationships for an industrial service firm; Study 2 extends these findings to a retail firm while controlling for the potential effects of subgroup identification; and Study 3 isolates and manipulates the construal level of organizational communication.
Preliminary researchAlthough construal level theory suggests that communicators will prefer low-level construal in internal communication and high-level construal in external communication, both the managerial intention in designing organizational communication as well as the perceptions of receivers are examined to further establish this association. A survey among 18 marketing managers (mean organizational tenure = 6.5 years) during an executive workshop pointed out that 86% of the participants described the internal brand communication of their company as more concrete than the external brand communication3. These participants were further asked to provide samples of “typical” internal and external brand communication which both communicate the same brand values, although towards an internal versus external audience. Eight paired samples of internal and external communication were received4. In a next step, any information that referred to the internal or external focus was excluded from the communication, and a randomized sample of these materials were presented to 20 marketing managers in a different executive workshop. The materials referring to the internal communication were evaluated as more concrete than the material referring to external communication (mean internal = 5.40, mean external = 3.33, t(1, 19) = 5.78, p < .01, α = .61/.69). Moreover, although the managers were blind to the internal or external focus, 75% (85%) of the participants suggested that the internal (external) communication is better suited to an internal (external) audience. Using the same samples, this study was re-run with 97 business master students from a Swiss university. The prior findings were replicated: The internal communication was rated as more concrete than the external communication (mean internal = 5.00, mean external = 3.62, t(1, 96) = 8.07, p < .01, α = .76/.76), and 78% (84%) of the participants suggested that the internal (external) communication is better suited to an internal (external) audience. Thus, consistent with prior research on the interrelatedness of construal level theory and communication, the results support the notion that internal (external) communication is associated with more concrete, low-level (abstract, high-level) construal.
Field studiesStudy 1Data for Study 1 was collected from 549 non-managerial employees in 24 departments of a multinational industrial service firm in Switzerland.5 These employees were located at the same site in a different city than the firm’s headquarters, and all employees stated that they have been exposed to both external and internal brand communications. Organizational brand values of the firm include “sustainability” and “passion”, and the corporate communications department located at headquarters is responsible for all organizational brand communication. Interviews with communication managers revealed that the internal brand communication is typically more concrete than external brand communication. For example, while internal communication focuses on how employees should behave in line with the firm’ brand values, external communication focuses on why the firm emphasizes certain brand values6.
MeasuresAll measures are displayed in Table 1. Organizational identification was measured with Mael and Ashforth’s (1992) well-established scale, and an objective measure of social distance to headquarters obtained from the HR department of the firm was used. Job profiles were categorized as “low distance” (61% of employees) for which the HR department indicated direct contact to headquarters, while all other job profiles without direct contact to headquarters were categorized as “high distance” (39% of employees). The measures of internal and external communication quality were based on Hughes (2013). Departmental size was used as a control variable because the span of supervisor control may influence organizational identification (Wieseke et al., 2009); tenure because the time an employee spends in contact with an organization may influence organizational identification (Cardador and Pratt, 2006); perceived external prestige because employees identify with an organization partly to enhance self-esteem (Mael and Ashforth, 1992); and perceived internal competition because intraorganizational competition may trigger a focus on the competing subunits (Mael and Ashforth, 1992). Table 2 presents descriptive statistics and intercorrelations.
Measurement of Constructs.
| Construct and measurement items | Study 1 | Study 2 | ||
|---|---|---|---|---|
| α / AVE | Loadings | α /AVE | Loadings | |
| Organizational identification (Mael and Ashforth, 1992) | .89 / .62 | .91 / .67 | ||
| .88 | .91 | ||
| .70 | .90 | ||
| .64 | .70 | ||
| .84 | .80 | ||
| .87 | .78 | ||
| Internal communication quality (based onHughes, 2013) | .95 / .76 | .85 /.51 | ||
| .74 | .81 | ||
| 84 | 89 | ||
| 88 | 95 | ||
| 92 | 38 | ||
| .93 | 39 | ||
| 91 | 64 | ||
| External communication quality (based onHughes, 2013) | .93 / .69 | .94 / .75 | ||
| .78 | .89 | ||
| . 77 | 88. | ||
| . 83 | .88 | ||
| 84 | — | ||
| .87 | 86 | ||
| 89 | 81 | ||
| Social distance to headquarters | n.a. | n.a. | n.a. | n.a. |
| Objective measure of the functional role of the employee obtained from the HR department (i.e., role with and without direct headquarters contact) | ||||
| Perceived external prestige (Mael and Ashforth, 1992) | .89 / .63 | n.a. | n.a. | |
| .84 | |||
| .74 | |||
| .83 | |||
| .81 | |||
| .73 | |||
| Perceived internal competition | n.a. | n.a. | — | — |
| The competition between different departments at [x] is intense. | ||||
| Identification with the store | — | — | n.a. | n.a. |
| I feel strong ties with the store I am working for. | ||||
| Physical proximity to headquarters | — | — | n.a. | n.a. |
| Measured as the absolute physical distance in miles between each store’s location and the location of the headquarters. | ||||
| Perceived social distance to headquarters | — | — | n.a. | n.a. |
| How would you rate the amount of contact with [x]’s headquarters? | ||||
Note: Translated items. All standardized loadings are significant at p < .01. n.a. = not applicable.
Descriptive statistics and correlations in the field studies.
| Variables | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Level 2: Department / Storea | |||||||||||
| 1. | Size (number of employees) | .10 | .01 | −.04 | — | −.06 | .02 | .00 | .05 | −.08 | |
| 2. | Physical Proximity (in miles) | — | −.20 | .03 | — | .01 | .04 | .06 | −.04 | .03 | |
| Level 1: Employee | |||||||||||
| 3. | Tenure (in years) | .07 | — | .09 | — | −.20 | −.23 | −.24 | − | −.13 | |
| 4. | Identification with the Store | — | — | — | — | .27 | .18 | .35 | –.28 | .43 | |
| 5. | Perceived Internal Competition | –.11 | — | .05 | — | — | — | — | — | — | |
| 6. | Perceived External Prestige | .09 | — | .07 | — | –.12 | .27 | .42 | –.28 | .36 | |
| 7. | Internal Communication Quality | –.17 | — | –.03 | — | –.06 | .22 | .33 | –.28 | .32 | |
| 8. | External Communication Quality | .07 | — | .02 | — | –.15 | .52 | .21 | –.36 | .44 | |
| 9. | Social Distance | .37 | — | .01 | — | –.11 | .10 | –.36 | .19 | –.34 | |
| 10. | Organizational Identification | –.18 | — | .05 | — | –.05 | .45 | .47 | .34 | –.29 | |
| Study 1: Mean | 35.55 | — | 4.25 | — | 4.68 | 5.18 | 4.42 | 4.38 | .61 | 5.15 | |
| Study 1: SD | 16.62 | — | 1.99 | — | 1.69 | 1.22 | 1.51 | 1.33 | .49 | 1.29 | |
| Study 2: Mean | 13.38 | 393.81 | 2.67 | 4.53 | — | 4.81 | 4.24 | 4.19 | .57 | 4.66 | |
| Study 2: SD | 11.38 | 188.36 | 1.49 | 1.33 | — | 1.03 | 1.10 | 1.14 | .50 | 1.16 | |
Correlations are based on scores disaggregated per employee. Study 1 (2) correlations are reported below (above) the diagonal. For Study 1, |r| ≥ .09 is significant at p < .05 and |r| ≥ .11 is significant at p < .01 (two-tailed tests), and all scales have been measured with 6-porganizational identificationnt Likert scales (except size, tenure, and social distance). For Study 2, |r| ≥ .09 is significant at p < .05 and |r| ≥ .11 is significant at p < .01 (two-tailed tests), and all scales have been measured with 7-porganizational identificationnt Likert scales (except size, physical proximity, gender, age, tenure, and social distance). Study 1: NDepartment = 24, NEmployee = 549; Study 2: NStore = 108, NEmployee = 553.
Squared correlations were lower than average variance extracted for any pair of constructs. A confirmatory factor analysis of the measurement model indicated good fit of the model with the data (χ2(203) = 468.73; CFI = .97; RMSEA = .05). Multilevel modeling with robust maximum likelihood estimation was used to accommodate the multilevel nature of the study, i.e. that employees are nested in departments. Log-likelihood difference tests were used to compare the nested models, and all predictor variables were grand mean-centered. The intraclass correlation coefficient indicates that 8% of the variation in organizational identification was related to departmental membership.
Table 3 displays the results of the structural models. Both internal and external communication quality were positively related to organizational identification. Results show a negative effect of the interaction between internal communication quality and social distance to headquarters on employees’ organizational identification (γ = −.23, p < .01) and a positive effect of the interaction between external communication quality and social distance to headquarters on employees’ organizational identification (γ = .28, p < .01). Organizational identification of employees with low social distance to headquarters is more strongly influenced by internal (γ = .38, p < .01) than by external communication quality (γ = −.02, n.s.), and a Chi-square difference test for parameter equality provides support for Hypothesis 1 (Δχ2 = 16.73, d.f. = 1, p < .01). Organizational identification of employees with high social distance to headquarters is more strongly influenced by external (γ = .26, p < .01) than by internal communication quality (γ = .15, p < .01; Δχ2 = 2.79, d.f. = 1, p < .10), supporting Hypothesis 2.
Results of the field studies.
| Study 1 | Study 2 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Model 1: Controls | Model 2: Main Effects | Model 3: Full Model | Model 4: Controls | Model 5: Main Effects | Model 6: Full Model | |||||||
| Level and Variable | γ | SE | γ | SE | γ | SE | γ | SE | γ | SE | γ | SE |
| Intercept | 5.15** | .07 | 5.15** | .06 | 5.06** | .07 | 4.66** | .05 | 4.65** | .05 | 4.69** | .05 |
| Level 2: Department / Store | ||||||||||||
| Size | –.01* | .00 | –.01a | .00 | –.01a | .00 | –.01 | .00 | –.01 | .01 | –.01a | .00 |
| Physical Proximity | — | — | — | — | — | — | .00 | .04 | .00 | .04 | .00 | .04 |
| Level 1: Employee | ||||||||||||
| Tenure | .02 | .02 | .03 | .02 | .03 | .02 | –.07* | .03 | –.02 | .03 | –.02 | .03 |
| Identification with the Store | — | — | — | — | — | — | .31** | .04 | .24** | .04 | .23** | .03 |
| Perceived Internal Competition | –.03 | .03 | .00 | .03 | –.01 | .03 | — | — | — | — | — | — |
| Perceived External Prestige | .51** | .05 | .35** | .04 | .34** | .04 | .23** | .05 | .12** | .05 | .10* | .05 |
| Social Distance | –.29** | .04 | –.23** | .04 | –.18** | .04 | –.20** | .05 | –.10* | .05 | –.11* | .04 |
| Internal Communication Quality | .21** | .03 | .24** | .03 | .17** | .04 | .22** | .04 | ||||
| External Communication Quality | .17** | .04 | .15** | .04 | .27** | .04 | .23** | .04 | ||||
| Interaction Effects | ||||||||||||
| Internal Communication Quality × Social Distance | –.23** | .07 | –.23** | .04 | ||||||||
| External Communication Quality × Social Distance | .28** | .07 | .11** | .04 | ||||||||
| Log-likelihood | –803.92 | –767.84 | –758.17 | –765.26 | –735.40 | –719.31 | ||||||
| (–2LL change) | 36.08** (d.f. = 2) | 9.67** (d.f. = 2) | 29.86** (d.f. = 2) | 16.09** (d.f. = 2) | ||||||||
| AIC | 1’623.85 | 1’555.68 | 1’540.34 | 1’548.52 | 1’492.80 | 1’464.63 | ||||||
| BIC | 1’658.31 | 1’598.76 | 1’592.03 | 1’587.29 | 1’540.17 | 1’520.61 | ||||||
Significance is based on two-tailed tests. Effects with robust standard errors are reported. Study 1: NDepartment = 24, NEmployee = 549; Study 2: NStore = 108, NEmployee = 553.
To test for the robustness of the results, it was tested whether employees with low and high social distance to headquarters differ in their attention to internal and external organizational communication. Therefore, all employees were asked whether they are able to recall the last time they encountered each type of organization communication. Employees with low (high) social distance recalled internal and external communication equally well (low distance: internal = 99% vs. external = 89%, n.s.; high distance: internal = 69% vs. external = 69%, n.s.). Thus, it is unlikely that the observed effects occur due to systematic differences in the attention to internal and external brand communication.
Study 2Study 2 replicates Study 1 in a different context—with 553 non-managerial employees from 108 stores of a German retailer. Organizational brand values of the firm include “customer focus” and “responsibility”. The corporate communications department at headquarters is responsible for all organizational brand communication, and confirmed that the internal communication of brand values is typically more concrete than the external communication. Data was collected from employees of various stores, while controlling for physical proximity of the stores to headquarters and employees’ identification with the store. We had to exclude 14 employees from the initial sample of 567 participants because these employees stated that they have not been exposed to external brand communication, a pre-requisite to assess the quality of communication.
MeasuresThe same measures as in Study 1 were used (see Table 1). In addition, Study 2 also controlled for physical proximity to headquarters because with increasing physical distance, the organizational as a whole may become less attractive as a target for employees’ identification (Wieseke et al., 2012); and identification with the store, because identification might exist at multiple levels independently from each other, and identification with proximal entities may be inductive to identification with more distal collectives (Wieseke et al., 2012). Further, employees’ perceived social distance to headquarters was measured to validate the objective measure obtained from the HR department of the retailer.
Analysis and resultsSquared correlations were lower than average variance extracted for any pair of constructs. A confirmatory factor analysis of the measurement model indicated good fit to the data (χ2(101) = 276.40; CFI = .97; RMSEA = .06). The intraclass correlation coefficient indicated that 6% of the variation in organizational identification was related to store membership.
Table 3 shows that both internal and external communication quality were positively related to organizational identification. Results further include a negative interaction between internal communication quality and social distance to headquarters on employees’ organizational identification (γ = −.23, p < .01), and a positive interaction between external communication quality and social distance to headquarters on employees’ organizational identification (γ = .11, p < .01). While organizational identification of employees with low social distance to headquarters is more strongly influenced by internal (γ = .42, p < .01) than by external communication quality (γ = .14, p < .05; Δχ2 = 8.57, d.f. = 1, p < .01), organizational identification of employees with high social distance to headquarters is more strongly influenced by external (γ = .36, p < .01) than by internal communication quality (γ = −.04, n.s.; Δχ2 = 22.08, d.f. = 1, p < .01). Thus, both Hypothesis 1 and Hypothesis 2 are supported.
Robustness analysesParticipants in the low distance condition rated perceived social distance to headquarters lower than those in the high distance condition (F(1548) = 575.51, p < .01), providing support for the objective measure. We replicated our analyses with the perceived social distance to headquarters measure. As predicted, we found a negative effect of the interaction between internal brand communication and perceived social distance to headquarters on employees’ organizational identification (γ = −.11, p < .01) and a positive effect of the interaction between external brand communication and perceived social distance to headquarters on employees’ organizational identification (γ = .10, p < .01). A further test examining whether subgroup identification affects the postulated relationships revealed no significant effects of identification with the store on the effectiveness of internal communication quality (γ = −.03, n.s.) and external communication quality (γ = −.04, n.s.).
Discussion of the field studiesTwo field studies support the postulated effect of construal fit. When social distance to headquarters is high (low), the quality of external (internal) brand communication is more important than the quality of internal (external) brand communication to increase organizational identification of employees. However, these findings are based on cross-sectional data, and the research design did not differentiate between the construal level and the internal versus external focus of the brand communication. To address these limitations, the following experimental study isolates and manipulates the construal level of organizational brand communication.
Experimental studyStudy 3 examines the causal effect of construal fit between organizational brand communication and social distance to headquarters on organizational identification using an experimental study with 219 participants. Actual organizational brand communication of the multinational industrial service firm from Study 1 was used as stimuli to increase the external validity of the study.
MethodStudy 3 used a 2 × 2 × 2 ([low vs. high social distance to headquarters] × [internal vs. external brand communication] × [brand communication with low-level construal vs. high-level construal]) between-subject design, yielding eight different scenarios. The experimental design was pretested with 6 participants who did not take part in the main study. These respondents were interviewed, and the wording of some of the manipulations was adjusted. 223 business master students from a Swiss university were recruited to participate in the study (44% female, mean age = 25.10 years). Importantly, 81% of these students were employed for more than six months prior to starting their master studies, while half of them is currently employed part-time. Notably, the average working experience of our sample is 1.35 years. Accordingly, we are confident that these students have already gathered some essential work experience that allowed them to realistically step into the given scenarios and provide answers with meaningful explanatory power (Aguinis and Bradley, 2014). Extant research in the field of economics provides evidence that the results of laboratory studies comparing students versus professionals are largely the same, concluding that the standard experimental subject pool, i.e., students, can be generalized to professionals (Frechette (2015)).
As an incentive, participants were entered into a lottery for movie gift cards. Participants were randomly assigned to read one of the eight scenarios. Because the experiment was designed with the multinational industrial service firm which participated in Study 1, the experimental setting may confound with students prior experiences and knowledge about that firm. Accordingly, participants were asked how they evaluate the perceived external prestige of the firm, whether they had known the firm prior to the study (all participants), and whether they were or are currently employed by the firm (n = 4; excluded from all analyses).
The stimuli are displayed in Fig. 1. Participants were exposed to specific information about their situation at work to manipulate social distance to headquarters. They were told to have direct contact with headquarters and a picture of the management board was presented (low distance), or they were told to have no direct contact with headquarters and a picture of the headquarters building from the outside was presented (high distance). To manipulate organizational brand communication, either an intranet article (internal brand communication) or advertisement (external brand communication) was presented which communicated one of the organizational brand values of the firm. Both internal and external brand communication either used a concrete “how” messages (low-level construal) or an abstract “why” messages (high-level construal) referring to the same brand value. Both messages have actually been used by the firm.
After participants had read one of the scenarios, they completed a short survey and responded to manipulation checks. Measures include organizational identification (Mael and Ashforth, 1992; α = .90), perceived social distance to headquarters (“How would you rate the amount of direct contact with [x]’s headquarters?”), communication type (“internal vs. external”), abstractness of the communication (“The brand communication is abstract”), perceived quality of the communication (Hughes, 2013; α = .82), and perceived external prestige (Mael and Ashforth, 1992; α = .83).
ResultsTable 4 displays means and standard deviations across all experimental conditions. The manipulation checks confirmed that all experimental conditions were introduced successfully. Participants in the low distance conditions rated social distance to headquarters lower than those in the high distance conditions (F(1 218) = 75.08, p < .01), all participants correctly indicated the type of communication, and participants in the low-level construal conditions rated abstractness of communication lower than those in the high-level construal conditions (F(1, 218) = 145.32, p < .01). A three-way ANOVA revealed significant effects of social distance to headquarters (F(1, 218) = 9.80, p < .01), perceived quality of the communication (F(1, 218) = 6.01, p < .05), and the predicted two-way interaction between social distance to headquarters and construal level of communication (F(1, 218) = 20.33, p < .01) on organizational identification. Importantly, neither construal level of communication, nor type of communication, nor any other interaction is significantly related to organizational identification.
Means and standard deviations across experimental conditions.
| Low social distance to headquarters | |||||
|---|---|---|---|---|---|
| Internal Brand Communication | Concrete | Abstract | |||
| Mean | SD | Mean | SD | ||
| Social distance to headquarters | 3.62 | 1.36 | 3.80 | 1.78 | |
| Perceived abstractness of communication | 2.16 | .90 | 5.04 | 1.70 | |
| Quality of communication | 4.26 | 1.07 | 4.72 | .75 | |
| Perceived external prestige | 5.02 | .81 | 4.33 | .83 | |
| Organizational identification | 5.43 | .71 | 4.92 | 1.05 | |
| External Brand Communication | Concrete | Abstract | |||
| Mean | SD | Mean | SD | ||
| Social distance to headquarters | 3.81 | 1.30 | 3.38 | 1.24 | |
| Perceived abstractness of communication | 2.08 | 1.20 | 5.02 | 1.29 | |
| Quality of communication | 3.58 | 1.05 | 3.66 | 1.16 | |
| Perceived external prestige | 4.49 | 1.12 | 4.79 | .88 | |
| Organizational identification | 5.44 | .93 | 5.05 | .84 | |
| High Social Distance to Headquarters | |||||
| Internal Brand Communication | Concrete | Abstract | |||
| Mean | SD | Mean | SD | ||
| Social distance to headquarters | 4.81 | 1.17 | 5.29 | .66 | |
| Perceived abstractness of communication | 2.73 | 1.54 | 4.07 | 1.63 | |
| Quality of communication | 3.96 | .88 | 4.60 | .85 | |
| Perceived external prestige | 5.39 | .86 | 5.00 | .89 | |
| Organizational identification | 4.66 | 1.00 | 5.20 | .93 | |
| External Brand Communication | Concrete | Abstract | |||
| Mean | SD | Mean | SD | ||
| Social distance to headquarters | 5.04 | .81 | 5.11 | 1.13 | |
| Perceived abstractness of communication | 2.48 | 1.40 | 4.54 | 1.40 | |
| Quality of communication | 4.39 | 1.18 | 4.43 | .96 | |
| Perceived external prestige | 4.68 | .96 | 5.34 | .84 | |
| Organizational identification | 4.43 | 1.00 | 5.29 | .89 | |
Note: The possible range of scores for listed values is 1–7, with higher values indicating more positive response.
Planned contrasts displayed in Fig. 2 revealed that concrete brand communication had a stronger impact on organizational identification compared to abstract brand communication when social distance to headquarters was low (mean concrete = 5.43, mean abstract = 4.99, F(1, 109) = 7.19, p < .01). A reverse pattern was found when social distance to headquarters was high: Organizational identification was affected more by abstract than by concrete brand communication (mean concrete = 4.55, mean abstract = 5.24, F(1, 108) = 14.69, p < .01). These effects occurred regardless of whether internal or external communication was used. Therefore, the experimental study provides support for the causality of the hypothesized effects of construal fit on organizational identification.
General discussionTheoretical contributions
The aim of this research was to investigate the relative effects of internal versus external brand communication on organizational identification, and to test for the moderating role of social distance to headquarters. Using a multiple method approach, findings across three studies in different settings were consistent: organizational brand communication is most effective to increase employees’ organizational identification in case there is construal fit between an employee’s construal of headquarters and the construal of the brand message.
The present studies are the first to offer a theoretical integration and empirical investigation of the combined effects of internal versus external brand communication on employees’ organizational identification. Historically, separate lines of research investigated the effect of organizational communication, and focused either on internal or external communication only. The present research combines those perspectives and point out that both types of brand communication may explain employees’ organizational identification above and beyond the effects of perceived external prestige and other control variables. Therefore, it adds to the emerging literature on the use of brand communication to ensure that employees are identified and aligned with the organization (e.g., Baker et al., 2014; Hughes, 2013), and provides a novel and insightful perspective on employer branding (e.g., Knox and Freeman, 2006).
In line with construal level theory, this study uncovers an association between the internal versus external focus of organizational brand communication and its level of construal. The present findings suggest that organizational communicators prefer low-level construal to address an internal audience, and high-level construal to address an external audience. Uncovering this association is meaningful to understand the relative impact of internal and external brand communication. The studies suggest that both types of organizational brand communication do not per se lead to different outcomes, but rather that the construal level qualifies the effectiveness of organizational communication on different employees. This is important because marketing scholars have repeatedly suggested using external communication to address an internal audience (e.g., Gilly and Wolfinbarger, 1998; Hughes, 2013; Wentzel et al., 2010). However, when doing so, the different levels of construal used in messages need to be explicitly considered because external communication with high-level construal is not equally effective for all employees.
To account for the relative effectiveness of organizational brand communication, the concept of construal fit was introduced and the interplay of organizational communication and social distance to headquarters was investigated. Previous research assumed that—depending on the research focus—either internal or external communication is effective across multiple contexts (e.g., Gilly and Wolfinbarger, 1998; Smidts et al., 2001). However, based on construal level theory, the present findings revealed that organizational communication is only likely to exert the desired effects on organizational identification when the construal level of the communication matches the social distance between an employee and headquarters. Thus, this research identified an important contingency factor, social distance to headquarters, which determines the effectiveness of organizational brand communication.
Practical implicationsComplementing the theoretical implications, the present findings also provide important insights for brand managers and employer branding. To date, previous studies have advised managers to use organizational communication to fuel organizational identification’s “perpetual work in progress” (Ashforth, 1998, p. 213). However, the research findings highlight that this effect might be more complex than previously assumed. In order to increase organizational identification among employees with low social distance to headquarters, managers should focus on brand communication with low-level construal. On the contrary, in order to increase organizational identification among employees with high social distance to headquarters, managers should focus on brand communication with high-level construal.
Because companies typically use more abstract messages with high-level construal in their external brand communication, the present results provide striking arguments for marketing managers to convince the top management to keep or increase external communication budgets in order to not only address potential customers but also employees who are psychological distant to headquarters. Therefore, it is recommended that managers consider the internal effects of external brand communication when deciding on their communication budgets. Moreover, managers should take employees’ perception of external communication into consideration when briefing agencies on new communication campaigns. For example, the multinational industrial service firm from Study 1 communicated its sustainable energy solutions to consumers and other publics with advertisements consisting of the message “we are able to reduce energy consumption of cities by 30%.” By doing so, a very complex technological solution is translated to an abstract statement, which communicates the firm’s organizational brand value “sustainability”. However, the firm internally communicated the same organizational brand value by an intranet article consisting of much more detailed information that covers technical and economic aspects as well as more specific examples on how and where its energy-saving products and services are applied. While such procedure may successfully address psychological close employees, the external communication could complement the internal communication to address psychological distant employees more effectively.
In addition, both field studies revealed that social distance to headquarters is negatively related to organizational identification. Thus, it appears to be a managerial priority to address employees with high social distance to headquarters in order to increase their organizational identification. The results reveal a substitution effect between organizational brand communication with high-level construal and social distance. From a managerial perspective, this substitution effect suggests that employees’ organizational identification can be enhanced by allocating resources either to abstract brand communication or to decrease social distance. However, considering that some employees have positions with infrequent or no exchange with headquarters (e.g., employees low in the organizational hierarchy), using brand communication with high-level construal to target these employees might be easier to prosecute than to decrease social distance for all employees. Managers might decide, for instance, to specifically target this internal audience with abstract advertisement.
Limitations and directions for future researchThis research is not without limitations, which also suggest promising avenues for future research. Regarding the two field studies, the theoretical model was tested by analyzing data collected in highly developed, de-regulated markets with strong competition. As organizational identification may be influenced by environmental factors (Wieseke et al., 2012), the generalizability of the findings beyond the population is limited. Hence, future studies may replicate the findings in different environmental conditions. Regarding the lab experiments, although the study samples covers a sampling domain (i.e., business master students) closely related to the sample of the field studies (i.e., actual employees), the different samples do not overlap. This shortcoming might reduce the external validity of these studies to some extent. Thus, future research should investigate the suggested causal relationships via a longitudinal or quasi-experimental study with actual employees of an organization.
More generally, while high construal level may dominate in internal brand communication and low construal level in external brand communication, the level of construal largely depends on specific goals for organizational communication campaigns. Although research on construal level theory would suggest otherwise (i.e., that the audience influences the construal of communication; Joshi and Wakslak, 2014; Joshi et al., 2016), different types of organizational communication may use different construal levels. For example, companies may potentially use internal communications to inform about a change in values with a high construal level, or external communications to inform customers about how to use the products with low construal level. Thus, our findings only hold for organizational brand communication where construal level is used in line with the audience. Moreover, it would be interesting to understand why employees who feel a greater social distance from the company are more influenced by communication that is more abstract. This relationship could result from the process of building the brand image, and provides an important research area for subsequent studies. Finally, the underlying cognitive processes of construal fit, which strengthen the brand communication-organizational identification link, were not explicitly examined. For instance, it would be interesting for future research to investigate the nature of employees’ mental organizational representation (including the strength and diversity of attributes) in conjunction with various levels of distance to the organization. Analyzing these mechanisms may yield deeper insights into the process of how construal fit affects organizational identification, and thus, would be a fruitful avenue for future research.
In line with Wieseke et al. (2012) and Wiesenfeld et al. (1999), distance to corporate headquarters is particularly likely to determine organizational identity. Moreover, the CEO is one representative of corporate headquarters but not the only one, and corporate headquarters represent a clear subgroup that possesses multiple interfaces with other organizational members (Reade, 2001).
It is important to note that identification can exist in the absence of direct interaction (Cardador and Pratt 2006). Thus, employees do not need to directly interact with headquarters in order to identify with the organization.
Measured based on three dimensions along 7-Point Scales: The internal (external) communication of my company is (1) concrete vs. abstract, (2) detailed vs. general, (3) specific vs. nonspecific (α = .69/.73)
Data were sampled in the following organizations: four manufacturers, two service providers, one retailer, and one university. The samples are available from the authors.
For this study, we had intentionally chosen only non-managerial employees because managers or any employees with higher responsibilities in our sample may have responded differently (e.g. more favorably) to internal communications not because of its low construal level, but simply because of their higher engagement with the organization.
Due to a non-disclosure agreement, the authors are not allowed to display examples of internal and external brand communication.