Marketing communications of brand-related stereotypes and consumer responses: a mixed-method approach Open Access

In recent years, research on stereotypes in general and on brand-related stereotypes in particular has been gaining momentum (Diamantopoulos et al., 2021; Gidaković et al., 2021; Kolbl et al., 2020). Broadly defined, stereotypes capture a “socially shared set of beliefs about traits that are characteristic of members of a social category” (Greenwald and Banaji, 1995, p. 14) and function as “energy-saving devices that serve the important cognitive function of simplifying information processing and response generation” (Macrae et al., 1994, p. 14). In the marketing literature, three distinct types of brand-related stereotypes can be identified that, respectively, refer to consumers’ shared and oversimplified beliefs about (a) the brand itself, namely, the brand stereotype (e.g. Kervyn et al., 2012; Davvetas and Halkias, 2019), (b) the country from which the brand originates, known as the brand origin stereotype (e.g. Diamantopoulos et al., 2017; Magnusson et al., 2019), and (c) individuals/groups perceived to be users or buyers of a particular brand, commonly referred to as the brand buyer/user stereotype (e.g. Antonetti and Maklan, 2016; Ziano and Pandelaere, 2018).

In practice, “marketers often depict brands, their origin and their users simultaneously in brand communications in an attempt to purposefully stimulate consumers’ stereotypical perceptions via utilizing stimuli that activate favorable associations in consumers’ minds” (Gidaković et al., 2021, p. 1925, added emphasis). Most extant research on brand-related stereotypes has focused, almost exclusively, on only one type of stereotype at a time (i.e. either brand stereotypes or brand origin stereotypes or brand buyer/user stereotypes – see Gidaković et al., 2021 for a summary of relevant research). Such neglect of potential cross-stereotype influences hinders a comprehensive understanding of how brand-related stereotypes affect consumer attitudes and overlooks potential synergies among stereotypes. Relevant studies in the field have also invariably adopted a consumer perspective to investigate the role of each brand-related stereotype, relying on consumers’ self-reports of their stereotypical perceptions of specific brands, their origins, and their buyers/users. As a result, the company perspective has been neglected: the question of how firms seek to portray stereotypes through their marketing communications remains unanswered. Such a perspective assumes a reactive rather than a proactive role for firms, even though it is the latter that intentionally creates and communicates brand-related stereotypes. It is not clear which brand-related stereotypes are deployed by companies (and with what frequency); what stereotype content (in terms of warmth and competence) is communicated; or how consumers respond to the communicated brand-related stereotypes (Septianto et al., 2022). Table 1 summarizes the research gaps and motivation of the study.

Against the background of Table 1, the current study seeks to address the following research questions by using a two-phase, mixed-methods research design (Christofi et al., 2024). The latter was deemed necessary to accommodate data collection needs from both company communications and generate empirical insights based on both perspectives:

• Which brand-related stereotypes do firms deploy in their marketing communications? What is the relative incidence of brand-, brand origin- and brand buyer/user stereotypes?

• What stereotype content do firms emphasize when communicating the three brand-related stereotypes?

• How does the content of the communicated brand-related stereotypes shape consumer attitudes toward the brand?

In the first phase, we focus on the firm perspective to identify what stereotypical information brands communicate to consumers and apply content analysis to identify brand-related stereotypes in a sample of print advertisements of brands in different product categories. In the second phase, we focus on the consumer perspective and use multilevel modeling as well as qualitative interviews to investigate consumer responses to the communicated brand-related stereotypes, identified in the first phase of our study.

To conceptually underpin our investigation and capture the content of the three brand-related stereotypes, we draw on the Stereotype Content Model (SCM), which is the most widely used model in stereotyping research (Fiske, 2018; Halkias and Diamantopoulos, 2020; Yzerbyt, 2016). The SCM offers a common conceptual denominator for conceptualizing and operationalizing different brand-related stereotypes (Gidaković et al., 2021), thus facilitating the comparison of stereotype content relating to specific brands, brand origins and brand buyers/users.

Our intended contribution is threefold. First, we contribute to stereotyping literature in a consumption context by offering the first assessment of how firms communicate brand-related stereotypes in practice. In doing so, we document the relative popularity of brand- vs brand origin- vs brand buyer/user stereotypes portrayed in print advertisements as well as the configurations reflecting different stereotype combinations. Second, we extend current knowledge on the SCM by revealing the extent to which firms emphasize the stereotypical dimension of warmth vs the dimension of competence in their communication of brand-related stereotypes. We thus offer a supply-side perspective on the SCM and the role of warmth and competence in determining stereotype content, the latter having previously been studied only from a demand-side perspective. Third, we investigate how the warmth/competence of the communicated stereotypes influence consumer attitudes toward the focal brand. From a managerial point of view, our study provides guidance regarding which particular brand-related stereotype (and which particular stereotypical dimension) to emphasize in marketing communications to maximize the positive impact on consumer brand attitudes.

The SCM (Fiske et al., 2002; Kervyn et al., 2021) is the most prominent theoretical framework for capturing the content of people’s stereotypical beliefs about “others,” the latter being both social groups and nonhuman/inanimate entities (Fiske, 2015, 2018; Fiske et al., 2007). According to the SCM, stereotypical beliefs are reflected in two fundamental dimensions: warmth and competence. These are “universal dimensions of social perception that endure across stimuli, time, and place” (Dupree and Fiske, 2017, p. 28). The former dimension relates to whether “others” have positive or negative intentions, while the latter dimension captures whether these “others” are actually able to enact these intentions (Fiske et al., 2002). Stereotyped entities are described as warm (cold), if they signal good (bad) intentions, and competent (incompetent) if they exhibit the capability (or lack of) to enact these intentions (Cuddy et al., 2008). Operationally, the dimension of warmth is captured by such items as “warm, trustworthy, friendly, honest, likeable, and sincere […]. Competence items include competent, intelligent, skilled and efficient, as well as assertive and confident” (Fiske, 2018, p. 68, original emphasis). Note that both stereotype content dimensions involve cognitive appraisals about “others”, although the warmth dimension is sometimes misinterpreted as capturing affect (e.g. Chattalas et al., 2008; Stokburger-Sauer et al., 2012). Note also that stereotypes can sometimes be “mixed”, that is, a particular target entity may score high on one dimension but low on the other (Fiske et al., 2002; Judd et al., 2005).

The SCM has been widely adopted in empirical research on brand-related stereotypes. According to the Brands as Intentional Agents framework (Kervyn et al., 2012, 2021) – which is the adaptation of the SCM in a branding context – consumers perceive the content of a brand’s stereotype in terms of its (good/bad) intentions (i.e. warmth) and its ability/inability to act on these intentions (i.e. competence). For example, consumers perceive Hershey’s as a brand with good intentions and high ability to enact them, whereas Marlboro is perceived as a brand with bad intentions and low/medium ability to actually enact them (Kervyn et al., 2012). A growing body of research shows that brand warmth positively and significantly influences consumer brand identification (Kolbl et al., 2019; Stokburger-Sauer et al., 2012), brand intimacy and passion (Davvetas and Halkias, 2019). Brand competence, on the other hand, is a particularly important driver of purchase intentions (Aaker et al., 2012; Jakubanecs et al., 2023), as well as consumer-brand relationship quality and brand loyalty (Valta, 2013). In recent years, numerous studies have investigated the roles of brand warmth and competence as predictors or mediators in a variety of consumer responses (Chua et al., 2024; Gong et al., 2021; Mazzoli et al., 2024; Meyer et al., 2024; Zhang and Wang, 2025; Zhang et al., 2022).

Regarding brand origin stereotypes, extant research has revealed that there are three different stereotypical clusters, which consist of countries perceived as high on warmth and low on competence (e.g. Portugal); countries medium on both warmth and competence (e.g. Sweden); and countries low on warmth and high on competence (e.g. Germany) (Cuddy et al., 2009). Country warmth positively influences consumers’ behavioral intentions of products from that country (Maher and Carter, 2011), even in the cases of product failure (Xu et al., 2013). Country competence, on the other hand, exhibits a positive impact on brand attitude and favorable product evaluations (Chen et al., 2014). Brand origin stereotypes influence consumers’ brand preferences, as well as the actual ownership of brands coming from these countries (e.g. Diamantopoulos et al., 2017; Halkias et al., 2016; Magnusson et al., 2019; Papadopoulos et al., 2018). Country stereotypes (captured by warmth and competence) also positively influence country-related emotions of admiration, which, in turn, result in customers intentions to visit the tourist destination (Micevski et al., 2021). In the context of destination branding, warmth is more diagnostic than competence in shaping destination advocacy and identification, with gender stereotypes also influencing such perceptions (Hamdy et al., 2024).

Regarding brand buyer/user stereotypes, these have received scant attention compared to brand- and brand origin stereotypes. Research suggests that consumers are often reluctant to buy socially responsible brands due to the pronounced warmth dimension ascribed to their buyers; instead, it is buyer competence which elicits envy and fuels “a desire to emulate a consumption group” (Antonetti and Maklan, 2016, p. 796).

Recent empirical findings have revealed that “consumers navigate their brand preference through simultaneously stereotyping brands, their origin and their users.” Therefore, investigating stereotypes in isolation of each other as documented in previous research “[…] can lead to a fragmented and possibly biased view of the role of brand-related stereotypes in a consumer behavior context” (Gidaković et al., 2021, p. 1941). This last point is of particular relevance for the current investigation, which seeks, on the one hand, to document how firms deploy brand stereotypes in their advertising and, on the other, assess how consumers’ attitudes are actually affected by the communicated stereotypes. For any one stimulus brand, up to three stereotypes have to be captured in terms of warmth/competence (relating to the brand itself, its buyers/users and its origin) and consumer responses – as reflected in brand attitudes – have to be linked to the combined content of the communicated stereotypes.

Our investigation was conducted in Austria, and our research design consists of two phases using a combination of research methods, as summarized in Table 2.

In Phase 1, we use content analysis to examine how firms portray brand-related stereotypes in their print advertisements, in terms of (1) which stereotypes are used, and (2) which stereotypical dimensions are particularly emphasized. Content analysis is the method of choice as this is a widely used approach in stereotyping research to capture the firm perspective [e.g. female role stereotyping – Plakoyiannaki and Zotos (2009); gender stereotypes in advertising – Grau and Zotos (2016); gender role portrayals in Japanese advertising – Ford et al. (1998); gender roles and humor in advertising; Eisend et al. (2014)].

In Phase 2, we investigate the influence of companies’ stereotypical portrayals of brands, their origins and their buyers/users (as established in Phase 1) on consumer attitudes. That is, we assess whether and how communicated ads with stereotypical portrayals influence attitudes toward the focal brand. Using the brands of Phase 1 as stimuli, we use multilevel modeling to relate brand stereotype portrayals (firm side) in ads to consumer attitudes (consumer side). To gain a deeper understanding of consumer responses to the communicated stereotypes, we complement the multilevel analysis with in-depth insights generated from qualitative interviews.

To answer the first research question in Table 2, we used a content analytic methodology which provides “a scientific, quantitative, and generalizable description of communications content” (Kassarjian, 1977, p. 10). In doing so, we followed Gaur and Kumar’s (2018) six-step procedure, which involves (1) selecting a database, (2) setting selection criteria, (3) developing a valid coding instrument, (4) coding the sample, (5) assessing coding accuracy, and (6) summarizing and interpreting the coded text.

We first created a pool of print advertisements of brands in five distinct product categories (soft drinks, furniture, sweets, online platforms, beer). We intentionally did not limit ourselves to a single product category but rather opted for a variety of product categories for generalizability purposes. Within each product category, we collected several print advertisements about a global and a local brand, as previous research indicates that consumers’ stereotypical perceptions of global and local brands may differ (see Davvetas and Halkias, 2019; Kolbl et al., 2019). The identification of the advertisements was rooted in Laroche et al. (2011), starting with the most recent campaigns and then going back in time.

In the second step, we applied two criteria to select the final set of ads for subsequent analysis. First, each advertisement had to depict at least one brand-related stereotype. Second, the advertisements were selected based on their temporal recency, with priority given to the most current ones available. The final database consisted of 10 different print ads per global brand and 10 ads per local brand in each product category, resulting in a sample of 100 advertisements for the 10 brands in the five aforementioned product categories (a further 20 ads were collected to serve as a pretest sample for our coding scheme). Table 3 lists the brands used in the main content analysis.

We then proceeded with the development of a theoretically-valid coding instrument. In line with the SCM, we identified items depicting warmth and competence in brand stereotyping research (Fiske et al., 2002; Kervyn et al., 2012; Halkias and Diamantopoulos, 2020). The pool of items consisted of 21 items for warmth and 14 for competence. Warmth included items such as friendly, good-natured, kind and warm; and competence items such as capable, competent, efficient, and intelligent. We refer to these items as the entity traits (an entity being a brand, a brand’s origin, and a brand’s buyer/user). In doing so, we follow Ford et al.’s (1998) approach, who used gender traits as one of the dimensions in their coding scheme. The coding of the ads was undertaken independently by two trained coders who first coded the pretest sample (N = 20) and then the main sample.

Consistent with established practices (e.g. Avery and Ferraro, 2000; Naderer et al., 2017), the content of the advertisements that the two coders identified was discussed with an academic expert in the area of investigation and, based on a common consensus, divided into three different categories. These categories, which serve as executional elements of the content were: slogan, pictorial information, and verbal information. The coders examined each individual ad and identified the relevant stereotype(s) (i.e. brand, brand origin, brand buyer/user) as well as the stereotype dimension(s) (i.e. warmth and/or competence) depicted in the ad (see  Appendix 1 for examples). Consistent with prior research on the SCM (Fiske et al., 2002; Kervyn et al., 2012), coders approached the brand-related stereotypes in the ads from a third-person perspective, reflecting a common view of society rather than the coder’s own perceptions of stereotypes. Ads were coded on the basis: “most people in Austria view [brand]/[brand origin]/[brand buyers/users] as: warm/competent.” We assessed coding accuracy by evaluating intercoder reliability. Cohen’s kappa produced a value of 0.98, which suggests a very high degree of agreement between the two coders (McHugh, 2012). The final step involved summarizing and interpreting the results of the content analysis, as discussed in the next section.

All three types of brand-related stereotypes (namely, brand-, brand origin- and brand buyer/user stereotypes) were found to be portrayed – albeit to a different extent – in the analyzed ads. This finding aligns with prior research from a consumer perspective, documenting the existence of these distinct brand-related stereotypes (e.g. Antonetti and Maklan, 2016; Kervyn et al., 2012; Maheswaran, 1994) as well as their coexistence (Gidaković et al., 2021).

Brand stereotypes were portrayed most often (in 91 out of 100 ads), followed by brand buyer/user stereotypes (in 43 out of 100 ads) and brand origin stereotypes (portrayed in 22 out of 100 ads). While approximately half the ads depicted only a single stereotype, the rest portrayed a combination of stereotypes (Figure 1).

Brand stereotypes on their own were most frequently portrayed (45/100), followed by combinations of brand- and brand buyer/user stereotypes (30/100); brand origin stereotypes in combination with brand stereotypes (9/100); a combination of all three stereotypes (7/100); brand origin stereotypes in co-existence with brand buyer/user stereotypes (4/100); brand buyer/user stereotypes on their own (2/100); and, brand origin stereotypes on their own (2/100).

To formally assess how the stereotypical dimensions of warmth and competence are deployed in print advertisements, we created a contingency table for each brand-related stereotype and compared the frequencies of the warmth/competence cues being simultaneously present (absent) in the relevant print advertisements.

Focusing on brand stereotypes (Table 4), there is a significant relationship between warmth and competence (x² = 50.07, df = 1; p < 0.001). The odds ratio of 0.0197 demonstrates that there is a negative association between the occurrence of brand stereotype warmth and competence: when competence is present, the odds of warmth being present are much lower than when competence is not present.

For brand buyer/user stereotypes (Table 5), the relevant contingency table is also associated with a significant result (x² = 4.42, df = 1, p < 0.05), with the odds ratio (0.14) indicating that when warmth is present, the odds of competence being present are much lower than when warmth is not present.

As far as brand origin stereotypes are concerned (Table 6), the contingency table does not show a significant relationship between warmth and competence (x² = 0.05, df = 1, p = 0.823). While the odds ratio of 1.3 suggests a 30% increase in the odds of warmth being displayed when competence is displayed compared to when competence is not displayed, the result is not statistically significant.

To summarize, in practice, firms seem to deploy all three brand-related stereotypes, but to a different extent and, importantly, in different combinations. Brand stereotypes on their own are the most widely used by firms in their communications, with a particular emphasis on the competence dimension. Brand buyer/user stereotypes follow in terms of popularity, mostly emphasizing the warmth dimension. Brand origin stereotypes are not widely deployed and, if they are, it is the competence dimension that gets highlighted. While there seems to be a negative association between the occurrence of the two stereotype content dimensions in the case of brand- and brand buyer/user stereotypes (i.e. when warmth is emphasized, competence is not – and vice versa), no such association is evident in the case of brand origin.

In Phase 2, we used multilevel modeling to assess how the content of brand-related stereotypes identified in Phase 1 influences consumer attitudes toward the focal brands. We complemented this analysis with a series of qualitative, open-ended interviews to gain in-depth insights into consumers’ understanding of stereotypes and (dis)confirm the findings of the content and multilevel analyses (Plakoyiannaki and Budhwar, 2021). Consistent with relevant literature (e.g. Fletcher et al., 2018; Saunders and Townsend, 2016), we opted for maximum variation sampling to collect data that would capture a wide range of consumer perceptions of, and responses to brand-related stereotypes. We recruited participants with various demographic backgrounds (i.e. age, gender and occupation) and generated a multilevel dyadic data set with a total of 84 consumers (Level 1) grouped by 10 brands (Level 2), with 8 to 10 consumers allocated to each of the 10 brands analyzed in Phase 1 (see  Appendix 2). In assigning brands to consumers, we made sure that respondents were familiar with the brand in question.

Respondents were first requested to fill in a short questionnaire and were subsequently interviewed face-to-face. Each encounter (filling in the survey and interview) took place in the presence of two trained research assistants, with one person interviewing the respondent and the other noting down other important observations (e.g. respondents’ nonverbal communication). The purpose of the questionnaire was to generate Level 1 data for the multilevel analysis and complement the Level 2 data obtained during Phase 1 (the latter comprising six (count) variables documenting the incidence of the two stereotype dimensions in each of the three brand-related stereotypes across the ten stimuli brands). The questionnaire included questions on attitudes toward the assigned brand (measured on the adapted Fuchs and Diamantopoulos’s (2010) scale; α = 0.87); perceived brand globalness and localness (measured, respectively, on Steenkamp et al.’s (2003) and Swoboda et al.’s (2012) scales; α_PBG = 0.80, α_PBL = 0.71); brand familiarity (captured by the item “How familiar would you say you are with [BRAND]?”, anchored at 1 = not at all familiar to 7 = very familiar) as well as demographic information (gender, age, education, etc.).

Following the completion of the questionnaire, study participants were interviewed according to Kvale’s (1996) seven stages: thematizing, designing, interviewing, transcribing, analyzing, verifying and reporting. We collected data through semistructured interviews following a conversational approach (Alvesson and Sköldberg, 2000) and enhanced the trustworthiness of our qualitative findings through the use of theory to develop the interview guide and the pretesting of the latter (Lincoln and Guba, 1985). At the beginning of each interview, the purpose of the study was explained, informed consent was obtained, and information was communicated to the participants regarding anonymity, confidentiality, recording and estimated duration of the interview. This was followed by “warm-up” questions to break the ice and establish a rapport between the researcher and participants. We continued the interview by exploring whether respondents notice brand-related stereotypes in the communication messages of companies and how they understand them.

In line with the theoretical underpinnings of the SCM (Fiske et al., 2002), we used the third-person technique in that we invited participants to elucidate how most people in their society perceive brand-, brand origin and brand buyer/user stereotypes instead of only elaborating on their personal stereotypical judgments. For example, participants were asked: “What do Austrians, in general, associate with France?”, or “What would most Austrians associate with brands coming from Italy?”, or “How would Austrians describe typical buyers of brand X?”. By providing the opportunity to respondents to talk about someone else, they were able to elaborate freely about stereotypical judgments that they might not necessarily admit holding themselves. This technique enabled us to elicit deep-seated stereotypical perceptions of participants, while minimizing social desirability bias. All interviews were transcribed and imported to Atlas.ti software following the 24-hour rule of Eisenhardt and Bourgeois (1988) to minimize recollection effects and, thus, capture as many details as possible of each interview encounter.

We used thematic analysis to analyze the interview transcripts, relying on the SCM for conceptual guidance and iterating between theory and data. The transcripts were read several times, after which we began to identify themes that emerged from our respondents’ narratives. Based on Corley and Gioia (2004) and Gioia et al. (2012), we started by identifying 1st order codes directly from the interviews. We then merged these codes into 2nd order theoretical level themes. Finally, further integration led to the aggregation into the three types of brand-related stereotypes. For example, 1st order codes included themes associated with brand attributes, brand origin and brand buyer/user characteristics descriptions. These themes were subsequently categorized into 2nd order codes around the themes of warmth and competence. Finally, brand stereotypes, brand origin stereotypes and brand buyer/user stereotypes formed the aggregate categories of our coding process. Investigator triangulation (Denzin, 1978), that is, the involvement of multiple researchers in the analysis of the interview transcripts, was used to enhance the trustworthiness of the qualitative findings. In doing so, we cross-checked our interpretations to improve our data structure and crystallize our analysis.

We conducted a multilevel analysis combining the data from Phase 1 (firm perspective) with the survey data from Phase 2 (consumer perspective). We estimated the model shown in Figure 2 to test the (cross-level) effect of the warmth and competence dimensions on consumer attitudes toward the focal brand. In line with recent evidence indicating interrelationships among each stereotypical dimension across different brand-related stereotypes (see Gidaković et al., 2021), we considered the effects of brand warmth, brand buyer/user warmth and brand origin warmth simultaneously (and did the same for competence). The resulting multilevel model assessing the impact of warmth (competence) contains three (count) variables capturing warmth (competence) at Level 2, with each variable capturing the frequency with which warmth (competence) was portrayed in the 10 ads of a particular brand (min = 0, max = 10).

Regarding Level 1, in addition to consumer attitude toward the brand (the dependent variable), we specify consumers’ perceptions of the focal brands’ globalness (PBG) and localness (PBL) and brand familiarity as predictors; age and gender are also included as control variables. The reason for including PBG and PBL as Level 1-predictors in the model is two-fold. First, PBG and PBL have repeatedly been shown to influence several consumer outcomes, including brand attitudes. As Kolbl et al. (2020) observe, “in addition to impacting brand quality and prestige (Steenkamp et al., 2003), PBG and PBL boost a brand’s identity expressiveness (Xie et al., 2015) and serve as strong attitudinal (Halkias et al., 2016) and brand identification drivers (Sichtmann et al., 2019).” Second, prior (consumer-based) research shows that PBG and PBL influence consumers’ stereotypical perceptions of brands in terms of brand warmth and competence (Davvetas and Halkias, 2019; Kolbl et al., 2019). By explicitly including PBG and PBL as drivers of consumer attitudes toward the brand at Level 1, we guard against overestimating the influence of the communicated stereotype dimensions in Level 2. Brand familiarity is also included as a Level 1 predictor since more favorable attitudes are likely to be held by consumers when faced with familiar brands. This is because “familiarity overcomes “fear of the unknown”; individuals are likely to be more circumspect/negative in their judgment of objects with which they are not familiar” (Diamantopoulos et al., 2011, p. 511).

Hierarchical linear modeling, with HLM v.8.0 software, was used to test the relevant cross-level effects (Castro, 2002; Hofmann and Gavin, 1998; Hox et al., 2017). Level 2 stereotype dimensions, PBG, and PBL, were grand mean centered; brand familiarity was group mean centered; and controls (age and gender) were uncentered. Pseudo R² coefficients were calculated according to Bosker and Snijders (2011). Table 7 summarizes the specifications and equations of the various models tested.

Our data set is of sufficient size for the purposes of multilevel analysis (N_Level1 = 84, N_Level2 = 10), in light of evidence that “estimates of the regression coefficients are unbiased, even in if the sample is as small as 10 groups of five units” (Maas and Hox, 2005, p. 91). Relevant psychometric information and descriptive statistics for all Level 1 and Level 2 variables in our model can be found in Table 8. Note that care should be exercised in interpreting the correlations at Level 2 in Table 8 since the relevant constructs capture the incidence/frequency with which warmth or competence is highlighted in each brand-related stereotype and not the strength/magnitude of the dimensions. For example, the −0.72 correlation between brand stereotype warmth and competence indicates that the more often warmth is depicted in the ads of a brand, the less often competence is depicted (and vice versa).

Estimation results are presented in Table 9. We first tested the intercept-only model [Table 6, equation (1)]. With an overall brand attitudes mean of 5.33 that differs significantly from 0 (Hox et al., 2017), the interclass correlation coefficient came to 0.08, indicating that 8% of the total variance in attitudes is explained at the brand level (Level 2).

We then assessed a baseline specification, that is, the regression-based model in a multilevel context [Table 7, equation (2)]. Consistent with prior literature (e.g. Batra et al., 2000; Davvetas et al., 2015), all three Level 1 predictors are positively and significantly related to brand attitudes; perceived brand localness (γ = 0.33, p < 0.001), brand familiarity (γ = 0.26, p < 0.001) and perceived brand globalness (γ = 0.13, p < 0.05). The baseline model explains 15% of the variance in brand attitudes.

Next, we examined the direct cross-level effects of the warmth dimension of all three stereotypes on brand attitude as captured by the cross-level model [Table 6, equation (3)]. The results show that brand warmth positively influences brand attitudes (γ = 0.14, p <0.05) and brand buyer/user warmth is negatively related to brand attitudes (γ = −0.10, p < 0.05), over and above the influences of the Level 1 antecedents. This model explains 29% of the variance in brand attitudes, which almost doubles the explanatory power in comparison to the baseline model.

We then repeated the analysis by testing the cross-level effect of the competence dimension of all three stereotypes on brand attitudes [Table 6, equation (3)]. Brand competence is positively related to brand attitudes (γ = 0.20, p < 0.05), as is brand origin competence (γ = 0.10, p < 0.001), while brand buyer/user competence is negatively related to brand attitudes (γ = −0.39, p < 0.05); again, over and above the influences of the Level 1 antecedents. The explanatory power (28%) of the model is comparable to the warmth-based model. None of the control variables were significant in any of the models tested.

In summary, the multilevel analysis reveals that both warmth and competence have notable diagnosticity in shaping consumers’ brand attitudes. Moreover, the two dimensions demonstrated comparable levels of explanatory power in accounting for variance in brand attitudes.

This part of Phase 2 explored further how consumer perceptions of brand-related stereotypes shape attitudes toward the brand. All interview participants were able to spontaneously recall the three brand-related stereotypes (brand stereotypes, brand origin stereotypes and brand buyer/user stereotypes) which were identified in Phase 1. For each stereotype, respondents were also able to make warmth and competence associations, which is consistent with prior stereotyping literature (Fiske et al., 2002). Participants evoked brand-related stereotypes to make sense of a brand both separately and in concert, meaning that such stereotypes are closely interconnected in consumers’ brand narratives. For instance, reflecting on the Austrian confectionary brand of Manner, a participant combined brand- and brand origin stereotypes:

Another respondent reflected on Coca-Cola drawing on a combination of brand buyer/user stereotypes and brand origin stereotypes:

In short, brand-related stereotypes not only co-occur in firms’ advertisements, as per Phase 1, but also in consumers’ brand understanding, as our interviews show. We now turn to the discussion of the individual brand-related stereotypes, their dimensions, and their influence on consumer brand attitudes.

Participants most often referred to brand stereotypes when elaborating on their assigned brand. Brand stereotypes were mentioned both in isolation but also in conjunction with other brand-related stereotypes. This pattern is consistent with the findings from the content analysis in Phase 1, where we show that companies rely heavily on portraying brand stereotypes in their communication activities.

The qualitative interviews revealed that participants used both dimensions of stereotype content, namely, warmth and competence, in order to make sense of a brand, with the competence dimension being the most salient one in their narratives:

While brand stereotypes were primarily linked to competence cues by most of the respondents, when associated with warmth, nostalgic cues were often brought to the fore. For instance, one participant noted:

Consistent with the literature on stereotyping (Fiske et al., 2002), the content of stereotypes can be positively or negatively valenced. Some participants highlighted negative aspects (“[Heineken] is a miserable beer […]. I know many beer drinkers who say that Heineken is no beer at all.”), others emphasized positive aspects (“I would definitely say that it [xxxLutz] is of high quality.”), and still others coupled negative (i.e. warmth) with positive aspects (i.e. competence) (“It [Coca-Cola] is unhealthy, but compared to other colas, it is a very high-quality brand.”), when making sense of the focal brand. In line with the multilevel analysis results, respondents elaborated on the relationship between brand stereotypes and brand attitude. For instance:

When elaborating on the origin of a brand, participants often drew on product category distinctions. For example:

The emphasis on the competence dimension when reflecting on the brand origin is aligned with the content analysis findings from Phase 1, showing that brand origin is most often expressed and depicted through competence. It is also consistent with the multilevel analysis, suggesting that brand origin competence is positively related to brand attitudes.

While participants most often relied on the competence dimension when considering a brand’s origin, the warmth dimension of the origin stereotype also featured in participants’ narratives as shaping positive brand attitudes:

The perceived relevance of brand origin warmth uncovered by the qualitative interviews should be seen in the context of the content analysis results in Phase 1. According to the latter, firms do not frequently highlight brand origin warmth in their marketing communications, relying instead on the competence dimension; neither did brand origin warmth significantly impact consumer attitudes in the multilevel analysis. The diagnostic value of the warmth dimension of the brand origin stereotype seems to lag behind the competence dimension, as also suggested by prior research from the consumer perspective (e.g. Diamantopoulos et al., 2017; Halkias et al., 2016).

In contrast to brand and brand origin stereotypes, participants rarely referred to brand buyer/user stereotypes when reflecting on a particular brand. This is surprising for two reasons. First, given that stereotyping is particularly prevalent when judging human groups (Fiske et al., 2002), one would expect that respondents would find it easy/natural to reflect on brand buyer/user stereotypes. Second, as the content analysis results in Phase 1 revealed, stereotypical portrayals of brand buyers/users are frequently used in firms’ marketing communications (see Figure 1). Our qualitative data suggest otherwise, with respondents either refraining from stereotyping buyers/users altogether or providing only broad/general descriptions:

When invoked by participants, brand buyers/users were often perceived as having a negative influence on brand attitude, which is consistent with our findings from the multilevel analysis. Participants often connected brand buyer/user stereotypes with avoidance groups (instead of reference groups) that consume a certain brand. For example:

In summary, the qualitative results paint a broadly consistent picture with the findings from the content analysis in Phase 1 and the multilevel analysis in Phase 2. Interview participants were able to spontaneously identify distinct brand-related stereotypes and distinguish between the warmth and competence dimensions of stereotype content. Their perceptions of the influence of these dimensions on brand attitudes supported the results of the multilevel analysis. The only divergence emerging from the interviews involved a greater perceived relevance of brand origin warmth and a lesser respondent recall of and attention to brand buyer/user stereotypes. A summary picture of the findings relating to both phases of our study can be found in the mixed-methods flowchart in Figure 3.

While a growing number of studies investigate consumers’ perceptions of brand-related stereotypes, extant research has yet to investigate the deployment of such stereotypes in the communication activities of firms and its impact on consumer attitudes toward the brand. To the best of our knowledge, our study is the first to address these research gaps by drawing on data from both firms and consumers and applying complementary research methods (content analysis, multilevel modeling and qualitative interviews). We thus respond to Septianto et al.’s (2022) call for an investigation of consumer responses to communicated brand-related stereotypes in real brand contexts. Moreover, we address calls for methodological pluralism (e.g. Christofi et al., 2024) by using a mixed methods research design to tap into both consumer and firm perspectives. Our findings (summarized in Figure 3) extend extant knowledge on brand-related stereotypes and have several theoretical and managerial implications as discussed below.

While prior research (e.g. Antonetti and Maklan, 2016; Diamantopoulos et al., 2021; Gidaković et al., 2021) implicitly assumes that consumers are exposed to multiple brand-related stereotypes, our study offers empirical evidence that this is indeed the case. Firms convey stereotypical information about the brand, its origin and the typical buyer/user and very often do so by depicting multiple brand-related stereotypes in their advertisements. Consumers also spontaneously evoke different brand-related stereotypes either independently or in conjunction, as the qualitative interviews revealed. This leads to a consistent picture from a supply-side (i.e. firm) and a demand-side (i.e. consumer) perspective, suggesting that research focusing on any one brand-related stereotype will inevitably provide only a partial (and thus possibly biased) explanation of how consumers understand and respond to stereotypical information relating to brands.

Regarding the relative prevalence of different brand-related stereotypes, both the content analysis results and the insights from the qualitative interviews clearly point to a predominance of brand stereotypes. The latter are the stereotypes most often deployed in firms’ marketing communications as well as most frequently evoked by consumers. Despite firms’ observed emphasis in communicating brand buyer/user stereotypes, rather than brand origin stereotypes, consumers tend to spontaneously refer to the latter much more often than to the former. This suggests that firms’ communications of stereotypical information might not be equally effective in reaching consumers but depend on the specific stereotype involved; we will revisit this issue in the Managerial implications section.

Turning attention to the content of brand-related stereotypes, both the warmth and the competence dimension influence consumer attitudes toward the brand, however, the nature of this influence varies depending upon the communicated brand stereotype. Whereas brand competence and brand warmth positively impact brand attitudes, the same dimensions have a negative effect when associated with brand buyer/user stereotypes. This implies that the impact of stereotypes content on consumer responses is not invariant and serves to emphasize the importance of considering multiple brand-related stereotypes, as noted earlier.

The picture painted above offers important insights on the diagnosticity of warmth and competence for consumer behavior. The debate on the relative importance of warmth vs competence has shifted from asking which dimension has a greater predictive power (e.g. Chen et al., 2014) to a more fine-grained and context-specific approach (e.g. Gidaković et al., 2021; Kolbl et al., 2020). Warmth is particularly relevant when it comes to consumer-brand identification and similar relational outcomes (Güntürkün et al., 2020; Kolbl et al., 2019; Stokburger-Sauer et al., 2012) as well as consumers’ functional and emotional value perceptions (Kolbl et al., 2020). Our findings on the positive effects of brand warmth on brand attitudes are aligned with this stream of research, while the observed absence of any impact of brand origin warmth is consistent with prior studies on the role of brand origin stereotypes (e.g. Diamantopoulos et al., 2017; Halkias et al., 2016). As far as brand buyer/user warmth is concerned, our findings resonate with Antonetti and Maklan (2016, p. 796), who conclude that warmth is not an attractive feature in a consumption context because “warm groups are not envied and envy plays a central role in fueling a desire to emulate a consumption group.”

The diagnosticity of the competence dimension has been repeatedly shown in prior marketing studies (e.g. Aaker et al., 2012; Gidaković et al., 2021; Grandey et al., 2005; Halkias et al., 2016; Kirmani et al., 2017; Kolbl et al., 2020; Marinova et al., 2018). Our findings add to the rich body of research by showing that brand-, as well as brand origin competence, positively influence brand attitudes. At the same time, our findings that not only brand buyer/user warmth, but also competence has a negative effect on brand attitudes, imply that irrespective of stereotype content, focusing on the brand buyer/user in marketing communications is likely to be risky. A possible explanation for this counter-intuitive finding is that consumers do not like firms to “impose” who their typical brand buyers/users are. Stereotypical perceptions of brand buyers/users are perhaps spontaneously formed through social interactions during brand use. Possibly this is also a reason why brand buyer/user stereotypes were not frequently evoked and elaborated during the qualitative interviews (i.e. consumers simply did not attend to communicated information about typical brand buyers/users). Having said all that, given that brand buyer/user stereotypes represent a rather under-researched area, these findings should be treated with caution and subjected to further research.

From a managerial perspective, our study offers insights regarding which brand-related stereotypes and stereotypical dimensions to use in marketing communications to generate favorable consumer responses. Our findings show that communication of stereotypical information by firms has the potential of influencing consumer attitudes toward the brand; however, care should be exercised regarding when deciding on which brand-related stereotype(s) to focus on and which stereotypical dimension to emphasize.

Regarding the former issue, portraying brand buyer/user stereotypes in marketing communications can be risky for two reasons. First, as our qualitative interviews revealed, brand buyer/user stereotypes are rarely invoked in the investigated consumers’ brand narratives. Despite firms’ frequent depiction of typical brand buyers/users in their ads (see content analysis results in Figure 1), respondents typically refrained from relying on such cues, focusing instead on brand- and origin-related cues. Second, and most importantly, both dimensions of brand buyer/user stereotypes were found to have an adverse effect on brand attitudes (see multilevel results in Table 9).

In light of the above, firms would be well advised to focus their communication efforts on portraying a combination of brand- and brand origin stereotypes and place particular emphasis on the competence dimension. Both brand competence and brand origin competence featured frequently in consumers’ brand narratives, and both were found to positively influence brand attitudes. Furthermore, highlighting brand warmth should further enhance the beneficial effect of brand competence, given that its influence on attitudes is also positive. While firms seem to trade-off between the competence and warmth dimension in their communication of brand stereotypes (i.e. when competence is emphasized, warmth is not and vice-versa), it seems advisable to deploy both dimensions in conjunction. Doing so is likely to result in a stronger overall effect on brand attitudes, given that both dimensions have distinct positive effects on the latter. Prior research in an advertising context shows that reliance on solely one stereotype dimension results in an innuendo effect (Kervyn et al., 2012), meaning that addressing only one dimension and leaving another out can lead to a negative evaluation of the omitted dimension, compromising advertising effectiveness (Peter and Ponzi, 2018).

Given that our investigation is the first to simultaneously approach brand-related stereotypes through multiple research lenses (content analysis, multilevel modeling and qualitative interviews), its findings should be considered as being suggestive rather than conclusive and subject to further research. There is a need for replication not only in additional/different product categories but also in different countries, given that stereotypes portray a consensus in a certain society and are culturally shared (Fiske et al., 2002).

Our investigation focused explicitly on brand-related stereotypes only. However, there are other types of stereotypes, which are not necessarily brand related, but may still play an important role in a consumer context. For example, prior research on gender stereotypes finds that “there is an association between product categories and female role stereotypes” (Plakoyiannaki and Zotos, 2009, p. 1411) and that gender and advertising practices reinforce brand stereotypes with “femvertising” trying to challenge traditional female stereotypes (Gomez-Borquez et al., 2024). When it comes to luxury (vs. nonluxury) brands, they rely more on stereotyped gender expressions and sexualization in their ads (Michaelidou et al., 2022). Brand stereotypes moderate the effectiveness of influencer endorsements, where competent/warm brands benefit more from expert influencers/equally with entertainers or informers (Ren et al., 2023). Future research could shed light on such additional types of stereotypes portrayed in marketing communications in different media (Mähnert et al., 2024).

Whether brand-related stereotypes and their dimensions portrayed in print advertisements are also similarly portrayed in other communication channels (e.g. TV ads and radio ads) and digital media (e.g. social media ads and content) is also open to investigation. It would be informative to explore how the content of stereotypes, as reflected in warmth and competence, varies across different types of communication elements and types of media. Could it be that pictorial elements are more likely to depict warmth, whereas the verbal elements would be more prone to portray competence?

Finally, the current study only focused on consumer attitudes toward the brand as the outcome variable influenced by the communication of stereotypical information. Attention to other consumer responses (e.g. brand ownership, loyalty and willingness-to-pay) would generate further insights into how brand-related stereotypes impact consumer behavior.

The authors thank Austria’s Agency for Education and Internationalization (OeAD; Project No. SI 11/2023) and the Slovenian Research Agency (ARRS; Project No. BI-AT/23-24-003; University of Ljubljana, School of Economics and Business) for supporting this research.

All authors have contributed equally to the manuscript.

Table A1 

Table A2