Introduction

During the last decade, researchers in the field of consumer–brand relationships have increasingly placed their interest in studying brand love (Bairrada et al. 2018; Batra et al. 2012; Luo and Hamlin 2022), and subsequently brand hate (Aziz and Rahman 2022; Brandão and Popoli 2022; Fetscherin 2019). In this study, the concept of brand refers to “a trade name/logo that identifies a product or firm, usage of which may be limited by legal structures and practice” (Avis and Henderson 2022, p. 366). In addition to physical products, a brand can be built on a corporate/organisation (Walker 2010), or service (Berry 2000), a place (Kaplan et al. 2010) or a person (Fournier and Eckhardt 2019). Consumer–brand relationships refer to the positive or negative bond between a consumer and a brand that may be fostered, weakened or even broken through different processes (see Alvarez et al. 2021). Brand love is a strong positive relationship, connecting a consumer with a brand through “multiple interrelated cognitive, affective and behavioural elements, rather than a specific, single, transient love emotion” (Batra et al. 2012, p. 6). Brand love offers many benefits for the loved brand, such as positive word of mouth (WOM), loyalty and resistance to negative experiences (Aro et al. 2018). By brand hate, in turn, we refer to a strong negative consumer–brand relationship (e.g. Kucuk 2019; Brandão and Popoli 2022). The literature on brand hate reveals that it is a strong brand relationship that may become visible, for example, as brand retaliation (Hegner et al. 2017) or even as the willingness to make financial sacrifices to hurt the brand (Fetscherin 2019). Thus, as behaviours linked to brand love may be beneficial and behaviours linked to brand hate may be detrimental, it is important to understand the mechanisms of brand love and hate, both for academics and practitioners. Both concepts represent extreme brand relationships and can often be viewed as opposite ends of the brand relationship scale (see Fetscherin 2019).

Although brand love and brand hate have been studied extensively as separate brand relationships, there is an apparent lack of studies that elaborate on these powerful brand relationships together in order to understand the interrelatedness of these two concepts (see Zarantonello et al. 2018). The few recent studies that combine brand love and brand hate have mainly focused on describing when or why a consumer’s brand relationship transforms (or does not) from brand love to brand hate (Kashif et al. 2021; Yang and Mundel 2022; Zhang et al. 2021). Thus, there is a lack of studies focusing on the interrelatedness of these two brand relationships. Existing brand love studies have mainly been conducted in the context of product (Hemonnet-Goujot and Valette-Florence 2022) and service (Ahuvia et al. 2022; Farmaki et al. 2021) brands, which also holds true regarding brand hate studies (Brandão and Popoli 2022). Although a few studies have focused on the effects that a celebrity’s personal brand has on the love felt towards the promoted brand (Zhang et al. 2020; Zhou et al. 2020), no studies have focused on brand love and hate relationships between the consumer and a personal brand. Further, even though consumers wish their brands to publicly take a stand in socio-political issues (Vredenburg et al. 2020), there is only one brand love (Banerjee and Chaudhuri 2022) and one brand hate (Banerjee and Goel 2020) study conducted in the political context. Thus, brand love and brand hate are rather neglected areas in personal brands and political contexts.

In the current political and media climate, there is a strong tendency towards polarisation in the public’s reactions. While familiarising ourselves with the data and existing literature, we observed that this study is also related to the concept of brand polarisation. Brand polarisation is conceptualised as “an affective phenomenon where beliefs and emotions of a significant number of people induce a simultaneous move to the extremes involving passionate positive and negative feelings and convictions towards the brand, like-minded consumers and opposite-minded consumers” (Osuna Ramirez et al. 2019, p. 620). Brand polarisation means that instead of having only loving or hating consumers, a brand simultaneously has both lovers and haters (Luo et al. 2013). Indeed, “Although nearly all of the existing research indicates that consumers form only, or primarily, a positive or negative relationship with a specific brand, the reality is somewhat different. Many brands simultaneously have a significant group of lovers and a substantial group of haters” (Osuna Ramírez et al. 2019, p. 615). In brand polarisation, feelings are extreme rather than moderate or neutral. Interestingly, brand polarisation can be considered an opportunity for a brand: consumers are not indifferent towards polarising brands, as they are well-established and evoke strong emotions (e.g. Osuna Ramírez et al. 2019). Another polarisation in the field of marketing is political polarisation (see Weber et al. 2021) where consumer’s own political identity guides their consumer behaviour in response to corporation’s/brand’s political choices (Weber et al. 2021).

To fill this gap, this study aims to increase the understanding of extreme brand relationships—namely, brand love and brand hate—and their conceptual interrelatedness. During her volunteer work with the social media account section of a personal political brand, the first author preliminarily surmised that expressions of brand love and brand hate may share commonalities. Thus, we began our study with a pre-supposition that brand love and hate might be conceptually interrelated and propose the following research question:

How are Brand Love and Brand Hate Conceptually Interrelated?

We considered a personal political brand—that is, a politician—to represent a particularly interesting type of personal brand (see also Armannsdottir et al. 2019). The politician’s brand image is likely to be amalgamated from issues such as how people feel about the politician herself and her ideology, the branding campaign’s success and how people feel about the politician’s party alliance, that is how people feel about the political organisation that personal brand represents. A personal brand may have both positive and negative reputation-related spill-over effects with organisations the personal brand presents (e.g. Rangarajan et al. 2017; Potgieter and Doubell 2020). A positive spill-over effect is employed when a celebrity brand’s reputation is used to increase the awareness of some organisation, brand or cause (see Özer et al. 2022), for example. A brand of a political party organisation may even cause challenges for building a brand of politician (Armannsdottir et al. 2019). A well-known personal political brand may both either enhance or hinder a country brand as well as organisation brands which are originated or related to the country that the politician presents (e.g. Fetscherin 2010). A term “political influencer” describes the impact a personal political brand may have in social media (Pérez-Curiel and Limón-Naharro 2019). However, it is also noted that celebrity endorsement is not effective without attachment between a consumer and a brand (Albert et al. 2017). Thus there should be a natural fit between a consumer and a brand to get benefits from celebrity—brand endorsement effects (Albert et al. 2017). Personal branding is becoming an increasingly relevant field of future branding research (see Santos et al. 2019).

Studies on personal political brands, or the brands of politicians, are fragmented, leaving this aspect of brand relationships unstudied. In the case of the brands of U.S. presidential candidates, it is noted that while the brand’s personality traits, such as honesty and intelligence, impact voter preference, these traits have less of an effect than candidates’ political orientation (Harrison et al. 2023). However, when parliamentary candidates lack the backing of political parties (as in Guernsey), personal political brand identities are built based on the candidates’ “personal characteristics, professional and personal experiences, legacy, and heritage” (Pich et al. 2020, p. 429). Further, it is noted that voter decisions in the U.S. may be impacted by anti-brand feelings; that is, the willingness to vote against a candidate (Newman 2007). As election systems vary between countries, it is important to study politicians’ brands in different cultures.

Thus, this study contributes to bringing personal brands and political brands into the academic discussion on brand love and brand hate. A personal political brand presents a good case study for investigating the conceptual interrelatedness of brand love and hate: it is a particularly unique type of personal brand (or human brand; see Osorio et al. 2020) that elicits emotions related to intellectual involvement. A politician brand differs from product brands, as it enables a brand’s core values to be expressed and brand identity to be fostered in a personal way.

Empirically, we study the online brand community of a new Finnish politician brand. The study employs abductive reasoning for theory building (Kovács and Spens 2005) and is qualitative in nature. Research iterating between deductive and inductive reasoning may be regarded as abductive (e.g. Eriksson and Kovalainen 2008). The primary data are gathered from the politician’s public Facebook page and complemented by an interview and participant observation. The evolution of the internet has enabled the ease of articulating consumer attitudes on brands and alerts us to the processes associated with brand love- and brand hate-related consumer expressions (Asmussen et al. 2013). Positive WOM is commonly connected to brand love relationships (e.g. Carrol and Ahuvia 2006) as negative WOM is to brand hate relationships (e.g. Zhang and Laroche 2020). We analyse consumers’ public online expressions of brand love and hate and the commonalities shared by these expressions, as well as their underlying themes. We consider electronic WOM (eWOM) to be a medium through which brand love and brand hate become visible in a manner that researchers can detect. In this study, voters represent consumers because they consume the ideas that the politician brand promotes (Black and Veloutsou 2017; Falkowski and Cwalina 2012). WOM is considered a powerful tool in marketing, as it has a great impact, whether positive or negative, on an individual's choices (Pagani and Mirabello 2012). Accordingly, it is important for brand managers to understand the circumstances in which people are willing to spread good or bad words about them.

Through our literature review (see Summers 2001), we noted that the concepts of brand love and brand hate share several mechanisms. In the current literature, these mechanisms are typically referred to as antecedents or consequences of brand love (e.g. Bairrada et al. 2018) and hate (e.g. Hegner et al. 2017). However, since we conceive of brand love and hate as brand relationships (see Alvarez et al. 2021)—that is, as dynamic phenomena—we consider these antecedents and consequences to be mechanisms that are part of brand love or hate, rather than phenomena that happen before (antecedent) or after (consequence) brand love and brand hate. Our study brings these mechanisms into focus. In this study, we focus on the mechanisms found in the literature on both concepts, namely brand love and hate, and also look whether novel mechanisms arise from the data. In addition, through this combined study of brand love and brand hate, we contribute to the scarce literature on brand polarisation (Luo et al. 2013). Thus, this study extends our understanding of the concepts of brand love, brand hate and brand polarisation. The study further contributes by capturing personal political brands in the investigation and discussion of brand love and brand hate.

The paper is structured as follows. First, theoretical elaboration is formed based on the literature on brand love and brand hate. Then, the methods and context of the study are described. Next, the findings are reported and discussed. Lastly, theoretical conclusions and managerial implications are presented, together with the limitations of the study and suggestions for potential future studies.

Theoretical Elaboration

Brand Love Relationship

People create loving relationships with non-human objects (Batra et al. 2012). In recent years, consumers’ love towards brands has gained increased research interest, starting with Carroll and Ahuvia’s (2006) study. Brand love can be conceptualised as a positively charged, strong and multidimensional relationship between a consumer and a brand, “which typically includes identification with a brand” (Aro et al. 2018, p. 73). Brand love offers many benefits for the brand, including loyalty (Rauschnabel and Ahuvia 2014), resistance to negative information about the brand (Batra et al. 2012) and willingness to pay a premium price (Albert and Merunka 2013) or invest money (Barraida et al. 2018). Further, anticipated separation distress is related to brand love when consumers think they will not be able to use the brands they love (Langner et al. 2016).

A brand’s self-expressiveness is strongly connected to brand love, thereby enabling an individual’s self-expression (Aro et al. 2023; Batra et al. 2012) and identification with the brand (Ahuvia et al. 2009), which refers to a consumer’s self-brand connection (Escalas and Bettman 2009). Further, brand love typically includes consumer’s connection to other users of the brand (Albert and Merunka 2013), the sense of community between these users (Bergkvist and Bech-Larsen 2010) and the social (interpersonal) relationships enabled by the brand (Ahuvia et al. 2022). Further, certain brand personalities can be more loveable than others; brands’ personality dimensions, sincerity and excitement are noted to enhance brand love and have a positive effect on positive WOM through brand love (Roy et al. 2016). In addition, identifiable brand gender may drive brand love (Machado et al. 2019).

Satisfaction with the brand as well as strong positive feelings and passion (Carroll and Ahuvia 2006), emotional connection with the brand (Batra et al. 2012), a feeling of closeness with the brand (Barraida et al. 2018) and engagement with the brand (Wallace et al. 2014) are all relevant parts of brand love. Brand love involves attachment to the brand and attitude valence—that is, a strong and positive attitude towards the brand (Batra et al. 2012). Positive brand experiences (Huang 2017) in line with positive service experiences (Long-Tolbert and Gammoh 2012) are connected to brand love. A brand’s good reputation and prestige (Ahuvia et al. 2009; Barraida et al. 2018), as well as brand trust (Banerjee and Chaudhuri 2022), are linked to brand love. Further, a brand lover typically views the brand’s values as being aligned with their own strongly held values (Batra et al. 2012).

Positive WOM is widely connected to brand love (e.g. Batra et al. 2012; Carroll and Ahuvia 2016; Kudeshia et al. 2016; Roy et al. 2016; Strandberg and Ek Styvén 2020), and loving consumers may even declare love towards a brand (Carroll and Ahuvia 2006). This WOM can happen either offline or online. Wallace et al. (2016) studied the connection between brand love and positive WOM in the Facebook context. The study noted that when individuals connect a brand to their inner self (how they see themselves), they are likely to love that brand and express positive WOM about it, for example, in Facebook comments, in addition to merely liking the brand’s page. By contrast, when individuals only connect the brand to their social self (how they present themselves to others), they may experience brand love and like a brand on Facebook, but they are less likely to spread good words about it any other way, for example, by commenting on or sharing the brand’s posts. Further, it is noted that a sense of brand community, in line with the self-brand connection, enhances brand love on the brand’s Facebook pages (Palazon et al. 2019). In addition to verbal WOM, a self-brand connection can be expressed through uploading photos—that is, visual WOM—on Facebook, and it is noted that uploading brand photos is related to brand love (Kaiser et al. 2020). As social media networks allow people to express their ideal self through brands that are not part of an individual’s material world (Schau and Gilly 2003), social media offers unique opportunities for self-expression (Wallace et al. 2014).

In addition, anthropomorphism enhances brand love (Huber et al. 2015). This refers to the fact that people are more likely to love brands in which they see human-like characteristics. Thus, “…anthropomorphism makes consumer–brand relationships more like interpersonal relationships…consumers love brands more when they see the brand not just as a person, but also as a person like themselves” (Rauschnabel and Ahuvia 2014, p. 388). Anthropomorphism is even more meaningful for brand love than a brand’s perceived quality (Rauschnabel and Ahuvia 2014).

Brand Hate Relationship

In recent years, scholars have started to focus on the negative aspects of brand relationships (e.g. Bryson and Atwal 2019; Dessart et al. 2020; Zarantonello et al. 2016). Indeed, brand hate can be considered an extremely negative brand relationship (Fetscherin and Heinrich 2015; Hegner et al. 2017) with multidimensional features (Fetscherin 2019; Rodrigues et al. 2020). Brand hate arises from a consumers’ negative past experience (Hegner et al. 2017), dissatisfaction (Bryson et al. 2013), ideological incompatibility (Banerjee and Goel 2020) and a violation of expectations (Zarantonello et al. 2016). Brand hate includes a strongly negative attitude towards the brand (Bryson et al. 2013). It is also characterised by strong negative emotions, such as anger, sadness and fear (Zhang and Laroche 2020). Moreover, negative emotional affect (Bryson et al. 2013), disgust, contempt and anger (Fetscherin 2019), a feeling of being betrayed (Grégoire et al. 2009) and passion directed against a brand (Osuna Ramírez et al. 2019) are incorporated into brand hate.

Consumers tend to avoid brands they feel represent an undesired form of self. Thus, taste systems related to incompatibility (Zarantonello et al. 2016) and symbolic incongruity between a consumer and a brand can enhance brand hate (Hegner et al. 2017). Consumers’ reactions to brand-related behaviours, both those of the brand and of other users of the brand, are linked to consumers’ attitudes towards the brand (Hegner et al. 2017). Negative stereotypes or images of people who use the brand that a consumer hates are connected to brand hate (Bryson et al. 2013), and these users might be regarded as rivals (Shuv-Ami et al. 2020). Brand hate includes a person being willing to identify with an outgroup of the hated brand or act as an antagonist of it (Osuna Ramírez et al. 2019). For such people, the brand represents a negative reference group (Hegner et al. 2017). Anti-brand communities provide a platform for consumers to fulfil their self-definitional needs when they oppose a specific brand (e.g. Popp et al. 2016). Further, members of these anti-brand communities maintain a relationship with the brand they feel negatively about instead of avoiding everything related to the brand (see Dessart et al. 2020; Rodrigues et al. 2020).

In addition, ideological incompatibility, for example, in terms of legal, social or moral corporate wrongdoing, is acknowledged as a strong driver of brand hate (Brandão and Popoli 2022; Zarantonello et al. 2016). Moreover, brand embarrassment may generate brand hate when the embarrassing brand harms a consumer’s desirable social self and the consumer does not want to identify with such a brand (Sarkar et al. 2020). In the context of a political party brand, the more interested an individual is in politics—that is, the field in which a brand is operating—the more likely they are to hate a certain party brand (Banerjee and Goel 2020). Additional drivers of brand hate include poor corporate social performance (Bryson et al. 2013) and negative rumours (Hashim and Kasana 2019).

A brand’s personality traits (see Osuna Ramírez et al. 2019) and country of origin (Bryson and Atwal 2019) drive brand hate when consumers hold a negative opinion of those brand characteristics. Contrary to the plausibility that individuals would shy away from a hated brand, negative engagement has been observed in brand hate relationships (Brandão and Popoli 2022). Thus, brand haters hold onto the brands they hate, retaining a connection with them. Brand haters commonly have informal negative discussions, either online or offline, about the brand in question (Hegner et al. 2017; Jain and Sharma 2019). Brand hate relationships include different forms of negative WOM, such as private and public complaining (Grégoire et al. 2009; Zarantonello et al. 2016) and protest (Bryson et al. 2013; Zhang and Laroche 2020). Brand hate can be more passive, cold and avoidance-like, or more active, hot and revengeful (Grégoire et al. 2009; Fetscherin 2019). Corporate wrongdoings are linked with higher levels of negative behaviours, determined as “attack” and “approach” strategies, for example, WOM and consumer complaining, while taste systems are connected with stronger “avoidance” strategies, such as patronage (Zarantonello et al. 2016). Brand hate may change over time (Zarantonello et al. 2018), as may brand love (Langner et al. 2016). In extreme cases of feeling betrayed, brand love is even noted to turn into brand hate (Grégoire et al. 2009).

Brand anthropomorphism is featured in expressions of brand hate; thus, consumers speak or write about and to the brand as if it were a human (Brandão and Popoli 2022). By contrast, the reverse anthropomorphism of a brand—seeing a brand as evil—is linked to brand hate (Kucuk 2020). Thus, if a company places its profits over its consumers’ welfare, it may lead to reverse brand anthropomorphism, in which consumers consider the brand to embody evil (Kucuk 2020).

Conceptual Interrelatedness of Brand Love and Brand Hate

In this section, we demonstrate how we identified the mechanisms shared by both brand love and brand hate by combining our initial insights from our data with current literature. Notable and expected is that the literature mainly presents opposing directions for these mechanisms, with brand love having a positive charge and brand hate having a negative charge. The shared mechanisms of brand love and brand hate are collated in Table 1, representing the framework of the study.

Table 1 Conceptual interrelatedness of strong brand relationships
Full size table

As illustrated in Table 1, for example WOM is included under both brand love and brand hate. However, in brand love, WOM is positive, whereas it is negative in brand hate. In the interpersonal context, love and hate include similar yet opposing emotional components (Sternberg 1986, 2005). Thus, we bring this notion to the brand relationship context and state that brand love and hate include similar yet (mainly) opposing mechanisms. However, we suggest that in the brand relationship context, these mechanisms are not limited to being emotional, but rather consist of multidimensional characters, as shown in Table 1.

In this study, we focus on the conceptual interrelatedness of brand love and brand hate. Based on the preceding discussion, we constructed a theoretical framework that shows that brand love and brand hate share similar mechanisms. Next, we chose one of these shared mechanisms, namely WOM, to study the interrelatedness of brand love and hate in a real-life context. We consider WOM to be a medium through which brand love and hate can occur in a form that is detectable by us as researchers. Thus, we consider WOM central to understanding brand love and hate.

Methodology

Research Design

The current study seeks to highlight the conceptual interrelatedness of brand love and brand hate and empirically ground the interrelatedness. We consider consumers’ actions as expressive behaviour that is based on images, dreams, myths and thoughts that are stored in their minds (O'Shaughnessy 2010). This study’s starting point was a combination of empirical and literature-based notions. In empirical terms during her work with a personal brand’s social media account, the first author noted that some postings gathered simultaneously both extremely positive and extremely negative responses. Thus, a singular posting on a brand's social media may spark loving responses from some individuals and simultaneously elicit hateful responses from others. However, she knew that the literature lacks studies combining these two extreme brand relationships. Throughout the analysis and writing process, the authors alternated between their empirical data and theoretical knowledge to form conclusions (see Kovács and Spens 2005).

This is a single-case study (Flyjvbjerg 2006) employing qualitative research methods. A case study was adopted to build a theory in the context of complex, real-life topics, allowing an in-depth exploration (Dubois and Gadde 2002; Flyjvbjerg 2006). A qualitative investigation was considered suitable for studying the contemporary social phenomenon within its real-life context (Grix 2002). Thus, the study does not aim at statistical generalisability; instead, it aims at expanding and generalising theories (e.g. Eisenhart 2009). The authors ensured that the data and method triangulation took place (Patton 1999; 2015) by collecting three types of primary data (1) Facebook data, (2) a personal interview and (3) participant observation data from the online brand community. Triangulation aids in the verification and validation of qualitative analysis. In this investigation, we examined the coherence of outcomes stemming from diverse data collection methods, a practice known as method triangulation. Furthermore, data triangulation took place as we examined the consistency of different data sources (e.g. Patton 1999). The Facebook data guided the analysis, whereas other types of data served as complementary data.

Facebook was the social media platform used in this study because it was the dominant global social media platform (see Machado et al. 2019). In addition, Facebook was linked to both brand love (Aro et al. 2023; Kaiser et al. 2020) and brand hate (Brandão and Popoli 2022) in earlier studies, and is therefore likely to represent a fruitful context for the data gathering in this study. Further, in the country where the study was conducted (Finland), Facebook was the most widely used public social media conversation platform during the data gathering period (Meltwater 2019) and the major singular platform for hate speech (Knuutila et al. 2019).

Study Case

The data were gathered from a public Finnish online brand community of a female politician brand. This is a purposely chosen and information-rich example representing an extreme case of an interactive brand community (Flyjvbjerg 2006; Patton 2015). We followed eWOM occurring in this online brand community, which acted as a forum for brand supporters’ and antagonists’ interactions through which it was possible to explore expressions of brand love and brand hate. In this context, consumers are represented by the electorate in the relevant constituency because they consume the ideas that the personal political brand promotes (Black and Veloutsou 2017; Falkowski and Cwalina 2012). The scrutiny of the online community employed netnography, which is simply defined as “ethnography on the Internet”. It utilises publicly accessible information in online forums to recognise and understand the needs and decision influences of target online consumer groups (Kozinets 2002, p. 62).

Finland has a multiparty political system with eight established parties. Most of the time varying amount of candidates endorsed by these parties are elected to parliament, while this is less certain for the candidates represented by a few alternative parties. The Finnish Parliament consists of 200 Members of Parliament (MPs) elected from 13 electoral districts. The population of an electoral district determines the number of MPs in that district. In the parliamentary election we focus on in this study, the largest district according to its number of MPs is Uusimaa, which had 36 MPs, at the time. Satakunta, with eight MPs, is the third smallest electoral district, Lapland is the second smallest with six MPs and Ahvenanmaa (Åland) is the smallest with only one MP. Thus, in addition to representing their party in parliament, MPs also represent their electoral districts, especially in the case of small districts. For this reason, parliamentary elections are considered important not only from a national or party perspective, but also from a regional or from electoral district perspective. In a Finnish referendum, one’s vote is given to an MP candidate. However, Finland’s proportional representation system means that, while one votes for a single MP candidate, that vote also goes to the party.

A novice female politician—a member of the Centre Party—was chosen for the study because she decided to build a brand to increase voters’ awareness of her during her first run for Parliament. The politician gathered a team with whom she created and executed her campaign, with a focus on her personal and professional strengths and her love for the electoral district in question, namely Satakunta. The campaign’s slogan was “From the love of Satakunta”. Thus, she intentionally made extreme emotions the focus of her political message. Her campaign colour was bright pink, which was remarkably different from her party’s typical green palette. Thus, she made conscious choices that presented her personal strengths and qualities, in addition to using her party’s logos and statements while building a brand. Her personal branding process can be considered successful because she was the only one in her party elected to Parliament that year from her electoral district. Further, the politician was the first female Member of Parliament elected from her party in her electoral district in 40 years. During the election year, her party suffered a significant national loss, losing 7.3% of its previous vote level and dropping from first to fourth place in the ranking of the biggest parties (Stat 2019). This politician, however, attracted more votes than any other first-time candidate in her electoral district.

Data Gathering

The first author gathered the Facebook data on an Excel sheet from the launch of the politician’s Facebook site until Election Day, with the timeline being approximately three months of Facebook data consisting of 110 posts, 606 comments, 8858 reactions and 369 shares. The study was planned, and the data were gathered after the election. Complementary data were gathered by conducting a qualitative interview. The politician was interviewed via Skype. The interview lasted for an hour. The second author conducted the recorded interview, while the first author observed. Further, participant observation took place (DeWalt and DeWalt 2011), owing to the first author acting as the social media manager of the politician’s personal brand. The position afforded an opportunity to note comments that were implicit in nature and to see comments hidden from the Facebook site. Further, being a participant in the campaign, the first author had an inside understanding of the discussion, slang and occurrences posted or referred to in the Facebook comments. This can be considered a strength as netnography requires a researcher’s interpretative skills (Kozinets 2002).

The interview with the politician provided background information and assisted us in conducting the data analysis. The politician responded to all comments on social media, including those with a critical tone, except for those that were purely malicious and/or likely generated by trolls. The campaign team concealed the most malicious comments, especially those directed in an extremely unconstructive manner towards certain groups, such as immigrants.

Data Analysis

The analysis of Facebook data broadly followed the principles of netnography (Kozinets 2009). Furthermore, we conducted qualitative thematic analysis, aiming to analyse qualitative data encapsulating patterns across the raw data and arranging it into pertinent themes (Thompson 2022). In our study, and according to abductive research design principles, “existing theory and literature is applied in tandem with the raw data to explain the patterns and story behind the data” (Thompson 2022, p. 1411). In other words, we followed broadly Thomson’s (2022) procedure for abductive data analysis, employing abductive reasoning (see also Kovács and Spens 2005) to review the online comments and examine their theoretical underpinnings.

The first step involved the first author familiarising herself with the data by reading it (Thompson 2022) to obtain a general view and make initial observations, such as the possibility that a posting may elicit both loving and hating responses from consumers. As a second step, she moved to analyse the most obvious content in the comments, noting and marking them as having either a positive or negative tone of voice. In this first-round of coding (e.g. Thompson 2022), we interpreted highly positive WOM expressed as supportive, solidarity and loving words that were sometimes fostered with loving, heart-shaped emojis to signify brand love (see Karjaluoto et al. 2016). Thus, highly positive expressions in the data constituted the brand love category and were marked in pink on the Excel sheet. Accordingly, we interpreted negative WOM expressed as hostile, humiliating and dismissive words that were sometimes fostered with hateful or disgusted emojis to signify brand hate (see Zarantonello et al. 2016). Thus, highly negative expressions in the data were part of the brand hate category and were marked as red on the Excel sheet. On Facebook, people use emoticons and emojis to highlight their meanings and emotions (Brito et al. 2019). Using emojis may even enhance a consumer’s brand attachment (Arya et al. 2018). The meaning of the majority of emojis (heart, angry face, womit, kiss, etc.) was easy to interpret, especially when used together with words; however, the meaning of some emojis (and stickers) was incoherent (see Brito et al. 2019). Further, as the logo of the politician’s party affiliation is the four-leafed clover, the emoji has a brand-related meaning in our study case. Although we thought it was important to include emoticon/emoji in the data, we excluded comments that were difficult to interpret and excluded comments that did not show clear positivity or negativity. The data on brand love and brand hate categories continued to the next stage.

In the third step, the second-round of coding was conducted by the first author who initially analysed the possible underlying content of the comments in the brand love and brand hate categories (e.g. Thompson 2022). Thus, she looked at the data of both categories together, and instead of focusing on whether the comments were positive or negative, she let content-related themes emerge from these expressions. Here, she noted that both expressions of brand love and brand hate had common themes that corresponded with the shared mechanisms of brand love and hate. Thus, it seemed that, in our data, brand love and hate were expressed through self-expression, strong emotions, identification with the brand, identification with users of the brand, values and prestige, attitude and engagement, brand experience and satisfaction. At this stage, Table 1 began to take shape as a result of analysing the data and allowing our initial observations to guide a fresh categorisation of brand love and brand hate literature, leading to the discovery of the shared mechanisms of these two extreme brand relationships. This stage was an iterative and flexible process of theme development. Theorisation from the data was conducted by reviewing the theoretical knowledge and frameworks related to brand love and hate and determining the extent to which these could explain our findings (Thompson 2022). Furthermore, a new mechanism, spill-over, was observed arising from the data, resulting in further theoretical development (Thompson 2022). To our knowledge, spill-over has not previously been mentioned in either the brand love or brand hate literature. In addition to these shared mechanisms, the first author observed that certain expressions were highly polarised, and that the same comment by the same commenter could encompass expressions of both brand love and hate. This led us to reflect that, instead of having a purely positive or negative relationship with a brand, an individual’s relationship with a brand may be polarised. We named this theme “co-existence of brand love and brand hate”.

The fourth step involved the second author reading the data that the first author organised, with both researchers identifying which discussions were particularly relevant to the study. Fifth, inter-coder reliability was ensured through member checks (Lincoln and Guba 1985), during which both authors discussed the data and the themes/narratives that emerged from it. During this last step, the original data were read once again to ensure that all parts meaningful to this study were taken into account.

The authors translated authentic extracts from Finnish into English and attempted to sustain the original tone of voice of comments. The comments are anonymised and the dates of the comments have been omitted to protect the commenters’ identities, even though they decided to write public comments. Further, we decided not to state the politician’s name in the study, even though the politician gave permission to publish her name. Next, we show our findings and the path of thematic analysis behind our notions. We adhere to Thompson's (2022) guidance, recommending the presentation of findings with headings representing each theme.

Findings

The development of Table 1 began, as noted, from our data showing that brand love and brand hate appeared to share similar mechanisms. By mirroring our notions from the data with the current literature, we were then able to construct a view of shared mechanisms in the concepts of both brand love and hate, as outlined and demonstrated in Table 1. Thus, the construction of Table 1 was based on the identification of common mechanisms underlying brand love and brand hate in the extant literature. Compilation of the table was informed by an extensive literature review, which revealed similarities between brand love and brand hate. Importantly, no previous literature has presented a similar table or identified such similarities between brand love and brand hate. This table thus represents a new theoretical development, thereby contributing to the literature on brand relationships. We further extended the findings presented in Table 1 by empirically showing whether and how these mechanisms are manifested in expressions of brand love and hate in the online brand community of a personal brand. This investigation thus provides new empirical information on these mechanisms in real-life settings.

WOM and Self-expression

We chose eWOM to represent the mechanism of brand love and hate that we studied empirically. Further, as eWOM is a way for a consumer to express themselves in the social media context, the comments we analyse are examples of brand love and hate related to self-expression, as well as examples of brand love and hate related to WOM. Through eWOM, we observed other mechanisms of brand love and hate. Thus, we demonstrate the way that other shared mechanisms of brand love and hate work behind expressions of brand love and hate on the Facebook page of a personal brand. We show how we interpreted our data and show that, on Facebook, brand lovers and brand haters communicate the underlying aspects of brand love and hate through eWOM.

In addition, we noted that brand love (and probably brand hate, although not shown in our data, except through emoticons and emojis) may be expressed through visual WOM. This was shown when support for the brand and identification as part of a group of supporters were expressed by a consumer who uploaded a brand-related photo to a comment on the brand’s posting. These visual expressions of brand love included photos of a consumer’s dog wearing the brand’s campaign brooch or a consumer’s child wearing the brand’s campaign bag. Next, we describe other mechanisms of brand love and hate that are expressed through eWOM. At the end of each quote, we use brackets to show the main mechanism we interpreted beneath the expression. In some cases, we interpreted the comment as being related to more than one mechanism, but we chose to show only one to keep the findings clear. However, as we consider it to be a finding itself, that there might be more than one mechanism behind the expressions of brand love and brand hate; we also provide a few examples in our findings.

Strong Emotion

We observed that strong emotions are shown in expressions of brand love and hate. These emotions are more manifold than mere liking/loving or disliking/hating the brand. For example, the expressions of brand hate that relate to money and capitalism might arise from feelings of envy or fear and worry regarding one’s own personal financial situation. On the contrary, expressions of brand love mentioning one’s own children and grandchildren bring love towards one’s own family into their expressions of brand love:

“If children and grandchildren are well, then Grandma and Grandpa are also happy!” (Positive emotions)

“Hey, come on, all that is tosh, everyone who is sane can see that. Read those happiness messages that real citizens share, and don’t drivel capital elite’s nursery rhymes to us!” (Negative emotions)

Further, our data show that strong emotions are a part of brand love and hate in the context of personal political brands.

Attitude

We observed that strong attitudes are shown in expressions of brand love and hate. For example, the expressions of brand hate that relate to money and capitalism might arise from negative attitudes towards a capitalist world, as well as people and things seen to represent that world. Furthermore, we observed that the brand's gender influenced some comments.

“Great idea. Without children, it is difficult to think about life. You are on the right road. This gives me the strength to think about tomorrow. Good luck for your work”. (Positive attitude)

“Good success [politician’s first name]. You belong to the category BRAVE WOMEN … BIG HUGS”. (Positive attitude, related to brand’s gender)

“There again go the empty promises!” (Negative attitude)

Further, our data show that strong attitudes are a part of brand love and hate in the context of personal political brands.

Engagement

Writing a public comment on a brand’s Facebook page requires some degree of engagement with the brand. Furthermore, we observed that some consumers repeatedly commented on the brand's postings, indicating frequent engagement with the brand. Thus, we interpreted our data as showing positive engagement in expressions of brand love and negative engagement in expressions of brand hate.

Identification with the Brand

We noted that identification with a brand is shown in expressions of brand love. This was shown in comments with a positive tone of voice towards the brand, indicating, sometimes repeatedly from the same consumer, that the commentator experienced a self-brand connection.

“That [breaking a glass roof to get a female Member of Parliament elected for the party from the electoral district] will succeed! We together do it for the best of Satakunta and FROM THE LOVE OF SATAKUNTA”. (Identification with the brand)

Further, our data show that incongruence between self-image, brand image and taste system-related incompatibility, that is, identification with the antagonist of the brand appears in expressions of brand hate, as shown in examples below:

“Empty speak … she is in management work in Helsinki … this [politician] herself has no problems”. (Incongruence with the brand)

In addition, our data show that identification/incongruence with the brand is part of brand love and brand hate in the context of personal political brands.

Identification with Users of the Brand

Identification with other users of the brand, along with a sense of community and we-spirit, demonstrated expressions of brand love in our data, as illustrated by the examples below:

“[The politician’s name] phenomenon keeps growing to its peak ” (Identification with users of the brand)

“It is great to be part of the journey [the campaign], now full speed to the goal!” (Identification with users of the brand)

Identification with users of the brand was also obvious in people’s positive reaction to each other’s comments, either with loving reactions or with words stating that they thought similarly to the other commentator.

Further, our data show that incongruence between self-image, brand image and taste system-related incompatibility, that is, identification with the antagonist of the brand’s users, appears in expressions of brand hate, as shown in examples below:

“No for kepulit [negative nickname for members of the Centre Party]. I will give my vote for supporters of Finland, not for eaters [meaning people who take what is others]”. (Incongruence with users of the brand)

In addition, our data show that identification/incongruence with users of the brand are part of brand love and brand hate in the context of personal political brands.

Values

Our data show that values are present in expressions of brand love and hate. Value congruity was shown, for example, in a comment in which the brand posted about the meaningfulness of reading to children in the same posting that she commented about a low birth rate in Finland:

“…It feels good that you have assimilated also this issue for us to tell. Reading is always worthy. It gives so much for the child and the reader. Thank you for good cause”. (Value congruity)

In line with this, our data show that value incongruity is linked to expressions of brand hate. This was shown, for example, when a commenter used hostile words to show how they did not accept liberal values:

“Like this speak of the Centre [Party], which allows Islamisation of our country, love for children and older people are far from that”. (Value incongruity)

Thus, we observed that values underlie expressions of brand love and hate. We noted that in the personal political brand context, incongruity is a mechanism of both brand love and brand hate.

Prestige

We noted that prestige is present in expressions of brand love and hate. Our data show that sponsored postings garnered more expressions of brand hate than organic postings. Sponsored postings are paid by the brand to be shown more often, reaching consumers who do not follow the brand’s Facebook page (see also Kruschinski et al. 2022). Organic postings are the brand's ‘normal’ content that is not paid for and, from an algorithmic perspective (as different social media platforms have their own algorithms), are more akin to individual users' postings. Thus, in our data, the brand's postings that appeared in the feeds of non-followers on Facebook attracted more negative feedback compared to non-sponsored postings. This was especially the case before the politician appeared on national television and was not yet known as a personal brand, but rather as a member of a certain party. Our data show that, alongside the campaign, the volume of expressions of brand hate decreased. This indicates that it is easier to hate unknown personal brands. Further, this highlights how a lack of brand prestige and a good reputation (which a new brand does not have in the beginning) may lead to expressions of brand hate.

In line with this, we noted that positive and loving expressions increased while the campaign proceeded, culminating in the campaign’s last posting, in which the politician thanked the voters. This single posting garnered 141 comments, all positive in nature. As a result, we observed that brand prestige and reputation lay beneath expressions of brand love and hate. We noted that in the personal political brand context brand prestige and reputation are parts of brand love and brand hate.

Brand Experience

We noted that brand experiences are present in expressions of brand love and hate. Our data show how positive brand experiences and satisfaction with the brand drive expressions of brand love, as shown in the comments below:

“Indeed, emphatic, intelligent, a person appreciating all others, [politician’s name] ”. (Positive brand experience)

Accordingly, our data show that negative brand experiences are present in expressions of brand hate. This is shown in the comment below:

“Wrong party . Achievements of Centre [Party] really show in the lives of citizens. Poverty has exploded, citizens are doing poorly ”. (Negative brand experience)

In addition, we noted that brand experiences include brand love and brand hate in the personal political brand context.

Satisfaction

We noted that dis-/satisfaction are present in expressions of brand love and hate.

“The joy of doing election work shows in you. You care to face everyone ”. (Satisfaction)

“You received a few extra votes today during your visit to Eura. Hopefully, you'll get elected, and the work for the good of Satakunta continues. We farmers are hoping for improved profitability in the future”. (Satisfaction)

Accordingly, our data show that dissatisfaction towards the brand are present in expressions of brand hate. This is shown in the comment below:

“Parliamentary work will stop as soon as one will be [elected] as a member of parliament, enjoying good benefits and salary”. (Dissatisfaction)

In addition, we noted that dis-/satisfaction includes brand love and brand hate in the personal political brand context.

Spill-Over

Notably, dissatisfaction, negative attitude, negative emotions, value incongruence and even negative brand experience might be expressed due to the politician’s party rather than the politician herself. Thus, emotions, attitudes and dissatisfaction towards something else and experiences gained from something other than the actual brand can be channelled into expressions of brand hate. This negative spill-over effect was shown, for example, in expressions of brand hate that indicated negative emotions and attitudes, and dissatisfaction and value incongruence were transferred from the party and prime minister (during the election, the politician’s party was the prime minister’s party) to the politician's brand, as shown below:

“X [name of the PM] has ruined all the efforts of the Centre Party members to enter the Parliament by making cuts and tightening the belt for the poor”. (Negative brand experience; Value incongruence)

“Oh Lord, no more kepulit [negative nickname for members of the party], this country cannot stand you anymore”. (Negative emotions; Negative attitude)

“Kepu [Finnish nickname for the Centre Party] betrays always. Last time, [they betrayed] their government partners”. (Dissatisfaction; Negative attitude)

“Wondering if she is trying to fill the shoes of [name of a former minister from the party] through the shadow of her father - I think those are too big boots to fill for a girl”. (Included also a sticker of a vomiting monster.) (Negative attitude, related to brand’s gender)

In our data, the positive spill-over from the brand’s stakeholders to the brand was a bit more delicately expressed. There were several expressions of brand love in which the emoji were used, thus showing support for the brand’s party affiliation. There were also expressions of joint brand love towards the politician and another politician from her party when posting pictures with party colleagues (first and second comments below). Some expressions concentrated on defending the acts of the Centre Party when others criticised the party’s actions and a comment showing joy for the party when the politician was elected to Parliament (third comment below):****

“Good spirit and success to election for both of you gorgeous candidates ”. (Positive attitude)

“[They are] The hopes for the fatherland ”. (Positive emotions; Value congruity)

“Big congrats, excellent campaign and fantastic outcome In addition, I can say that you saved the Centre [Party] in Satakunta as here we [the party] lost percentually least when considering all electoral districts! It was so sad that [name of other candidate from the party] was left outside in finishing lines ” (one other candidate was almost elected also, but the last few votes went to another party) (Identification with brand and users of the brand)

Thus, our data show negative spill-over from brand stakeholders in expressions of brand hate and positive spill-over in expressions of brand love. This caused us to note that, in addition to the shared mechanisms we found through theoretical elaboration, there is also one more shared mechanism of brand love and hate, namely, spill-over. Further, this made us to note that, expectedly, political polarisation relates to expressions of brand love and hate in the context of personal politician brand, where it is more natural to consumer to love and express love towards a politician from consumers “own” party than to party from opposite political pole.

Co-existence of Brand Love and Brand Hate

Interestingly, we noted that an expression might include both brand love and brand hate. This was present in cases where brand hate was expressed towards the party brand and brand love was expressed towards the personal brand, in comments such as:

“Smart girl and we even have the same family roots, but the party is wrong”.

“[Last name of Prime Minister] chips  Insipid and unpredictable. Games of Kepu [Finnish nickname for the Centre Party] are over. You are in the wrong party. However, good luck to your life ”

These comments show how a consumer may simultaneously have both a positive and a negative, and thus a polarised, stance towards a brand. Our data clearly indicate the co-existence of brand love and brand hate within individual consumers, representing a form of brand polarisation. In addition to these observations, our overall Facebook data demonstrate that while some individuals express affection for the brand, others harbour disdain for it, and both are simultaneously expressed through similar mechanisms in the same platform. This phenomenon represents another form of brand polarisation. Our data demonstrate that negative comments are not always harmful to the brand. The politician stated in her interview that she noted that negative comments helped in the branding process to a certain extent, in that such comments provided her with an opportunity to defend her knowledge and values. She also stated that she built herself a brand by using SWOT analysis but noted that it is the fans and media that determine “how the brand is seen”.

Discussion and Conclusions

The purpose of this study was to increase the understanding of extreme positive and negative brand relationships, namely brand love and brand hate, and their conceptual interrelatedness. The purpose was achieved by investigating the mechanisms of these concepts, both theoretically and empirically. The study’s theoretical contributions are in advancing the understanding of extreme brand relationships and the connections between them. To the authors’ knowledge, no previous studies, either on brand love or brand hate, have focused on personal brands in general or personal political brands in particular. Therefore, the study contributes to the intersection of extreme brand relationships (brand love and hate), personal branding and political branding and marketing.

Theoretical Contributions

The first theoretical contribution lies in unveiling the conceptual interrelatedness of brand love and hate through an elaboration of the current literature on the topic. We derived insights from our data, indicating that brand love and brand hate share comparable mechanisms. Subsequently, we constructed the framework presented in Table 1 through an abductive process (e.g. Thompson 2022). Building upon the interplay between the literature and our empirical findings, we demonstrate that both concepts are underpinned by the same mechanisms. These comprise WOM, self-expression, strong emotions, brand and user identification (and their opposites), values, prestige, attitude, engagement and brand experience, as well as satisfaction and dissatisfaction. This finding adds to the current literature, which has mainly focused on studying brand love (e.g. Albert and Merunka 2013) or brand hate (e.g. Fetscherin 2019) separately. Hence, our findings show that brand love and hate are conceptually closely related, rather than representing opposites of each other. However, the shared mechanisms mainly have a positive charge when related to brand love and negative charge when related to brand hate. Our findings resonate with the notions presented in the context of brand trust and distrust (Mal et al. 2018, p. 945), which suggest that “brand trust and brand distrust are polar opposite constructs, as the same characteristics and behaviours inform both brand trust and brand distrust beliefs”. This demonstrates that two brand-related opposite concepts may actually share underlying mechanisms.

The second theoretical contribution lies in the study’s context, noting that brand love and brand hate include the aforementioned mechanisms in the personal political brand context. As there does not appear to be any previous brand love or brand hate studies in the personal brand context and only one study of both in the political brand context (Banerjee and Chaudhuri 2022; Banerjee and Goel 2020), this study broadens the discussion of brand love (Batra et al. 2012) and hate (Zarantonello et al. 2016) as it is conducted in the context of personal political brand (Armannsdottir et al. 2019). In addition, due to its context, this study brings political polarisation (see Weber et al. 2021) into discussion of brand love and hate as political identity of a politician brand was reflected in expressions of brand love and hate. In some expressions of brand love, match with politician’s political affiliation was shown whereas in some expressions of brand hate the mismatch with politician’s political affiliation was shown.

This study’s third theoretical contribution is in noting how expressions of brand love and hate broadly show emotions, attitudes, engagement, values, cognitions and experiences related to these brand relationships. Thus, expressions of brand love and hate not only constitute declarations of love (I love this!) or hate (I hate this!), but instead include expressions of all mechanisms of brand love and hate. In addition, our findings show that these mechanisms are intertwined. This was shown in the fact that several expressions of brand love and hate consisted of more than one underlying mechanism; for example, one expression of brand love may simultaneously include a positive attitude, satisfaction and positive emotions. Thus, our study fosters earlier studies that connect positive WOM to brand love (Bairrada et al. 2018) and negative WOM to brand hate (Zhang and Laroche 2020), and studies connecting brand love (Ahuvia et al. 2009) and brand hate (Sarkar et al. 2020) to self-expression and engagement (Brandão and Popoli 2022; Wallace et al. 2014). Our findings support an earlier study (Kaiser et al. 2020) that noted that, in the social media setting, brand love can be expressed with visual WOM, not just verbal. Further, our study implies that brand hate can also be expressed with visual WOM. Thus, our findings support earlier literature noting that brands offer a vehicle for expressing emotions (Lee 2021) and that brands are vehicles of self-expression on social media (Wallace et al. 2014).

The fourth theoretical contribution lies in the notion that regardless of a brand’s own actions, sometimes a brand is hated or loved because of the actions or image of its stakeholders. Thus, this study demonstrates that brand love and hate with a personal brand may have spill-over effects with organisations and groups that the personal brand presents, and the other way around. These organisations may be political parties or corporations funding and supporting election campaigns, and groups may be formed of members and supporters of the party, for example. Our findings further support Schönhage and Geys’ (2023) notions regarding how a personal scandal of a politician may impact the brand of other politicians in a same Party as well as to the Party’s brand. Accordingly, we found a new shared mechanism of brand love and hate: spill-over, which is shown in Table 2. Thus, our findings support earlier studies indicating that the formation of brand love might be enhanced by attractive celebrities promoting the brand (Zhang et al. 2020; Zhou et al. 2020) and that the formation of brand hate might be driven by brand-related aspects, such as country of origin (Bryson et al. 2013), by noting that the stakeholders may have a part in the formation and expressions of brand love and hate. Our notion that the politician’s party affiliation leads to expressions of brand hate is in line with the notion that party affiliation plays a major role in political branding (e.g. Bennett et al. 2019).

Table 2 Interrelatedness of brand love and brand hate
Full size table

The fifth theoretical contribution demonstrates that consumers are willing to publicly engage with polarised brands, which corresponds with the existing research (Osuna Ramírez et al. 2019). Our findings advance the knowledge of brand relationships by showing that a consumer may have a polarised brand relationship in which brand love and brand hate co-exist. A consumer might, for example, love the politician but hate their party, thus having a polarised relationship with the politician’s personal brand. This finding confirms the notion that “the opposite of love may not be hate, since the two feelings can co-exist, and one can love some things about one’s partner and hate others at the same time” (Sprott 2004, p. 304). Thus, we noted two forms of brand polarisation: one in which a consumer has a polarised relationship with a brand, and one in which a group of consumers maintain polarised brand relationship, with some loving the brand and others hating it. A recent study (Aro et al. 2023, p. 101) noted that “brand lovers can also distance themselves from aspects of the brand they consider incongruous with their identity or that do not fit with their inner self”. Our study furthers this notion of paradox in brand relationships by showing how a consumer may have a polarised relationship with a brand. In addition, our findings suggest that people who love or hate a certain brand might have an interest in the context (in our case politics) in which the brand acts (see Banerjee and Goel 2020). Our notion extends Osuna Ramírez et al.’s (2019) finding that negativity towards a brand can actually be an opportunity for a brand, as our interview data showed that answering negative comments may help a brand express its strengths.

Managerial Implications

This study suggests that a new brand might profit from the negative comments (when comments are constructive, not hate speech) it receives from consumers (or from the media) if the brand is willing and able to answer these negative comments and, therefore, show its actual strengths. Thus, when building a brand, answering some negative critiques might open an organic channel to demonstrate the brand’s substance. Therefore, brands should not be too alarmed by all comments of brand haters’. Further, haters who visit a brand’s page to publicly comment on the brand’s postings offer the brand a touchpoint, presenting an opportunity to persuade them to change their mind about the brand, especially if the brand hate arises from a misunderstanding, misperception or negative spill-over. Thus, also in brand hate, there is a relationship between a brand and a consumer.

The findings indicate that sponsored posts gather more negative WOM than non-sponsored posts. This is explained by the fact that, on Facebook, the brand followers usually feel connectedness with the brand, and sponsored posts are also visible to people who do not follow the brand. This offers insights for the brands in planning which posts to sponsor and which would be more powerful when presented to the followers of their page. In addition, as the algorithms of different social media platforms change and consumers may suddenly move from one platform to another, brands should actively observe what platform their target groups use and be alert to algorithm changes. According to this study’s findings, strategic and active branding is also beneficial in personal brand and political brand contexts. When the brand gains more publicity and has more opportunities to communicate and show “what the brand is made of”, the number of negative comments may diminish. Thus, a new brand should actively seek to express and communicate its authentic values and strengths (see also Vredenburg et al. 2020).

In addition, even though brand building is important in moving towards an ideal brand identity, it is the WOM, rumours, public media, consumer’s brand experiences, etc., which eventually determine how the brand is seen; in other words, the brand’s image. For this reason, it is meaningful for a brand, especially a new one, to have direct and/or personal relationships and communication channels with consumers and representatives of the media to convey the brand’s own message to the audience. Thus, in addition to being visible in the media and participating in events, interactively and actively maintained social media accounts might help build a (political personal) brand.

Hate speech is an emerging phenomenon in contemporary conversation culture. There is a societal need to find ways to tackle this problematic issue, particularly on social media. For example, in the UK, there are cases where MPs' lives and families have been threatened, causing MPs to consider whether the work is worth the risk (BBC 2019). Also in Finland, many female politicians have unfortunately become targets of hate speech, with former Prime Minister Sanna Marin being perhaps the most well-known example. When private party videos of her dancing spread internationally, she faced judgement. However, simultaneously, she garnered strong support both nationally and internationally. For instance, BBC News, The Guardian, CNN and Reuters showed their solidarity with Prime Minister Marin. Thus, there is a possibility that a negative attitude towards a politician's personal brand may also spark something positive. However, it is important to oppose and combat all forms of hate speech and bullying and to defend a respectful discussion culture. This study is one of the first attempts to apply the lens of brand relationships to understand the mechanisms behind hate speech. Therefore, the study plays a humble part in addressing the global problem of hate speech by keeping the phenomenon in academic discussion and bringing it into the field of brand relationships.

Limitations and Future Studies

Although the empirical research findings are most applicable for personal political branding, they can prove useful to the area of personal branding and political branding as well, particularly regarding expressing brand love and brand hate towards personal brands in social media. Further, the findings of this study—that is, the shared mechanisms of brand love and hate—could be tested and studied in the context of other brand types to see whether and how these mechanisms are apparent. As our study focused on a personal brand, the manifestation of brand anthropomorphism was not evident in our data, given that the brand itself is human. However, existing literature links anthropomorphism to brand love (Rauschnabel and Ahuvia 2014), and reverse anthropomorphism to brand hate (Kucuk 2020). Hence, this shared mechanism may be identifiable within the context of other brand types. The current study provides rich data from a brand from one country. Therefore, future studies connecting brand love and brand hate, both qualitative and quantitative, could cover other types of brands and contexts in other countries and cultures. In addition, as this study showed the co-existence of brand love and brand hate, and given that brand polarisation is an unstudied field of research, further studies relating to the concepts of brand polarisation and polarised brand relationships are encouraged.

Finally, studying brand hate as a consumer–brand relationship remains a neglected area of study. Although brand hate is considered a consumer–brand relationship (e.g. Brandão and Popoli 2022), there is a lack of studies actually conceptualising and studying it as a relationship. Our study attempts to bring the relationship view into brand hate studies by introducing the dynamic mechanism approach; however, further studies are needed to fully understand the dynamic relationship nature of brand hate. In line with this, the dynamic nature of brand love deserves further study.