Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por email
Citado por SciELO 
Similares em
SciELO 
Permalink
1. Introduction
As the number of social network sites (SNS) users continues to grow, platforms such as Facebook are increasingly adopted by marketers to interact with their target audiences. Despite the recognized opportunities offered, the challenge of standing out and earning users' attention and time is also clear. Guerrilla marketing is an advertisement strategy that uses surprise and unconventional interactions to impact the consumer unexpectedly and to create strong memories and interactions while generating word-of-mouth. Jay Conrad Levinson was the first to coin guerrilla marketing and stressed that technology was essential to its success (Levinson, 2007), particularly for campaigns' rapid spread and reach. Hence, SNS stand out due to their mechanisms facilitating communication reach, namely by sharing and exposing users' interaction with content in the murals of their network.
Considering that, ultimately, the effectiveness of a guerrilla campaign in SNS depends on the active participation of users through sharing and overall interaction with its content, it is essential to understand the factors that explain interaction with this type of marketing communication and also the possible impacts on brand image. However, the contributions in the literature are still scarce, as shown by Yan, Rashid, and Mohamed's (2017) systematic review on unconventional advertising, who found in the Scopus and Web of Science databases no more than 20 articles for guerrilla advertising, 27 articles for ambient advertising and 39 articles for ambush advertising published in the 2010-2017 period. Interestingly, the number of guerrilla marketing studies has risen in recent years. The Web of Science database reports a total of 100 journal articles on guerrilla marketing and related topics (i.e., "guerrilla marketing" OR "unconventional marketing" OR "ambush marketing" OR "ambient marketing" OR "invisible marketing") published between 1989 and 2011, of which 42 were published between 2018 and 2021.
This study aims to explore the impacts of guerrilla marketing campaigns on Facebook on brand image and content sharing intentions. After reviewing the main contributions in the literature on the subject, which is presented in section 2, a mixed-method study was carried out, comprising focus groups and an online survey. Section 3 describes the methodology adopted, section 4 is dedicated to the analysis of results, and finally, section 5 includes conclusions, implications for managers, and suggestions for future research.
2. Literature Review
Guerrilla marketing is an advertisement strategy in which a company uses surprise and/or unconventional interactions to promote a product or service. The term was introduced by Jay Conrad Levinson in 1984, comprising unconventional marketing campaigns that allow to overcome media saturation and achieve a high return on investment by communicating with a limited number of consumers in an impactful, unconventional, and unexpected way that will lead to target consumers to spread the advertising message and ultimately draw the attention of a larger audience.
Levinson (1984) proposed five characteristics of guerrilla advertising: (i) unexpectedness, namely related to the location and the surprise effect that is created in the target public, raising emotional reactions and fostering memory and attention; (ii) creativity, by adopting innovative communication approaches, which will also have a positive impact on targets' attention; (iii) cost-effectiveness, both because of being less expensive than traditional advertising, and because of the gains that result from targeting only a few individuals, and getting them to help spread the message; (iv) interactivity, as typically a guerrilla marketing campaign, will rely on interaction with customers and prospects to excel its spread and impact; and (v) simplicity, as guerrilla marketing campaigns should provide simple messages to be easily understood by consumers, despite, for instance, the unexpected settings and the creative approaches that might distract them.
Several authors (e.g., Hutter, 2015; Hutter & Hoffmann, 2011; Sajoy, 2013) stress that the success of guerrilla marketing is built on surprise, diffusion, and low cost. The fundamental principle of the guerrilla approach is to surprise consumers with unexpected activities and to draw attention to the advertising message. The surprise effect results from the divergence between perceptions and expectations (Hutter, 2015). Advertisers can use different methods to surprise consumers, such as nonsense, humour, and shocking messages and placing advertisements in unexpected locations. It is assumed that the surprise effect will potentiate the message diffusion effect (Hutter & Hoffmann, 2011), which is another essential success factor. The Internet is particularly useful for the diffusion effect. As stressed by Castronovo and Huang (2012) and Levinson and Gibson (2010), online channels, namely SNS, help improve the efficacy of guerrilla marketing by enabling users' interaction and by facilitating content virality. So, usually, marketers adopt integrated marketing strategies, combining online and offline efforts, to excel the efficacy of guerrilla campaigns, being also possible to design strictly online campaigns, namely when dealing with low budgets and targeting a segment of consumers that are particularly active online, as demonstrated by Levinson and Rubin (1995).
Guerrilla communication campaigns comprise various strategies, including ambush, ambient, and invisible advertising. Ambush advertising was defined by Chadwick and Burton (2011) as a form of capitalizing on awareness and other benefits by an unauthorized association with an event or property, using others' venues to spread advertising messages. Ambient advertising refers to placing unconventional advertisements in an unusual location (Hutter & Hoffmann, 2011), which results in creating incongruous stimuli by placing the campaign in an unfamiliar environment (Hutter, 2015). Hutter (2015) calls it the guerrilla version of outdoor advertising, stressing that to be innovative, the ad must have both a temporal and unusual location and often have an unconventional execution. Invisible communication was defined by Finne and Strandvik (2012) as hidden and embedded messages intending to involve consumers without letting them notice it is a commercial message. Hence, all these forms of guerrilla communication campaigns have in common the fact that they intend to surprise the consumers and trigger emotions.
The arousal of powerful emotions will make consumers more willing to share the experience with peers. In fact, as Holbrook and Batra (1987) argue, emotions are mediators of consumer responses to advertising, and Olney, Holbrook, and Batra (1991) demonstrated that advertising content appeal and uniqueness trigger emotional responses such as pleasure and arousal. Moreover, Ladhari (2007) showed that pleasure and arousal have significant effects on willingness to generate word-of-mouth communication. Creating a surprise effect seems to be particularly effective, as intense, positively and negatively surprising experiences clearly impact sharing details with others (Derbaix & Vanhamme, 2003).
Extant literature shows the positive relationship between emotions and word-of-mouth communication in general (Keiningham, Rust, Lariviere, Aksoy, & Williams, 2018), which also holds for guerrilla marketing characteristics, namely its inherent surprise effect. Dahlén, Granlund, and Grenros (2009) demonstrated that consumers perceive guerrilla campaigns as valuable by consumers and positively impact word-of-mouth communication behaviour. Dinh and Mai (2016) propose that factors explaining a guerrilla campaign's effects include novelty, relevance, aesthetics, humour, clarity, surprise, emotional arousal, and message credibility. This is according to other contributions in the literature that support some of these factors as explaining word-of-mouth communication behaviour. One conclusion of their study pointed to a positive impact of creativity in the intention to share impressions about the advertisement with others while demonstrating no impact of humour and emotional arousal on word-of-mouth communication (Dinh & Mai, 2016). Hence, more studies on emotional appeals in guerrilla marketing and its effects on consumer behaviour are needed to validate these expected positive relationships further.
The literature reports other factors that explain the impact of guerrilla marketing campaigns, such as attitude toward the brand (Rosengren, Modig, & Dahlén, 2015), brand image, loyalty, and purchase intention (Hutter & Hoffmann, 2011). As stressed by Fournier (1998), consumers' relationship with brands and purchase decisions are explained not only by functional and utilitarian meanings but also by psychological and emotional ones, so emotional arousal plays an important part in the efforts made by brands toward their target customers. For instance, marketing activities and electronic word-of-mouth are strongly connected with perceived trust, which in turn determine purchase intentions (Garg & Kumar, 2021). Considering that controversial guerrilla marketing often gives rise to both positive and negative emotions, the impact on the brand has to be carefully anticipated. Again, further studies are needed to fully understand the outcomes of guerrilla campaigns for brands.
Another interesting aspect to consider is the profile of the consumers targeted by guerrilla marketing. Some studies have approached this topic, being Dinh and Mai (2016) one of the most prominent ones. These authors studied guerrilla campaigns' effects on the Millennials, who are considered a unique consumer segment due to their propensity to the ubiquitous presence of technology in their lives since they were born and have been using the Internet more than any other generation (Dimmick, Ramirez Jr, Wang, & Lin, 2007). Overall, it is widely accepted that consumers do not react equally to communication appeals, and demographic, and personality traits are often tested to better define the most appropriate targets for innovative communication approaches. For instance, Strutton, Taylor, and Thompson (2011) found that Millennials engage more on SNS than Generation X, but both generations present similar technical skills and propensity to spread online word-of-mouth communication - Generation X will use more e-mail.
Hence, by depending on the active interaction of the users, online guerrilla marketing is expected to be particularly successful amongst the most intensive users of online platforms. These users are also more prone to publish and share on SNS, as current internet research often uses both frequency and types of activities performed on SNS to measure the usage intensity, being the two highly correlated indicators. Still, Chun and Lee (2016) verified that advertisement content matters, as well as the involvement with the brand, concluding that hedonic content has a particular interesting impact on consumers with low involvement with the brand being advertised, meaning it is especially adequate for targeting new customers and new market segments.
3. Method
The success of digital guerrilla marketing campaigns depends on users' willingness to collaborate by sharing their experiences and advertisement content with other consumers. So, it is essential to grab consumers' attention and to provide a powerful experience that makes her/him interact and share the guerrilla content. This type of campaign is also expected to affect brand image and content sharing intention. Considering the contributions provided by the literature on the topic, this study defined two research questions:
RQ1: What factors influence the interaction of Facebook users with guerrilla marketing content?
RQ2: How does a guerrilla marketing campaign impact consumer behaviour?
This study adopted a mixed-method approach, comprising one exploratory and the qualitative first phase, which was followed by a quantitative phase that enabled hypotheses testing. Thus, the mixed-method approach consisted of the systematic integration of qualitative and quantitative methods in the same study to provide multiple perspectives on the research problem, taking advantage of the complementarity of the two approaches and using the results of the first stage to inform and support the second one. It also enables methodological triangulation since two different methods are used to study the same phenomenon to respond more accurately to the research questions by combining conclusions from the two methods (Bryman, 2016).
3.1 Focus Groups
The study's first phase comprised four focus groups with 26 Facebook users, using a convenience sampling technique. Focus groups are frequently used in exploratory research in cases of scarce literature on the topic and when it is intended to determine characteristics and factors explaining behaviour. They comprise conversations about a topic proposed by the moderator to gather participants' experiences, opinions, expectations and views. In this case, the main objective was to understand facilitators' and consumers' motivations to interact with guerrilla marketing content on Facebook and the impact on brands associated with the campaign. The motto was provided by two guerrilla campaigns: (i) Sagres, a Portuguese beer brand and sponsor of the Benfica Football team created a campaign making fun of a failure of the rival's goalkeeper Rui Patrício during a match between Belenenses and Sporting , and (ii) the TV series Dexter guerrilla campaign examples, including placing human body parts at a butcher shop .
The four focus groups were organized to guarantee diversity in terms of age, gender, education, and intensity of Facebook usage. Participants (26 in total) were both male and female with ages ranging between 18 and 48 years, including undergraduate, graduate and post-graduate Facebook users, both frequent and sporadic. The focus groups were audio-recorded, then transcribed, and subjected to theory-driven content analysis.
3.2 Online survey
The qualitative study provided contributions to the survey that was developed afterwards. Surveys are particularly adequate for testing relationships between variables, such as the ones implicit in the research hypotheses proposed for this study. Online surveys have various advantages and disadvantages (e.g., low cost, convenience, targeted mainly at online consumers and more intensive users of the Internet). Still, considering that the study focused on guerrilla marketing campaigns on Facebook, an online survey was considered the most adequate method. Three research hypotheses were defined:
RH1: Interaction with a guerrilla campaign is positively associated with the intensity of Facebook usage;
RH2: The type of interaction with a guerrilla campaign is associated with the perceived content characteristics;
RH3: Perceived content characteristics have a positive effect on brand image.
For the quantitative phase, a guerrilla marketing campaign was developed for a fictitious brand of a restaurant. Firstly, a Facebook page was created, announcing the soon-to-be-launched new restaurant in Aveiro, a medium-sized city in Portugal (78,450 inhabitants in 2011). The guerrilla campaign was launched after the brand page reached 250 followers. The campaign (Figure 1) consisted of posting an adaptation of the well-known Leonardo da Vinci's painting "The last supper" by replacing the heads of the characters with the faces of Portuguese public figures all named José (e.g., José Saramago, José Mourinho, José Sócrates, José (Zeca) Afonso), creating political and religious controversy. The image was published with the text (in Portuguese): "By chance, it was my idea to get this little group together at Friends Aveiro. But speaking of something else, hopefully, this will not be the last supper of anyone!!!". Moreover, the copy included links to two pages on Facebook with a high number of followers, which is also a guerrilla technique (invisible marketing), aiming to increase the reach of the post and attract followers from local communities. Another guerrilla technique, ambush marketing, was also used to disseminate the post by posting the link of the campaign on Facebook communities and discussions chosen according to the target audience (i.e., residents in Aveiro). The campaign was online for six days (13-18 October 2015).
This campaign reached 3309 people (study's population), of which 533 interacted with the post by liking, commenting or sharing. After being reached by the campaign, Facebook users were asked to answer an online survey. The sample is formed by 256 individuals of the population, aged between 15 and 55 years, who answered the questionnaire.
The survey featured six types of variables (Table 1): (i) sociodemographic characteristics, including age, gender and academic qualifications of the user; (ii) Facebook usage intensity; (iii) activities carried out on Facebook; (iv) type of interaction with the guerrilla campaign; (v) perceived characteristics of the guerrilla content, and (vi) brand image.
The responses were processed and subsequently analyzed through the application of statistical methods, including descriptive and exploratory analysis. The hypotheses were tested by performing non-parametric tests like the Chi-Square test of independence, Mann-Whitney U, Kruskal Wallis and Gamma coefficient (Beatty, 2018; Verma, 2012). For the hypothesis tests, we present the observed results for the test statistics and the corresponding p-value, the latter value being presented in parentheses. We considered the decision rule that establishes the rejection of the null hypothesis if the p-value is less than or equal to the significance level 5%. The statistical analysis was done using IBM® SPSS® (IBM, 2010), version 19.
4. Results
4.1 Qualitative Data
4.1.1 Diversity of emotional reactions to guerrilla campaigns
This study provides evidence of the various emotional reactions guerrilla marketing produces in consumers. Participants in the focus groups showed positive and negative emotions regarding the two campaigns. Some participants classified both campaigns as fun, amusing, and humorous, while others considered them offensive and shocking. Sagres' campaign was particularly amusing for Benfica supporters (the team sponsored by Sagres and a rival team of Sporting), while the Sporting's fans found the advertisement offensive. As one of the participants suggested, "I think they are humiliating Sporting (...) this was very vulgar" (FG1).
Similarly, in the case of the Dexter campaign, the participants who found it fun were those who usually watch the TV series or at least were familiar with it. One example from the dialogue in FG4:
- As a fan of the series, I'm amused by [the campaign], but if my mother went to that butcher shop, she would not be happy at all! And she would not want to watch Dexter on TV.
- I liked it too, but I know what the series is about. Obviously, someone who is not familiar with it will be frightened. Oh, the old ladies in the butcher, poor thing!"
Additionally, the participants that considered the Dexter campaign scary and shocking were mostly individuals who had never watched the TV series and that do not like shocking and violent TV shows. Some opinions shared included disliking being exposed to the campaign, feeling scared ("I would not like to be in the shoes of anyone who was caught by that campaign, it would have scared me" FG3), and feeling offended ("anyone who would not know what that was about would feel scared and scandalized" FG3).
Thus, the emotions generated depended on the profile of each participant, namely the involvement they had with the brands and topics associated with the campaign. For instance, participants that are not interested in football said that they would not pay attention to the Sagres campaign since they felt neither involvement nor identification with the subject. Dexter's campaign generated similar results, although the shocking nature of the images raised attention. Still, it would generate a greater impact on consumers more involved with the brand and the campaign's theme.
4.1.2 Factors influencing online sharing
Both lack of involvement and negative emotions would result in the absence of interaction with the guerrilla content. Participants unanimously considered that they would not share or interact in any other way with the content in such instances. As summarized in FG1 "What you share is victory and when you feel good. Because if you are losing or having negative feelings about something, you don't share". Regarding comments, only some participants that considered the campaign offensive would eventually leave comments, "If I saw any offensive remarks, I would state my point of view. I'm not being a fundamentalist, but I'm pointing to the facts" (FG1). Sharing behaviour also depends on the novelty of the campaign, as stated by one participant in FG4: "I have no interest in sharing, it is very rare that I do so. And most of all, I do not share what has already been shared. I only put content that I have not yet seen".
Hence, sharing guerrilla content on SNS depended on four cumulative factors: having an active presence on Facebook, associating positive emotions to the campaign, being interested in the topic (product, service, brand, communication theme), and the content being relevant for their peer network. Participants stressed that what they share depends on the interests of their networks, "My group of friends is typically Benfica supporters, so in this case, I would probably share it just for fun (FG2). It also depends on their identity and the self-image they project on the SNS, as explained by one participant: "I would not share it because it has not connected with who I am (FG3).
4.1.3 Impact on brand image
Both campaigns analyzed by focus groups' participants seem particularly effective in reinforcing involvement with customers and less able to attract new customers. The participants with some familiarity with the brand and topic showed more interest in the campaign, were more willing to interact with it, were more able to understand the inner meanings of the guerrilla campaign, and consequently enjoyed it more. For instance, amongst the participants that felt positive emotions after seeing Sagres' campaign, some recognized an increased involvement with the beer brand and were willing to defend it. One example was provided during FG1:
The video motivated me to drink a Sagres beer that day. Not because of the video itself, but because of the reactions of Sporting fans, as soon as I read their comments saying "I'll never drink Sagres again", I had to drink one Sagres beer!
The same applies to the Dexter campaign, according to the participants in the focus groups. The campaign was particularly effective in increasing fans' interaction and making them want to watch the second season of the TV series. As one participant puts it "I think it has a great impact on the people that usually watch the show (…) it impacts fans, if I saw this, I'd be super happy" (FG4).
4.1.4 Summary of findings
This qualitative exploratory research provides some interesting cues on the impact of guerrilla campaigns on consumers. First, there is no unanimity regarding a guerrilla approach generating negative emotions. For some consumers, especially those with greater brand involvement, shocking campaigns might be regarded as amusing, as happened with the two cases used for the focus groups. Moreover, according to the participants' reactions and opinions, humour is shown as one of the most effective characteristics of guerrilla campaigns.
Considering that the creative, surprising, and unconventional nature of guerrilla content is harder to be immediately understood by consumers, it is more appealing to actual customers, namely consumers with some involvement with the brand, at least by using unexpected approaches, as it was the case of Dexter campaign. It is also interesting to note that if there is coherence between the campaign motto and the brand or product being advertised, as it was the case in the Dexter campaign, it should be effective in attracting prospects if positive emotional reactions occurred; individuals that did not know the TV series and found the campaign amusing would become interested in watching it. In such cases, the guerrilla campaign can improve brand awareness and future consumption intentions.
Also, guerrilla campaigns may call attention and create emotional reactions, but the impact is not reflected in interaction with guerrilla content on SNS, especially in regards to sharing. This study demonstrates that the surprise effect triggers attention but may not result in online word-of-mouth communication. Since the success of guerrilla campaigns depends on target individuals' collaboration by helping to disseminate the campaign, this study also shows that only a few will share some campaign-related content on SNS. Campaigns must trigger positive emotional reactions in individuals that are frequent users of SNS that regularly share content with their network; the campaign (brand, motto, characteristics) must be adequate to one's self-image and considered interesting, relevant, and timely for their peers.
4.2 Quantitative Data
4.2.1 Sample characteristics
The respondents presented homogeneous ages and varied around the observed mean age, which was 28.5 years, with a standard deviation corrected to 7.7 years. A higher frequency of responses was observed for the younger ones, with half the respondents aged between 23 and 32 years. The distribution of the respondents by gender is balanced, with 53% of the female gender. Most respondents were graduates.
Regarding Facebook usage, most respondents (83%) accessed Facebook several times a day, and the activities most frequently performed were reading friends' posts and messaging. The distribution of the respondents concerning the interaction with the guerrilla content on Facebook is quite balanced, observing that 48% of the users only visualized the publication, and 52% effectively interacted with it by liking, commenting, and sharing. Interestingly, male participants were the ones who most interacted with the campaign, as most female users only viewed the post.
4.2.2 Guerrilla content perceived characteristics
Regarding the opinion on the image published by Friends Aveiro brand, most respondents (52%) considered the guerrilla content fun, 50% of the respondents agreed that the post was interesting, 57% considered it controversial, 46% considered it not offensive, and 59% considered that the publication was different. Hence, none of the characteristics was unanimously perceived by the participants in this study.
Kruskal Wallis and Mann-Whitney U tests show that the characteristics perceived in the content do not differ significantly with age. However, the tendency to consider that the content of the publication is fun, interesting, different and offensive differs significantly between the two genres. Only for the controversial classification no significant differences among gender were detected (Table 2).
Table 2 Perceived characteristics of the content according to Facebook user profile: age and gender
| Funny | Interesting | Controversial | Offensive | Different | |
| Kruskal Wallis ª | 3,9 (0,4) | 1,8 (0,8) | 3,3 (0,5) | 3,7 (0,4) | 7,3 (0,1) |
| a. groups: Age category | |||||
| Mann-Whitney Ub | 4506,0 (0,0) | 6089,5 (0,0) | 7639,5 (0,4) | 5638,0 (0,0) | 5795,5 (0,0) |
| b. groups: Gender |
4.2.3 Hypotheses' test
Regarding the association between the type of interaction with the publication and the characteristics of the guerrilla content, the Chi-square test and the Gamma coefficient allowed us to conclude that: i) there is a statistically significant tendency for the participants to interact more with the content they found more fun; ii) there is also a significant tendency to have less interaction with guerrilla content by those who consider it more offensive (Table 3).
Table 3 Type of interaction with the publication versus characteristics of the guerrilla content
| Funny | Offensive | |
| Chi-square | 240,67 (0,0) | 128,22 (0,0) |
| Gamma | 0,97 (0,0) | -0,54 (0,0) |
To test whether the perceived characteristics of the content had significant differences according to the type of interaction, the Mann-Whitney U-Test was used. It was concluded that, on average, the perception of the content as different, interesting, offensive and funny differs significantly between the participants who interacted and those who only viewed the publication. In contrast, for the controversial nature of the post, no significant differences were recorded (Table 4). This supports RH2, showing that fun content generates more interaction while offensive content is associated with lower interaction by users.
Table 4 Agreement with the characteristics of the content according to the type of interaction with the publication
| Funny | Interesting | Controversial | Offensive | Different | |
| Mann-Whitney Uª | 366,0 (0,0) | 2430,0 (0,0) | 7788,0 (0,5) | 4328,0 (0,0) | 3700,5 (0,0) |
| a. groups: Interaction |
The study also provides some results that support RH3. The chi-square test leads to the rejection of independence between considering the fun guerrilla content and the image of the brand, and there is even a statistically significant and positive association (Gamma coefficient), which means that there is a tendency to consider guerrilla content to be more fun by users that have a more positive brand image. The same tests allowed concluding that there is an association between perceiving the content as offensive and the brand's public image. However, there is a statistically significant association only for the sophisticated image. As this association is negative, we conclude there is a greater tendency to consider guerrilla content offensive to users who find the brand less sophisticated (Table 5).
Table 5 Characteristics of guerrilla content versus brand image
| Modern | Sincere | Innovative | Competent | Fun | Unexpensive | Diversified | Enthusiastic | Sophisticated | ||
| Funny | Chi-square | 190,83 (0,0) | 109,45 (0,0) | 180,33 (0,0) | 125,42 (0,0) | 264,90 (0,0) | 131,19 (0,0) | 177,55 (0,0) | 161,93 (0,0) | 88,87 (0,0) |
| 0,27 (0,0) | 0,30 (0,0) | 0,26 (0,0) | 0,31 (0,0) | 0,24 (0,0) | 0,28 (0,0) | 0,28 (0,0) | 0,27 (0,0) | 0,37 (0,0) | ||
| Offensive | Chi-square | 83,47 (0,0) | 107,62 (0,0) | 110,44 (0,0) | 120,88 (0,0) | 185,24 (0,0) | 107,12 (0,0) | 135,83 (0,0) | 131,26 (0,0) | 113,02 (0,0) |
| Gamma | 0,03 (0,6) | -0,07 (0,2) | 0,07 (0,2) | -0,10 (0,1) | -0,01 (0,9) | 0,05 (0,5) | 0,04 (0,5) | -0,03 (0,6) | -0,25 (0,0) |
Moreover, there is a positive and significant association between the frequency with which Facebook activities are carried out, and the interaction with guerrilla content since the Gamma coefficient gives positive values for all activities. It is concluded that there is a tendency for greater interaction with guerrilla content by users who carry out the mentioned activities more frequently (Table 6).
Table 6 Facebook activities versus interaction with guerrilla content
| Read | Write | Share | Messages | Videos | Photos | Comment | Groups | Links | |
| Chi-square | 28,04 (0,0) | 86,91 (0,0) | 59,43 (0,0) | 18,59 (0,0) | 66,33 (0,0) | 65,69 (0,0) | 38,48 (0,0) | 70,63 (0,0) | 82,01 (0,0) |
| Gamma | 0,56 (0,0) | 0,43 (0,0) | 0,49 (0,0) | 0,54 (0,0) | 0,43 (0,0) | 0,45 (0,0) | 0,42 (0,0) | 0,46 (0,0) | 0,42 (0,0) |
For some aspects inherent to the brand's image, there is a positive and statistically significant association with the interaction with guerrilla content on Facebook. There were positive values for the Gamma coefficient, accompanied by p-values of less than 5% for sincere, competent, inexpensive, diversified, enthusiastic and sophisticated, indicating that there is a tendency for more interaction with guerrilla content by users who recognized these aspects in the brand image. In contrast, the modern, innovative and fun aspects do not present a statistically significant association with interaction with the guerrilla campaign on Facebook (Table 7).
Table 7 Brand image versus interaction with guerrilla content
| Modern | Sincere | Innovative | Competent | Fun | Inexpensive | Diversified | Enthusiastic | Sophisticated | |
| Chi-square | 102,18 (0,0) | 90,46 (0,0) | 106,99 (0,0) | 92,88 (0,0) | 149,38 (0,0) | 81,58 (0,0) | 100,40 (0,0) | 112,5 (0,0) | 64,23 (0,0) |
| Gamma | 0,17 (0,1) | 0,29 (0,0) | 0,14 (0,1) | 0,33 (0,0) | 0,07 (0,4) | 0,26 (0,0) | 0,17 (0,0) | 0,2 (0,0) | 0,46 (0,0) |
The association between the Facebook user profile and the interaction with the guerrilla content was studied, and the conclusions are not identical for all characteristics of the profile. It has been found that Facebook's intensity of use is positively and significantly associated with the interaction with guerrilla content, showing that the interaction with guerrilla content tends to be greater for those who use Facebook more often, providing empirical support to RH1. There is a statistically significant association between interaction with guerrilla content on Facebook and the gender of users. However, there is no significant association between the academic qualifications or between the age of Facebook users and the interaction with guerrilla content on Facebook (Table 8).
4.2.4 Synthesis of the quantitative results
The results show that there is some balance between the number of users who interacted with the guerrilla content published on Facebook (134 users) and those who only viewed the publication (122 users). For the sociodemographic characteristics of the respondents, there was some variability regarding the age and academic qualifications of respondents. However, the tendency is to be younger and have academic qualifications at the level of higher education. As for gender, there was a balance between male and female users.
Analyzing the results makes it possible to draw some conclusions regarding the hypotheses considered relevant to the study. The guerrilla content is shown as a decisive factor in the type of interaction. As for the characteristics of guerrilla content, there was a tendency for users who found the content to be more entertaining to interact with it more. In the case of users who have characterized the publication as offensive, there is a tendency to only visualize the content (RH2). For the age and academic qualifications, it was verified that these had no significant association with the type of interaction with the guerrilla content on Facebook, but for the variable gender, it has been proved it is significantly associated with the type of interaction. Users who considered guerrilla content fun tend to have a positive brand image. In contrast, users who consider the content offensive tend to have a negative brand image (RH3). It is also concluded that the individuals that access more frequently to Facebook present a greater tendency to interact with guerrilla content (RH1). It was also verified that the frequency with which the activities are carried out during each session is significantly associated with the type of interaction with the content.
It is concluded that the characteristics of the guerrilla content, as well as Facebook user habits, are associated with user interaction with guerrilla content on Facebook. It also resulted that the characteristics of guerrilla content are associated with the image that the consumers create in relation to the brand. Finally, it was confirmed that user interaction with guerrilla content on Facebook is associated with the image that the consumers create in relation to the brand.
5. Conclusion
Despite the relevance of guerrilla marketing strategy in overcrowded communication media in online settings, the number of studies exploring the success of guerrilla marketing approaches is still limited. This study aimed to provide contributions that help fill this gap. For that, we have defined two research questions, the first regarding the factors that influence the interaction with guerrilla marketing content on social network sites (RQ1), and the second referring to its impact on the brand image (RQ2). This article provides contributions relevant to academics and practitioners interested in guerrilla campaigns.
Our results show that the interaction with guerrilla campaigns on social network sites will depend on several factors (RQ1), including users' characteristics. One aspect that stands out is that consumers who use social network sites more frequently are more willing to interact with guerrilla content. It was also found that the type of activities performed during each session is significantly associated with the type of interaction with the guerrilla content. The relationship with the brand, content perceived characteristics, and peers' needs and profiles are other factors that explain the interaction with guerrilla content.
Guerrilla communication strategy is built on surprise, diffusion, and low cost (e.g., Hutter, 2015; Hutter & Hoffmann, 2011; Sajoy, 2013), often depending on the collaboration of their targeted consumers through word-of-mouth to increase its reach and achieve success. Guerrilla campaigns are expected to stand out and grab consumers' attention due to their unconventional and unexpected nature, leading them to talk about it with other consumers and consequently help reach a bigger audience (Levinson, 1984). In this regard, two aspects are of utmost importance for guerrilla campaigns: the impact it has on the members of the target audience that have contact with the campaign and their willingness to spread the word, as these campaigns usually rely on consumer collaboration to reach a larger audience. Overall, this article demonstrates the difficulty in accomplishing targets' collaboration with the guerrilla campaigns on social network sites, being the major challenge in generating likes, comments, and especially shares. Again, while shock and surprise are important to stand out, success will depend markedly on the positive impact on consumers, who are clearly more prone to interact with funny content and often refuse to interact with provocative content that may be interpreted negatively.
One aspect that stands out in both the qualitative and the quantitative phases of this research is the fact that consumers perceive guerrilla content characteristics differently, and while some might consider a particular campaign offensive, others will find it amusing. This adds risk to conducting guerrilla campaigns, especially the ones intended to shock their target audience, and deserves the careful attention of marketers preparing guerrilla approaches.
Another conclusion particularly relevant for practitioners is the fact that guerrilla campaigns are easily understood by consumers with some degree of involvement with the brand, thus showing greater conversion potential amongst former clients than within new ones. While Hutter and Hoffmann (2011) are among the authors that stressed that guerrilla marketing is associated with brand image, our study adds evidence that, with an unknown brand, guerrilla content perceived characteristics (e.g., fun, offensive) affect brand image (RQ2). Hence, the qualitative study showed that consumers attached to the brand often found guerrilla content amusing and were more willing to interact with it, while the quantitative study demonstrated that positive emotions were associated with a more favourable brand image. About the impact that guerrilla content can have on the brand image, it has been found that users who find guerrilla content entertaining tend to have a positive brand image. In contrast, users who consider the content offensive tend to have a negative brand image.
Despite the interesting cues provided, this study is not without limitations, one of which refers to the fictitious brand that was used for the quantitative research. Moreover, the sampling method prevents the extrapolation of results. Considering the exploratory nature of this study and the interesting results obtained, it is recommended that future studies replicate the tests using different samples and well-known brands to validate these results and thus help practitioners to develop more effective guerrilla marketing strategies. Other suggestions include studying guerrilla marketing in business-to-business contexts.
