Abstract
Resource branding has become a cocreated process carried out within virtual brand communities (VBCs), which boost the social connectivity of users and contribute to placemaking. However, due to the very nature of VBCs, branding strategies rely on user engagement to succeed. The present study has performed a multidimensional and mixed-method analysis of almost 4,000 entries in the Facebook and Twitter VBCs of two renowned European music festivals, Sónar and Primavera Sound (Barcelona), to show the benefits of understanding the different components that comprise festival engagement and the relationships between them. Results show the importance of considering different festival brands and their main VBCs to develop a strategy based on the structure of each VBC user as well as the most prominent dimensions of user engagement and the inherent relationships between them. In addition, the consideration of “ongoing” engagement is a potential research pathway for exploring brand engagement beyond mere attendance.
Introduction
Festivals have been recognized as having relevance far beyond purely economic impacts and are considered to be one of the most important contemporary forms of social and cultural participation, whose performative character is conducive to people meeting and relating to each other (Giovanardi, Lucarelli, & Decosta, 2014; Morales Pérez & Pacheco Bernal, 2018). In addition, their importance for leisure and tourism is undisputed. Festivals are currently one of the main draws for tourism destinations, and several authors (Pechlaner, Bò, & Pichler, 2013) have indicated that this combination of the destination and festival brand engagement influences the decisions made by tourists, as well as their level of satisfaction. Indeed, festivals have been considered to be one of the main tools for repositioning a range of destinations (Stipanovic, Rudan, & Peršin, 2015) and academics have analyzed the potential benefits of the associations of festival branding in these contexts (Dreyer & Slabbert, 2012).
However, festival branding has undergone a huge transformation in recent years, led by several trends linked to the impact of the Internet, especially those related to social media (SM). Getz and Page (2016) highlighted, among other things, “performative leisure” (where people celebrate their involvement in events and enjoyment by sharing their experiences via SM) and the “accumulation of social capital” (where consumers want to celebrate their achievements and participation in key events and festivals and this is part of the desire to accumulate experiences for their social capital repertoire).
These trends are in continuous development because SM platforms are used to create a range of virtual communities around the festival brands—the festivals’ VBCs. These communities are transforming the traditional relationships between festival managers and organizers and other users, empowering the latter (Hvass & Munar, 2012) and enabling their engagement with the festival, other members of the community and the destination. The emerging picture indicates that these VBCs are producing new forms of festival branding as a result of the specific characteristics of SM platforms in terms of communicating emotions (affective variables), information (cognitive variables; Hudson & Hudson, 2013) and different forms of engagement behavior of users.
Moreover, and following Getz and Page (2016), it comes as no surprise that user communication through these VBCs goes beyond a simplistic consideration of festival branding since it also reflects users’ needs to use them to generate potentially valuable social capital. Neither is it surprising that these VBCs enhance the close relationship between the projection of the festival and that of the destination brand through placemaking (Richards, 2015). Finally, sufficient importance should also be given to the exact moment in which the communication in these VBCs occurs (MacKay, Barbe, Van Winkle, & Halpenny, 2017).
As such, if VBCs can be considered to have facilitated a greater diversification of the cocreation of festival narratives (Gyimóthy & Larson, 2015), including branding through user-generated content (UGC), they arguably represent a unique marketing opportunity for festival organizers and destination marketing organizations (DMOs). Taking this into account, this research characterizes VBCs in the context of festivals and considers the different dimensions of their engagement with the festival brand, the moment in which this engagement is constructed and the profile of the users leading these processes.
Background Literature
Contextualizing Festivals’ VBC and User Drivers for Participating in Them
Developments in ICT have dramatically transformed the operational and strategic practices of tourism organizations and have had an impact on their competitiveness around the world (Buhalis & Law, 2008). The Internet has enabled tourism businesses expand their customer base, with most of the transformations based on the particular way that the relationship between the organization and their customers has developed. Nowadays, SM platforms are leading this development by allowing the formation of virtual communities based on the creation and exchange of UGC. Within this framework, stakeholders—whose profile is becoming increasingly differentiated and who come from a growing variety of SM-based communities—are constantly cocreating narratives and transforming place branding (Kavaratzis & Hatch, 2013). This produces a relatively fragmented image over which DMOs have no control in the traditional sense, making it necessary to build a feeling of belonging to a community. Place branding thus becomes a point of reference for content creation and for community-building in particular.
Festival organizers are also increasingly using SM to generate VBCs as a way of managing their relationships with other users to foster a stronger sense of belonging to the festival brand (Montanari, Scapolan, & Codeluppi, 2013). In fact, this motive has been detected in different VBCs in terms of event organizers’ necessity to improve brand promotion, brand acknowledgement and experience and brand loyalty and satisfaction (MacKay et al., 2017). Moreover, while implementing an SM strategy does not guarantee success, it does provide a clear opportunity for harnessing the marketing intelligence that is generated when people are being social and for using this social connectivity to increase attendee numbers (Lee, Xiong, & Hu, 2012; Marine, Martin, & Daries, 2017).
However, various authors have emphasized the need to understand this social interaction beyond the boundaries of economic impact and explore social, cultural, and spatial outcomes (Mair & Whitford, 2013). Social connectivity is one of the key characteristics of VBCs in these SM contexts, as users are socializing while capturing, displaying, retelling, and reliving engaging experiences related to the festival (Flinn & Frew, 2014). Along these lines, MacKay et al. (2017) revealed significant efforts to build relationships just before and during the festival. This is connected to the “accumulation of social capital” trend that Getz and Page (2016) highlighted, where participation in VBCs is celebrated as an experience and probably considered as part of the social capital repertoire. At this point, it is important to remember that social capital is one of the ingredients for constructing civic commitment and is an indicator of community health, since it allows the resource of one individual (in the case of VBCs, basically information) to be transformed into a collective attribute (Van Staveren & Knorringa, 2007).
In the tourism context, some interesting studies have demonstrated the capacity of tourism activities to stimulate the creation of social capital among users (Heimtun, 2007). In the festival context, Arcodia and Whitford (2007) showed that attendees developed social capital by supplying the community with specific opportunities for accessing and developing resources, improving social cohesion, and providing a focus for celebration. For Crespi-Vallbona and Richards (2007), festivals can enhance their bridging social capital by connecting with the world beyond the immediate community, while Curtis (2010) considered that festivals can engender the sense of belonging. Finkel (2010), meanwhile, connected social capital creation with the identification with a place and more recently, Wilks and Quinn (2016) proposed a heterotopian concept of festivals as a kind of sacred celebration.
However, festivals are also in the center of destination branding strategies (Richards & Wilson, 2004) that seek to transform cultural capital into a competitive advantage. For Prentice and Andersen (2003), they are crucial to transform a destination’s brand or even modify the image of an entire region, with a decisive contribution to placemaking (Richards, 2015). Therefore, if VBCs represent one of the main avenues for publicizing festivals (and hence destinations), this aspect should also be considered by festival organizers and DMOs.
Going a step further, literature focusing on festival branding has often mentioned the importance of user engagement. However, in most cases this engagement is hard to characterize. In fact, there has been little analysis of how users engage with festivals, or with the territory or community (Garay & Morales Pérez, 2017), in the particular spaces formed in SM, which have now become the most important places in terms of festival communication strategies. Therefore, in the festival context, there is a knowledge gap in relation to engagement characteristics of VBCs created in SM. However, other areas and research topics have considered this question and next we are going to observe how their findings may be useful for making future proposals.
How Community Users Act and Interact: User Engagement in Festival VBCs
Consumer engagement has attracted much attention recently due to its relationship with different indicators that measure business results (Harrigan, Evers, Miles, & Daly, 2017; Hollebeek, Glynn, & Brodie, 2014; Moliner, Monferrer-Tirado, & Estrada-Guillén, 2018). Originating in the public relations domain, the idea of user engagement has looked at the interaction between consumers and brands. In their seminal work, Brodie, Hollebeek, Jurić, and Ilić (2011) affirmed that consumer engagement is a multidimensional concept comprising cognitive, emotional, and/or behavioral dimensions, and plays a central role in the process of relational exchange where other relational concepts are engagement antecedents and/or consequences in iterative engagement processes within the brand community.
For these authors, consumer engagement has a high component of interactivity and is based on several behavioral manifestations, including learning, sharing, advocating, socializing, and codeveloping, that can lead to many performance indicators, highlighting loyalty and satisfaction.
In their exhaustive literature compilation, Hollebeek et al. (2014) demonstrated the multidimensional character of engagement conceptualization in the marketing literature, highlighting the emotional (affective), cognitive, and behavioral dimensions. From this review, they propose that in a consumer/brand interaction, the affective attributes represent the consumers’ degree of positive brand-related affect and the cognitive attributes are related to the consumers’ level of brand-related thought processing. Moreover, they also add the behavioral dimension, called “activation,” which is defined as a consumer’s level of energy, effort and time spent on a brand in a consumer/brand interaction. Meanwhile, Dessart, Veloutsou, and Morgan-Thomas (2015) identified three key engagement dimensions (cognition, affect, and behaviors) but, more interestingly, they also proposed two foci of this engagement: brand focus and community focus.
Customer engagement has become a widely discussed topic in tourism and hospitality literature, focusing on its conceptualization and measurement, particularly through the search for its antecedents and consequences. The first attempts to shed light on these aspects can be found in the work of Choi and Sirakaya (2005) and Kim, Ritchie, and McCormick (2012), who proposed comprehensive scales for measuring attitudes toward different tourism products and experiences. Consistent with these proposals but also mapped, largely, onto the customer brand engagement dimensions offered by Hollebeek et al. (2014), a new reference by So, King, and Sparks (2014) provided a scale that can be used to examine the effects of customer engagement on various different key consumer behavior outcomes. Of particular relevance are the five constructs that they proposed as customer engagement antecedents for tourism brands, namely “identification,” “enthusiasm,” “attention,” “absorption,” and “interaction.”
Only 2 years after their seminal article, So, King, Sparks, and Wang (2016) proposed an integrative model for hotel and airline brand managers, which added “service brand evaluation” to customer engagement to make up brand trust and ultimately affect brand loyalty. More recently, Ahn and Back (2018) deconstructed engagement into three dimensions based on traditional image attributes (cognitive, affective, and behavioral). They proposed four antecedent constructs, namely sensory, affective, behavioral, and intellectual experiences, proving engagement’s influence on behavioral intention.
In connection with these references, tourism and hospitality literature focusing on customer engagement has also given the SM context a prominent role. In this regard, Harrigan et al. (2017) tested the dimensions proposed by So et al. (2014) within the context of tourism brand SM to confirm the previous assumption (Hudson & Hudson, 2013) that customer involvement leads to customer engagement, which in turn leads to brand loyalty. Going beyond loyalty, Dijkmans, Kerkhof, and Beukeboom (2015) showed whether and when a company’s online activities to acquire engaged consumers have a positive impact on another aspect: corporate reputation.
Interestingly, Cabiddu, De Carlo, and Piccoli (2014) adopted an affordance perspective that led them to identify three distinctive SM affordances for customer engagement in tourism. “Persistent engagement” represents the possibility of maintaining an ongoing dialogue with customers, even when they are not physically at the property; “customized engagement” represents the possibility of interacting with customers based on prior knowledge of individual-level information; and “triggered engagement” represents the possibility of instigating customer encounters based on an external, customer-initiated event.
For Sigala (2018), the collaborative activities afforded by SM in the tourism industry move customer relationship management (CRM) toward a process of engaging, rather than managing, people in the cocreation of value within online communities/ecosystems involving many different stakeholders (Trunfio & Della Lucia, 2018). Definitively, today’s SM networks have the capacity to engage tourists with their peers on a global level, influencing their consumption on a large scale (Tussyadiah, Kausar, & Soesilo, 2018) and having an impact on brand loyalty, especially regarding (passive) SM engagement and affective loyalty (van Asperen, de Rooij, & Dijkmans, 2018).
However, as previously mentioned, engagement in SM and related VBCs has not been prevalent in the festivals literature (MacKay et al., 2017). SM has been seen more as a promotional tool and there has been an overemphasis on pre-event research (Leung, Law, Van Hoof, & Buhalis, 2013), overlooking their fundamental essence, which is to facilitate interaction and engage consumers precisely through the creation of VBCs. Nevertheless, authors such as Flinn and Frew (2014) have considered that engagement in SM across all phases of festival production and consumption benefit both the organizers and the attendees, as their experiences are constantly captured, displayed, retold and relived in these spaces.
More recently, MacKay et al. (2017) offered a seminal analysis of the scope and degree of engagement on festival SM pages through a variety of basic indicators (retweets and likes; hashtags; @mentions; links and types of link), finding that engagement reached its highest levels before and after the festival and suggesting that organizers manage engagement to improve positive festival reputation. Laurell and Björner (2018) have recently demonstrated that festivals enjoy considerable levels of digital engagement and that this also generates engagement for their venues. Finally, in a recent study that precedes the present one (Garay & Morales Pérez, 2019), we analyzed user engagement in the virtual brand community (VBC) of a particular festival, Sónar, evidencing how its stakeholder structure influenced the prevalent attributes, behaviors, and focuses of this engagement. The study also found diverse associations between these elements, highlighting the relevance of sociality and the importance for the festival to maintain an ongoing communication with these stakeholders.
In relation to this background and our original objective, we formulated the following research questions:
To answer these questions, this study uses a multidimensional and mixed-method analysis of the Sónar and Primavera Sound (hereinafter, “Primavera”) music festivals, two of the most prominent festivals in Europe in terms of attendees, impact on their venue (Barcelona) and international recognition.
Methodology and Data Analysis
A Multidimensional Characterization
The Sónar and Primavera festivals were selected for this study because of their considerable audience numbers and international projection. In 2018, Sónar and Primavera received 126,000 and 220,000 visitors, respectively, and are among the most highly attended festivals in Europe. In addition, they were chosen because of their strong link with and significance in Barcelona, one of the most important European tourist destinations. Sónar is an electronic and advanced music festival founded in 1994, which takes place both day (Sónar by Day) and night (Sónar by Night), and is accompanied by Sónar+D, a 3-day congress dedicated to creativity. Over its 25-year history, Sónar has contributed substantially to Barcelona’s cultural sector, but it also has a well-known international presence, having hosted events in 65 cities worldwide. All Sónar events feature a strong creative technological component.
Primavera was created in 2001 as a showcase for the international independent music scene, but today it includes a wide range of musical styles and has become a meeting point for artists and spectators, spanning several generations. The festival has undergone continuous growth in recent years, but one of its main objectives is to be recognized as an international brand. In fact, its worldwide projection has resulted in attendees from 134 countries and attracted the interest of numerous global brands, thereby increasing its sponsorship income.
Moreover, in addition to their importance in quantitative terms, the variety of experiences created around these events and their technological and innovative profile have also been considered. Both events have a long history of VBC presence on different SM platforms. As Leung et al. (2013) have indicated, most tourism literature (including festivals) has explicitly consider the analysis of a single platform, Facebook, while festivals’ are privileged contexts to observe the characteristics of VBC user engagement on various recognized SM platforms (Sónar has nearly 500,000 followers on Facebook and 150,000 on Twitter, while Primavera has nearly 360,000 followers on Facebook and almost 98,000 on Twitter).
This study involves a multiperiod analysis because, as Leung et al. (2013) revealed, few studies have addressed all the different periods of the tourism experience (festival experience included) and most only consider information and planning during the period before the service is provided. Nonetheless, more recently, different authors (Hudson & Hudson, 2013; MacKay et al., 2017) have finally incorporated a primary multiphase consideration, while others (Gyimóthy & Larson, 2015) have stressed the importance of ongoing communication to festival success. A vision of the “consumption” cycle of the festival is therefore necessary for a better understanding what its branding strategy may be over time.
Likewise, festival branding literature also indicates that this is a progressive negotiation process (María Munar, 2011) between different users with various drivers (Sevin, 2013). However, there is limited scholarly work examining these perspectives, or where it does exist, the stakeholder analysis is rather basic, using only rudimentary categories (MacKay et al., 2017). By using the Sónar and Primavera festivals, this study has allowed for a multiuser study, in addition to the multiplatform and multiperiod analysis. The data analysis has also followed a sequential and mixed-method strategy (Neuman & Robson, 2014).
Following Creswell and Plano Clark (2017) mixed-method proposals, we carried out an exploratory sequential design whereby the first step was to collect and analyze qualitative data (interpreting and coding tweets and posts). From this first step, we obtained qualitative data, which were reinterpreted and recodified to ultimately form the different dimensions of engagement (affective and cognitive attributes, behaviors and drivers). The next step was to quantify each dimension to transform qualitative data into quantitative data, and this data analysis then generated quantitative results.
Data Collection
Sampling was purpose-driven and aimed to obtain complete and accurate information about conversations in VBCs managed by the festival organizers in their official accounts. We have not used the festivals’ hashtags, as most social platform APIs (application programming interfaces) do not allow the capture of hashtags or subject conversations more than 1 month old, but they do permit a lengthy capture of a single user’s conversation. In this context, it is festival organizers who mainly manage VBC conversations (Garay & Morales Pérez, 2017), posting and tweeting new content and responding to and retweeting other users’ contributions. Three weekly captures were made during a period that began 9 months before the 2016 edition and continued until a month afterward, beginning in October 2015 and finishing in July 2016 in the case of Sónar, and from July 2015 until July 2016 for Primavera.
More than 38,000 entries (5,007 tweets and retweets and 6,992 comments and posts in Sónar’s Twitter and Facebook VBCs and 975 tweets and retweets and 25,342 comments and posts in Primavera’s Twitter and Facebook VBCs) were captured and collected in XLS using the web scraping tool Ncapture® and were found to be suitable for data analysis using the software NVivo10®.
Modes of Analysis
Before streamlining the databases for each VBC and festival, information was divided into different time periods: “Ongoing,” including all the data captured between 11 months and 1 month before the festivals’ edition; “Before,” starting a month before until the beginning of the event; “During,” capturing the data released the days the event was held; and “After,” with all the data captured from the day after the event ended until 1 month later. From these time-period categories, a random sample with over 4,000 time-period entries was created (see Tables 3 to 6) that met the criteria of a 5% sample error, a 95% confidence level, and a distribution of responses of 50%.
The next step was to categorize each entry (tweet, retweet, post, and comment) into the user categories reported in recent specialized literature (Van Niekerk & Getz, 2016). For this purpose, the profiles of the users and the content of each entry were analyzed, finding differences between the two festivals and their respective VBCs. Then, we proceeded to undertake a new content analysis for each VBC (Twitter and Facebook), which consisted in coding each entry in accordance with an initial proposal for a coding map based on existing literature on VBC users’ engagement attributes, behaviors, and drivers.
Numerous references (see Tables 1 and 2) guided a progressively developed, more inclusive and differentiated conceptual framework for capturing and organizing categories. The resulting and definitive coding map includes some of the affective and cognitive engagement attributes and behaviors that are reported in the literature, although these have been adapted and new categories and grades have been proposed for some of them. Regarding these categories and grades, we found that while on Twitter users mainly projected affectively positive messages, some of the user messages in Facebook were clearly negative. This could be related to the disengagement category proposed by Brodie et al. (2011) but in this case explicitly manifests an affective discontent rather than a passive disengagement.
Coding Map: Engagement Affective and Cognition Attributes
Source: Authors’ own work.
Coding Map: Behavioral Manifestations of Engagement and Conversational Drivers
Source: Authors’ own work.
Moreover, while the two positive dimensions of affection found (excited and pleased) could be close to those found in the previous literature, our analysis indicated that there were a great number of messages that did not express any affect and thus were emotionally passive. In terms of the cognitive attributes, we also found a possible graduation of cognitive messages, ranging from those that were passive in this sense to others that reflected a real dedication to expressing something (see, e.g., posts in Table 1). Also, the initially proposed uninterested code/category (adapted from the disengagement of Brodie et al. [2011]), although potentially present in other kinds of VBCs, did not appear in our samples.
Regarding engagement behaviors, the content analysis provided a varied panorama, from simple forms of conduct where users were asking for or sharing information, to more elaborate behaviors involving the evaluation of someone or something. Probably the most important finding here relates to the importance of dialogue, as identified in some previous works (So et al., 2014), in the form of socialization or interaction and which, on most occasions, may represent an advanced manifestation of community engagement as it necessarily implies a call to other users. Finally, in relation to engagement drivers, the qualitative analysis validated not only the presence and relevance of the festival’s branding but also the motivation to socialize by mentioning or interacting with other users. Furthermore, although in a more marginal position, the content analysis indicated the need to consider the placemaking driver, as the descriptions of places were explicit and likely to relate to other variables.
In addition to detecting which users were intervening in each period, this qualitative interpretation enabled a quantitative joint analysis of all these categories, as each tweet, retweet, post, and comment was coded with one of the definitive affective and cognitive engagement attributes, as well as with one or several engagement behaviors, and drivers (assigning a 1 if the subdimension was present and a 0 if it was not). Therefore, by quantifying all of these elements, we obtained the quantitative results (descriptions and associations) discussed in the sections to follow.
Results
VBC Stakeholder Structure
Figures 1 and 2 show for both festivals that Twitter VBC is managed by the festival organizer but with considerable participation from music/cultural sector agents and with little or no participation from other users, meanwhile the Facebook VBC comprised event attendees and other Facebook users (what we called “general users”). On Facebook, although the organizer started the conversation through its posts, its leading role tends to dissipate as general users took over the conversation. The VBC stakeholder quantitative composition therefore indicates a two-pronged communicative strategy: while Twitter is dedicated to communication with music/cultural sector agents, Facebook is managed more to engage attendees and other general users interested in the event.
Sónar Festival: Stakeholder Categories by Period and VBC: Contribution Percentages
Primavera Sound Festival: Stakeholder Categories by Period and VBC: Contribution Percentages
Moreover, the first result of coding subdimensions by periods is that the organizers of both festivals are in continuous contact with the different stakeholders through their different VBCs and they have a direct impact on the engagement process. Additionally, there are no important temporal differences in the relative weight of each stakeholder with two exceptions. In Sónar’s Twitter VBC, while the relative weight of the festival manager increases when the event has finished and during the rest of the year, musicians acquire more relative importance in the conversation during the event. However, in Primavera’s Twitter VBC, the musicians increase their prominence during the rest of the year while festival managers decrease their relative participation, this being especially high just after the event took place. This is related to the importance that Sónar places on the events it promotes in other periods of the year, while Primavera makes good use of all the material (especially images) produced during the festival to maintain engagement during the weeks after it takes place.
Engagement Analysis: Affective and Cognitive Engagement Attributes
Beginning with the affective and cognitive attributes (see Tables 3 and 4), the VBCs for both festivals reveal an emotionally passive conversation, although it is also interesting to mention the appearance of disaffection in both Facebook VBCs and more affective intensity in both Twitter VBCs. In the case of Facebook, general users intervene more spontaneously in the conversation, while the festival organizer has a minimal leadership role. In the case of Twitter, where the music/cultural sector performs a most significant role, affection has more space in short (and especially visual) tweets, while negative or null subdimensions were absent. Another interesting element to note is that as the festival date approaches, the relative weight of the most passive affective attributes diminishes, while the most critical visions—also the most affectively engaged—increase. In relation to cognitive attributes, in both Facebook VBCs cognitively passive conversations are also predominant, but both festivals’ Twitter VBCs also use more attentive engagement, including a relatively high weight of dedicated engagement within Sónar’s Twitter.
Sónar Festival VBCs: Engagement Affective and Cognitive Attributes by Period and VBC: Contributions (Tweets or Posts) and Percentages
Note: VBC = virtual brand community.
Source: Authors’ own work.
Primavera Sound Festival VBCs: Engagement Affective and Cognitive Attributes by Period and VBC: Contributions (Tweets or Posts) and Percentages
Note: VBC = virtual brand community.
Source: Authors’ own work.
If we look at developments over time, on the one hand, we can see how in the period between the annual events, disaffection decreases, while emotionality increases (especially in the “excited” degree) in most VBCs. On the other hand, in the period just before the event and for its duration, the emotionality degree decreases, while the cognitive level increases. With regard to cognitive attributes, the “attentive” degree in both festivals’ Twitter VBCs is even more important in the period after the event. Moreover, while Twitter is the channel for maintaining a more cognitively active conversation during the long period between each edition of the events, Facebook maintains a more cognitively passive character during this period.
Also important is that there is a significant association between affection and cognition attributes in all VBCs, but especially in Primavera’s Facebook VBC (χ2 = 243,795, p = .00), as we found more emotional content (in a negative or positive way) in contributions with a higher level of cognition. In the case of Primavera’s Facebook VBC, this is seen in messages sharing a disaffected opinion about the festivals or, on the contrary, a very positive attachment to them through “excited” and even more “pleased” affection.
Finally, regarding stakeholders, in both festivals’ VBCs, while organizers and media professionals are more affectively passive, musicians and other professionals are the most emotionally engaged. Regarding cognitive attributes, while in the Twitter VBCs most of the stakeholders displayed elevated degrees of cognition (with a minor level in the case of musicians), in the Facebook VBCs the predominant general users are more cognitively passive.
Engagement Analysis: Engagement Behaviors
Tables 5 and 6 show that in both festivals’ Twitter VBCs, the most usual behavior is sharing—100% in the case of Primavera and nearly 97% in the case of Sónar. In parallel, dialogue is the main behavioral aspect observed in both festivals’ Facebook VBCs, ranging from 75% of the contributions in the case of Sónar to 83% for Primavera. Moreover, while the sharing behavior on Twitter does not vary much throughout the different periods observed, the dialoguing behavior in Sónar’s Facebook VBCs takes on a more relative strength during the festival, while in Primavera’s Facebook VBCs, it does so before the event takes place. It is also important to note that the evaluative behavior has a secondary but relevant weight in both platforms, in particular during the “after” period, when users recalled the festivals highlights and shared their experiences during the performances. Asking other users’ questions is relatively marginal, although it reaches a more relevant weight in Primavera’s Facebook VBCs (10%).
Sónar Festival VBCs: Engagement Behaviors and Drivers by Period and VBC: Contributions (Tweets or Posts) and Percentages
Note: VBC = virtual brand community.
Source: Authors’ own work.
Primavera Sound Festival VBCs: Engagement Behaviors and Drivers by Period and VBC: Contributions (Tweets or Posts) and Percentages
Note: VBC = virtual brand community.
Source: Authors’ own work.
Analyzing the relationships between these behaviors and the affective and cognitive attributes, in Sónar’s Twitter VBC, we discovered a significant association between sharing behavior and the affectively passive (χ2 = 13.494, p = .01) and the cognitively pleased and dedicated attributes (χ2 = 11.394, p = .03). In Sónar’s Facebook, VBC dialoguing behavior has a positive association with the cognitively passive attributes (χ2 =115.585, p = .00), and in Primavera’s Facebook VBC, it is positively associated with the affectively (χ2 = 140.046, p = .00) and cognitively (χ2 = 40.042, p = .00) passive attributes.
In turn, evaluating behavior has a noteworthy association with higher levels of affection (excited) and cognition (attentive), as seen in Sónar’s Twitter VBC (χ2 = 100.728, p = .00 with the affective and χ2 =15.260, p = .00 with the cognitive), and Primavera’s Twitter VBC (χ2 = 40.714, p = .00 with the affective and χ2 = 6.900, p = .00 with the cognitive). In addition, the same occurs in Sónar’s Facebook VBC (χ2 = 318.240, p = .00 with the affective and χ2 = 69.316, p = .00 with the cognitive) and Primavera’s Facebook VBC (χ2 = 530.140, p = .00 with the affective and χ2 = 131.401, p = .00 with the cognitive).
Finally, with regard to the stakeholders, there are no significant differences in their sharing or dialoguing behaviors on Twitter, but there are in the evaluating behavior (χ2 = 16.887, p = .05) in which the festival organizer highlights certain points. However, in the Facebook VBC, we can find notable differences. Here, sharing behavior is significantly associated (χ2 = 501.672, p = .00 in Sónar and χ2 = 153.589, p = .00 in Primavera) with the festival organizers’ contributions, while the dialoguing behavior is particularly associated with the general users’ comments (χ2 = 244.313, p = .00) in Sónar. In Primavera’s Facebook VBC, dialoguing is associated with general users’ comments and with the festival organizers’ and other professionals’ comments (χ2 = 15.255, p = .05).
Engagement Analysis: Engagement Drivers
Turning attention to the engagement drivers, Tables 5 and 6 not only show how festival branding is the main focus of the conversation in both festivals’ VBCs, except in Sónar’s Facebook VBC, but also that social capital has a considerable relative weight (between 70% and 85% in these VBCs). Finally, placemaking occupies a secondary role, except in Primavera’s Twitter VBC, where it accounts for 25% of the sample. Regarding developments over time, it is possible to identify some trends from the significant associations yielded by these results. This is the case of festival branding, where there are significant associations with the period just after the event was held, for example, in the Sónar Twitter (χ2 = 22.900, p = .00), Sónar Facebook (χ2 =26.803, p = .00), or Primavera Twitter (χ2 =16.211, p = .00) VBC. Meanwhile, the social capital driver acquires more weight during the days the events are held, with significant associations found in the case of the Sónar Twitter (χ2 =7.996, p = .05), Sónar Facebook (χ2 =30.015, p = .00) and Primavera Facebook (χ2 = 95.446, p = .00) VBCs.
Considering the associations of the main drivers with the attributes, results show that festival branding is significantly associated with the affectively excited and pleased attributes in Facebook (χ2 = 278.759, p = .00 in Sónar and χ2 = 116.887, p = .00 in Primavera) and with the cognitively attentive and dedicated attributes in Twitter (χ2 = 30.612, p = .00 in Sónar and χ2 = 7.089, p = .02 in Primavera) and in Facebook (χ2 = 181.834, p = .00 in Sónar and χ2 = 157.492, p = .00). As for social capital, there are a few significant associations in the case of affective attributes, specifically with the affectively passive one in Facebook (χ2 = 101.972, p = .00 in Sónar and χ2 = 129.511, p = .00 in Primavera). Social capital is also associated with the attentive and dedicated attributes in Sónar’s Twitter (χ2 = 30.612, p = .00) and Facebook (χ2 = 6.981, p = .03) and a more passive cognition in Primavera’s Facebook (χ2 = 19.083, p = .00).
Results show also new relevant relationships between these drivers with engagement behaviors. One of the most important is the fact that festival branding is significantly associated with evaluating in both festivals’ Facebook VBCs (χ2 = 141.694, p = .00 in Sónar and χ2 = 94.736, p = .00 in Primavera) and in Sónar’s Twitter VBC (χ2 = 4.352, p = .03). It is also associated with sharing in both festivals’ Facebook VBCs (χ2 = 60.970, p = .00 in Sónar and χ2 = 27.183, p = .00 in Primavera). In turn, social capital correlates with the dialoguing behavior in both festivals and platforms (χ2 = 8.574, p = .03 in Sónar’s Twitter, χ2 = 159.417, p = .00 in Sónar’s Facebook, χ2 = 3.983 p = .04 in Primavera’s Twitter and χ2 = 254.306, p = .00 in Primavera’s Facebook).
In addition to these associations, the results show others appearing between the drivers themselves. In this context, although festival branding and social capital appear together in a high percentage of the tweets and post comments, they have a negative association because in the few cases where one of them does not appear, the other usually does. Therefore, we can find this kind of negative association in both festivals’ VBCs (χ2 = 254.306, p = .00 in Sónar’s Twitter, χ2 = 49.094, p = .00 in Sónar’s Facebook, χ2 = 16.408, p = .00 in Primavera’s Twitter and χ2 = 70.267, p = .00 in Primavera’s Facebook). Moreover, there is an especially interesting association between festival branding and placemaking in the Facebook VBCs (χ2 = 8.358, p = .00 in Sónar and χ2 = 4.135, p = .04 in Primavera).
Finally, the results also reveal different associations between these drivers and the stakeholders who emit them. In the case of festival branding, the results show associations with the festival manager in Sónar’s Facebook VBC (χ2 = 59.505, p = .00), while social capital is associated with the activity of online journalists and music professionals in Sónar’s Twitter VBC (χ2 = 35.172, p = .00) and with the festival manager’s participation in Sónar’s Facebook VBC (χ2 = 11.647, p = .00). Another interesting aspect is the finding that placemaking is associated with the activity of festival organizers in the case of Sónar’s Facebook (χ2 = 90.502, p = .00) and with the action of public organizations (χ2 = 27.222, p = .00) in Primavera’s Twitter and with Internet journalists in the case of Primavera’s Facebook (χ2 = 20.870, p = .00).
Discussion and Conclusions
This study characterizes the user engagement of various VBCs for two of the most prominent music festivals in Europe, offering several methodological and practical contributions. In relation to measuring engagement, our results complement and enrich previous quantitative approaches and provide a new, qualitative classification of VBC attributes, behaviors and drivers. In the case of attributes, this contribution is made by complementing previously reported subdimensions and proposing a scale or gradation, and for the behavioral and driver dimensions, by adding new subdimensions that allow for a broader understanding.
This study also offers an in-depth analysis of the relationships between engagement dimensions and consideration of the user profile in each VBC and which spans a whole year, thereby observing user engagement in periods long before and after the events were held. Regarding the first research question, the study strengthens the need to analyze stakeholders’ role in these processes (María Munar, 2011; Sevin, 2013) while also considering the stakeholder composition of the different festival VBCs. This is relevant because, as seen in the results, the dominant profiles condition the type of engagement characteristics appearing in each VBC and oblige organizers and other interested users to modulate their narrative appropriately and accordingly. In relation to our second research question and building on previous literature (Brodie et al., 2011; Dessart et al., 2015; Hollebeek et al., 2014; So et al., 2014), the results show how important considering different brands (in our case festivals) and their main VBCs is in understanding the need to develop an engagement strategy based on awareness of the VBC user structure and how engagement is constructed within each of these communities.
Cocreation processes are significantly affected by users’ engagement characteristics when considering different festivals and different VBCs. In this respect, the importance of stakeholder composition in the VBC characterization seems to guide the predominance of one attribute or another, especially regarding cognitive attributes. On Twitter, music professionals must be diplomatic in their relationship with the brand and need to explain their shared projects. On Facebook, generic users do not share this need and can express themselves in a more direct, dispassionate, and straightforward way.
The study has also shown the presence of a variety of engagement behaviors depending on the VBC profile. Regarding this, while on Twitter the stakeholders use this VBC for sharing information and with marketing/promotion motivations (Marine et al., 2017) in relation to the music industry, on Facebook, organizers still share information but general users demonstrate dialogic engagement and are at the heart of the socialization dynamics (Flinn & Frew, 2014). Moreover, while sharing and dialoguing behaviors are mostly related to passive affection and cognition, evaluation is associated with higher degrees of these attributes.
In relation to the engagement drivers, the study has not only confirmed the idea (Dessart et al., 2015) of the predominance of festival branding but has also shown the importance of social capital and the emergence of placemaking as other relevant drivers in these VBCs. In this context, Twitter serves to reinforce professional sector-based relationships and the need to manage resources in a more sustainable way. Facebook, by its side, contributes to the construction of civic commitment. This is consistent with the idea of Flinn and Frew (2014), whereby capital is now an experience captured in a narrative that is cocreated in a VBC.
Moreover, although placemaking appears as a secondary driver, its projection confirms the theory (Richards, 2015) that such important festivals contribute to placemaking, even generating engagement processes for places. The qualitative study has also shed light on this aspect and the underlying strategy therein. While Primavera, especially through Twitter, becomes a kind of loudspeaker for the city of Barcelona and its neighborhoods, Sónar, an internationalized festival is a global placemaker. However, albeit from different points of view and on different scales, the two festivals are active agents in the creation of placemaking and therefore give Barcelona an advantage in terms of the international competition between global cities.
In terms of these drivers, while on Facebook festival branding is constructed through higher degrees of affection and cognition and social capital through more passive ones, on Twitter both festivals show higher degrees of affection and cognition for both drivers. It is also interesting to highlight the fact that festival branding is basically constructed through the sharing and evaluation behaviors of festival organizers, while social capital is mainly constructed through the dialogic behavior of different stakeholders such as general users, professionals, and public organizations. Festival branding also has an interesting association with placemaking in the Facebook VBCs, indicating that festivals are creating a direct relationship with places and contributing to connecting places and their related images to festival attributes and related emotions, and vice versa, in a direct relationship, confirming some ideas advanced by previous literature (Laurell & Björner, 2018; Richards, 2015). Thanks to the VBC narratives, emotional connections are created with the place as a result of memories that are continually recalled in the period between festivals.
Finally, regarding the third research question, this study has analyzed a range of engaging experiences across all phases of festival production and consumption, adding a new period that covers an entire year between editions. This has been useful in differing from and extending the conclusions of previous literature (e.g., MacKay et al., 2017), as while the study does not show engagement peaks, it demonstrate how engagement dimensions and subdimensions are constructed and change over a whole year. Both Sónar and Primavera maintain communication with stakeholders via VBCs after the festival, so they sustain engagement with their brands to keep them alive in users’ memories and facilitate affective attachment with them and their venue.
This confirms the idea of “persistent engagement” put forward by Cabiddu et al. (2014), where an ongoing dialogue is maintained with customers, beyond the cycle of the trip, that is, even when they are not physically present. This aspect could be a potential research pathway for exploring brand communication beyond purchases and travel processes. It is worth highlighting the continuous communication in which all the stakeholders are immersed throughout the different editions of the festivals and the positive impact that can be inferred in terms of reputation and brand loyalty.
Festivals’ VBCs can be dynamic spaces within which to develop destination promotion strategies. Definitively, these findings demonstrate that festivals are not only branded during periods close to the dates on which they are held, as indicated in existing literature (e.g., Hede, Jago, & Deery, 2005).
These findings have several implications for festival organizers and DMOs to be considered. In the first place, the need to establish a fluid and ongoing communication with the main stakeholders with interests and/or implications in these festivals. This communication must necessarily be multidirectional and must seek the connection (what some would call “social bond”) of them with the festival (and logically their brand). Second, if this communication seeks to connect the stakeholders with the festival, the affective attributes must remain privileged, although the transmission of cognitive information adding qualitative value must also be considered. Third, festivals (and even more those that relate directly to culture) move people because of their brand awareness (the result of branding) but, as has been shown, because festivals facilitate and promote the generation of social capital (maintaining and generating stakeholders’ contacts, i.e., constructing networks) and their stakeholders increasingly tend to consider and promote the identity of the places where they are held (placemaking). Finally, all these elements should be monitored in the future to take advantage of the possibility of cocreating narratives around the festival: VBCs should purposefully be used as a medium for cocreation of the festival experience for attendees, as well as a means of reinforcing the desirable attributes/values of the “neo-tribe” event through their online conversations. The previous elements are those that will favor the creation and survival of a community (in our case a VBC) really engaged with the festival, with the various benefits that can be derived from these processes.
This study is not without limitations but overcoming these could form the basis for some future research proposals. First, although these festivals illustrate unique activities and processes, to enhance generalizability future research should consider other events based on similar characteristics and extend this to other parts of the cultural industries. It would also be interesting to add a qualitative analysis to studies such as the present to discover the existence and influence of virtual communication strategies in engagement constructions.
Moreover, it would also be interesting to know whether the use of a certain social network platform by the organizers to communicate can be influenced by the cohort that uses each platform the most, especially regarding the age of those attending the festival and of other stakeholders, such as artists or bloggers. In this respect, different studies (Smith & Anderson, 2018) show that young people aged 18 to 24 years are substantially more likely to use platforms such as Snapchat, Instagram, and Twitter, while Facebook is used by people across a wide range of age groups. Another area to explore relates to the images uploaded through users’ posts and their evaluation in terms of the context and dynamics involved in the communication. Finally, following recent findings (Papadimitriou, Kaplanidou, & Apostolopoulou, 2018), an additional aspect would be whether festival attendance impacts on the engagement process and the possible relationships with the destination’s engagement,
Summary
As a summary, this study had the objective of understanding the characterization of users’ engagement in festivals’ VBCs and how this engagement impacts on these festivals’ branding strategies. Using a multidimensional and mixed-method analysis and selecting the cases of two VBC (the ones created in Facebook and Twitter) of two international festivals (Sónar and Primavera Sound), we have been able to show the diverse relationships that are established between a range of users’ engagement dimensions: affective and cognitive communication, engagement behaviors and engagement drivers. In this context, the present study has shown the similarities and differences for each VBC and each festival, highlighting the prevalent role of affective attributes, sharing behaviors, and branding motivations but also the increasing importance of cocreation, social capital, and placemaking. Finally, this study has also shown the importance for these festivals’ organizers to maintain a communication inside these VBC that goes beyond the simple festivals’ attendance to construct branding during the whole year and through a diversified range of activities accompanied by promotional messages.
