Co-creation experiences in social media brand communities: Analyzing the main types of co-created experiences

Abstract

Purpose

This paper presents an in-depth investigation on how brands may concur to the co-creation of consumers’ experiences. In particular, the purpose of this paper is to clarify the main types of co-created experiences that consumers may encounter as a result of social media brand communities.

Design/methodology/approach

To identify the main types of co-created experiences, a digital investigation has been used as the main method of analysis. The authors draw their digital investigation on the digital methods paradigm.

Findings

Four principal types of co-created experiences have been identified and conceptualized, namely, brand’s products’ individual usage experiences, auto-celebrative experiences, brand’s products’ communal usage experiences and collective celebration experiences.

Originality/value

Results stress the importance for brand strategists to involve members of social media brand communities to stimulate co-creation experiences. Specifically, it emerges that the simultaneous interaction among members of the community and the brand may directly affect co-creation experiences.

Propósito

La presente investigación se propone analizar en profundidad cómo las marcas pueden estar de acuerdo con la co-creación de las experiencias de los consumidores. En particular, el objetivo de la investigación es aclarar cuáles son los principales tipos de experiencias co-creadas que los consumidores pueden experimentar debido a su participación en las comunidades de marcas de redes sociales.

Diseño/metodología/enfoque

Para hacerlo, en primer lugar, se han identificado los factores que influyen en la co-creación de las experiencias de los miembros de las comunidades de marcas. En particular, el punto de partida de esta investigación está representado por el papel de otros consumidores y de la marca en la co-creación de experiencias. Con el fin de identificar los principales tipos de experiencias co-creadas, se ha utilizado una investigación digital como el principal método de análisis. Dibujamos nuestra investigación digital en el paradigma de Métodos Digitales.

Hallazgos

Se identificaron y conceptualizaron cuatro tipos principales de experiencias co-creadas.

Originalidad/valor

Los resultados enfatizan la importancia de que los estrategas de marca involucren a los miembros de las comunidades de marcas de medios sociales para estimular la co-creación de experiencias. Específicamente, surgió cómo la interacción simultánea de otros miembros de la comunidad y la marca puede afectar la co-creación.

Palabras clave:

Co-creación de valor, Comunidades de marca, Experiencias de los consumidores, Experiencias co-creadas, Investigación digital, Marketing experiencial

Citation

Rialti, R., Caliandro, A., Zollo, L. and Ciappei, C. (2018), "Co-creation experiences in social media brand communities: Analyzing the main types of co-created experiences", Spanish Journal of Marketing - ESIC, Vol. 22 No. 2, pp. 122-141. https://doi.org/10.1108/SJME-03-2018-0011

Introduction

Communities were initially thought as small, homogeneous, closely acquainted social groups sharing a sense of gemeinschaft (Tönnies, 1887). It has consequently been revealed that members feel psychologically united as a result of “emotional and familial bonds” shared with others (Thomas et al., 2013, p. 1011). Such a classic conceptualization, however, is now inadequate in representing the complexity of contemporary communities (Husemann et al., 2015). As human relationships have become progressively digitized (Arvidsson and Caliandro, 2016), modern communities are currently characterized by how dynamic they are. As humans need to participate in some form of community to develop their social skills (Tajfel, 2010), the classic conceptualization of community be considered the cornerstone of any research exploring communities.

Because of the importance and prevalence of various communities in everyday life, consumers’ participation in any form of community has widely been explored in consumer research (Muniz and O’Guinn, 2001). Indeed, it is commonly accepted that participation within a community significantly impacts consumers’ behavior (Schouten and McAlexander, 1995). Consumption communities, therefore, are amongst the most explored social groups in marketing literature (Thomas et al., 2013); particularly brand communities. Brand communities are consumption communities formed by consumers who share a similar passion toward a specific brand (Muniz and O’Guinn, 2001). They have been extensively explored by literature, as they are the most exploitable kind of community in strategic marketing (Zaglia, 2013).

Building on these premises, pertinent literature has explored how brand communities – and their digital counterpart, social media brand communities – may play a fundamental role in value co-creation practices; particularly with regard to the assessment of the importance of specific types of community (Carù and Cova, 2015). From this perspective, it has emerged that the dyadic relationship between consumers and brands (or service provider) is strengthened by consumers’ participation in a community. This is evidenced by Carù and Cova (2008), who seminally assessed that brand may contribute to the co-creation of consumers’ experiences through constant interactions. It became apparent that consumers had the power to influence the experiences of other members of the community, thus co-creating whole new experiences (Triantafillidou and Siomkos, 2014). Pertinent literature has since explored which consumption behaviors stimulate experience co-creation, and which factors may allow consumers to obtain a satisfactory experience (Gentile et al., 2007).

This notwithstanding, the literature on experience co-creation in brand communities is limited in at least several aspects. On the one hand, the majority of literature has observed the phenomenon from a brand-based perspective (Carù and Cova, 2008). There is thus a need to explore experience co-creation from a consumer-orientated point of view (Gentile et al., 2007). Second, the main types of co-created experiences are yet to be fully conceptualized (Ismail et al., 2011). Indeed, to the authors’ best knowledge, apart from Nuttavuthisit (2010), very few authors have attempted to investigate whether a member’s individual consumption activity – instances in which a consumer uses a product individually and then shares feedback with other community members (Carù and Cova, 2015) – or communal consumption activity – in which a consumer participates with other consumers in an activity sponsored by the community (Triantafillidou and Siomkos, 2014) – correspond to different types of co-created experiences. Then, there is a need to understand how individual or communal consumption activities (or both) are related to the co-creation of satisfactory experiences. Finally, in the current digital era, it is necessary to investigate and conceptualize how brands may strategically exploit the phenomenon of experience co-creation in marketing strategies (Triantafillidou and Siomkos, 2014). This research will try to provide insight into how social media brand communities may be exploited for experience co-creation.

Building on these premises, this research aims to identify the main types of co-created experience that members of the brand community may develop after different kinds of consumption activities. For this, a consumer-focused observation has been adopted. However, because the majority of brand communities nowadays have at least a digital offspring (Zaglia, 2013), social media brand communities have been considered as the setting of the research.

Apart from the introduction, this research is structured as follows: the next paragraph deals with existing literature on brand communities, their progressive transformation in social media brand communities and experience co-creation; a digital investigation on the principal European runners’ communities is then outlined (Arvidsson and Caliandro, 2016); finally, results, conclusions and suggestions for future research are presented.

Theoretical background

Consumption communities’ evolution in the digital era: from subcultures of consumption to social media brand communities

Consumption communities are formed by consumers who share a common commitment toward a specific consumption activity (Schouten and McAlexander, 1995). The three main types of consumption communities are as follows:

1. consumption subcultures (Schouten and McAlexander, 1995);

2. brand communities (Muniz and O’Guinn, 2001); and

3. consumer tribes (Canniford, 2011a).

Any kind of consumption community shares three common characteristics denominated makers of the community (Algesheimer et al., 2013):

1. consciousness of kind;

2. shared rituals and traditions; and

3. moral responsibility (Muniz and O’Guinn, 2001).

“Consciousness of kind” suggests that the community members share a communal sense of belonging; they feel different from “outsiders” and are hostile toward potential intruders (Latour, 2005). “Shared rituals” means that members share recurring practices, routines and common jargon when communicating with each other (Schau et al., 2009; Thomas et al., 2013). “Moral responsibility” indicates that members tend to help one another (Zaglia, 2013).

Beyond these three common characteristics, consumers show certain patterns in forming intentions to join consumption communities. They are often motivated by economic advantages and psychological well-being (McAlexander et al., 2002) – they may join consumption communities to learn about products and brands before making final shopping decisions (McAlexander et al., 2002). Otherwise, they may join consumption communities to identify themselves with the community’s values and symbols (Muniz and O’Guinn, 2001). This adheres to social identity theory (Tajfel, 2010), which claims that individuals are naturally prone to join groups they esteem to achieve social identification.

Although the three characteristics are common to all consumption communities (Canniford, 2011a), communities may vary in terms of power distribution, community origins, social positions and locus regarding raison d’être, whether the members are linked by consumption activities or brand values (Canniford, 2011b). Consumption subcultures share strong interests in a specific consumption activity, which becomes the community locus (Schouten and McAlexander, 1995). Consumers directly establish consumption subcultures, with governance based on member hierarchy and rigid internal structure (Canniford, 2011b), in which members resist authority while showing barbarian and outlaw behaviors. In contrast, brand communities are “non-geographically bound […] based on a structured set of social relations among admirers of a brand” (Muniz and O’Guinn, 2001, p. 412). Members share a common interest, passion or love toward a specific brand (McAlexander et al., 2002) and are frequently managed by a brand manager, are slow to change, express the mainstream culture of their geographical area and foster consumers’ activities around a brand (Habibi et al., 2014a). Finally, the third type of consumption community is the consumer tribe, formed to develop social ties with others who value products or services as their locus (Canniford, 2011a, 2011b). Unlike other kinds of consumption communities, they are multifarious, transient, playful, entrepreneurial and characterized by a diffused internal governance (Cova et al., 2012; Goulding et al., 2009).

Over the recent years, it has been possible to observe the transformation of many brand communities in social media brand communities (Habibi et al., 2014a, 2014b). Indeed, according to Rialti et al. (2017a), social media brand communities differ in terms of dimensions and are frequently formed by several thousand members. Moreover, as a consequence of the interactivity of social media, social media brand communities allow brand strategists to share content with members and receive constant feedback from them (Laroche et al., 2012). Social media brand communities thus offer members different and innovative ways of consuming content generated by brands, as well as a new path through which to share new user-generated content with others (Habibi et al., 2014a, 2014b).

In relation to social media brand communities’ characteristics, strategic marketers frequently target these latter ones (Hofacker and Belanche, 2016; Husemann et al., 2015). In a similar fashion to traditional brand communities, social media brand communities are the most exploitable of current communities in terms of marketing strategy (Zaglia, 2013). Marketing strategists can therefore foster members’ engagement with brands (Habibi et al., 2014b) by establishing bonds based on physical, emotional and cognitive involvement in the community (Patterson et al., 2006). Actually, if consumers are identified and engaged with the community, they will be most likely to develop a bond with the brand (Algesheimer et al., 2005; Hollebeek, 2011); particularly in communities that offer the opportunity for consumers to participate in brand-related activities (Schau et al., 2009). Marketers can thereby encourage consumer or brand engagement through social media brand community engagement.

Pertinent literature has explored the ways in which brand strategists may use both traditional and social media brand communities as a source of product and brand innovation (Cova and Dalli, 2009). Indeed, both offer brand strategists a unique platform through which to collect information and feedback on products from deeply committed consumers. Furthermore, they may be instrumental in advance testing the launch of new products or new branding campaigns (Ramaswamy, 2008). Such communities are vital in engaging consumers in word-of-mouth marketing (Cova and Dalli, 2009). Co-creation is therefore frequently used as a principal lens through which to explore brand community’s potential in strategic marketing (Ramaswamy, 2008; Zwass, 2010). This is particularly true when considering social media brand communities (Habibi et al., 2014a). In fact, in social media brand communities, the interactions between brand strategists and consumers are often more articulated than in traditional communities (Zaglia, 2013).

From this perspective, scholars have recently started to explore social media brand communities as a vehicle to facilitate the creation of unique experiences for consumers. This has been deemed possible as a brand may contribute to co-creating consumers’ experiences by interacting with consumers’ themselves (Carù and Cova, 2008; Cova and Dalli, 2009). Second, it emerged that consumers may influence the experiences of other members of the community, thus co-creating brand new experiences together (Triantafillidou and Siomkos, 2014). Such a phenomenon has been labeled in existing literature as experience co-creation (Carù and Cova, 2015).

Experience co-creation in social media brand communities

Gentile et al. (2007, p. 397) identified consumers’ experiences as a personal evaluation of a product or service based on “the comparison between a customer’s expectations and the stimuli coming from the interaction with the company and its offering in correspondence with the different moments of contact or touch-points”. When the stimuli and the personal perceptions deriving from a consumption experience overcome the original consumers’ expectations, consumers will experience a satisfactory experience. Otherwise, the experience will be negative (Shaw and Ivens, 2005). Over the past decade, the consumers’ experience construct has been unpacked by several scholars. For example, Brakus et al. (2009) suggest that the consumers’ experience derives from sensorial, emotional, cognitive, lifestyle and relational stimuli linked with the consumption experience. Therefore, marketing strategies should stimulate several perceptive spheres of consumers to foster the development of positive experiences (Rialti et al., 2016a; Rialti et al., 2016b; Zollo et al., 2018). Similarly, according to O’Loughlin et al. (2004), a positive consumer experience derives from consumers’ perceptions of the brand, the quality of transactions in product or service acquisition and the quality of the relationship with other consumers of the brand. A satisfactory experience for consumers may arise from consumption activities being capable of stimulating joy and pleasure (Triantafillidou and Siomkos, 2014).

Recently, the capability to stimulate a satisfactory consumers’ experience has emerged as a fundamental aspect of modern marketing strategies (Pine and Gilmore, 1998; Rialti et al., 2016c). Consumers obtaining such experiences as a consequence of a consumption activity, in fact, are more loyal to the brand and more prone to advocate it (Brakus et al., 2009). Consumers’ experience has been widely explored by value co-creation streams of marketing literature (Prahalad and Ramaswamy, 2004). Specifically, a positive experience may be capable of creating permanent memories in consumers’ minds which is a form of intangible value for consumers (Gentile et al., 2007). In turn, the nostalgia related to their memories may stimulate consumers to replicate the consumption experience (Luna-Cortés, 2017; Triantafillidou and Siomkos, 2014). Loyal consumers replicating their consumption experience may thus re-generate revenue streams for the brand (Schmitt, 1999).

However, as stressed by Gentile et al. (2007), to foster consumers’ positive experiences, the touch-points between the brand and consumers constitute an extremely important role. These touch-points, actually, have recently been identified as one of the main components of the so-called context of experience (Akaka et al., 2015), which, according to pertinent literature, is the ensemble of the ecosystem in which consumers interact with brands and all the actors influencing the creation of experience (Chandler and Vargo, 2011). Indeed, observing experience creation, building on the notion of context, may allow scholars to properly consider both the ecosystems in which the phenomenon occurs and actors/individuals that influence the process of experience creation (Akaka et al., 2015; Vargo et al., 2008). In this sense, drawing from both the definition of context of experience and the characteristics of social media brand communities, a number of studies have identified the latter as a possible context for experience formation (Ismail et al., 2011). To reinvigorate this assumption, Akaka et al. (2015) have advocated consumption communities (characterized by proper sets of value shared by consumers), frequent dialogic interaction between brand and consumers and, finally, frequent consumer to consumer interaction (Muniz and O’Guinn, 2001).

Consumers’ experiences arising from their participation in brand communities have been identified as co-created experiences (Carù and Cova, 2015). Indeed, experiences deriving from participation in brand communities are influenced by not only consumers’ perception of brand and brand’s products but also brand managers and the intervention of other consumers (Triantafillidou and Siomkos, 2014). First, it has been assessed that the opportunity for the brand to initiate a dialogue with consumers is fundamental to fostering the co-creation of experiences (Rialti et al., 2016b). Indeed, brand-consumers’ interactions may positively influence the overall experience of consumers (Ramaswamy, 2008). Second, within a community, consumers may participate in communal activities. Specifically, other consumers’ interventions may influence the memories of consumers and shape the experience of the individual consumer (Helkkula et al., 2012; Jaakkola et al., 2015).

Despite the attention toward experience co-creation, some gaps in this stream of literature still exist: while literature has explored the factors fostering the co-creation of positive experiences (Ramaswamy, 2008), little attention has been paid to identifying whether different kinds of experiences may arise from different factors influencing the process (Carù and Cova, 2015). In fact, while scholars such as McColl-Kennedy et al. (2015) and Nuttavuthisit (2010) have tried to configure the different kinds of co-created experiences arising in relation to different consumers’ motivations, it has not been considered whether different experiences are co-created because of the involvement of different actors. Again, research in this area has not determined whether different experiences arise from the individual’s use of a product or through participation in a consumption activity with other consumers – i.e. involvement in a sponsored event reserved for members of a community. As such, this research focuses on the different experiences that may emerge as a result of the influence of different actors. It then focuses on the importance of the kind of consumption activity.

Building on such a gap, the aim of this research is to identify and categorize the principal types of co-created experiences of members of brand communities. The results will then be placed into a framework based on the actors influencing the co-creation of experience and on two different kinds of consumption activities, namely, individual consumption and participation in communal activities.

To accomplish this, we have analyzed the content concerning consumption activities shared by members of social media brand communities formed by runners. These communities are quite peculiar in that they are created by brands to involve runners in communal activities (Thomas et al., 2013). In fact, runners’ communities initiated by the brand have a nested system of interrelated virtual identities. However, they are also populated by social media managers engaging in dialogue with consumers (Guinalíu and Jordán, 2016; Rialti et al., 2017b). Moreover, such communities have frequently been considered by scholars exploring co-creation-related phenomenon (Ramaswamy, 2008).

Method

To answer our research questions, we conducted a qualitative digital investigation on several runners’ social media brand communities (Caliandro and Gandini, 2017). We elected to use a method capable of capturing relevant quantities of social media data (Caliandro, 2017; Rialti et al., 2016a). The selected method is articulated as follows. First, information sources were identified and data were collected and divided according to their origin. Then, the data were processed through an automated data analysis and, among the results, a snap of the graphical structure of the network of members was obtained. Finally, the results were interpreted (Arvidsson and Caliandro, 2016).

Setting the stage: identifying information sources

To gather the required data, we identified several runners’ communities existing on Twitter. We studied six specific runners’ brand communities related to two prominent brands in the running market: Nike and Adidas. These six runners’ brand communities are the European runners’ brand communities, characterized by the most elevated number of tweeted Tweets in the observation period (5 October 2016 to 5 April 2017). The following procedure was followed to identify the six communities. First, we followed four Twitter hashtags functioning as worldwide community catalysers: #nrc (Nike), #nikerunclub (Nike), #nikerunningclub (Nike) and #Adidasrun (Adidas). Then, moving on from the first four catalyzing hashtags, we identified the three European local brand communities for each brand: #whyirunchampselysees (Paris-Adidas), #nrclondon (London-Nike), #whyrunmadrid (Madrid-Adidas), #werunamsterdam (Amsterdam-Nike) #nrcbcn (Barcelona-Nike) and #whyirunfrankfurt (Frankfurt-Adidas). We selected these six local communities, as they allowed us to analyze a significant number of tweets created by consumers.

Following this procedure, a grand total of 102,430 tweets were collected for analysis. Building on existing literature (Caliandro, 2017; Rialti et al., 2016a), we selected the tweets that received at least one mention (@s.) or were retweeted (RT) at least once. These tweets are ostensibly more important than others, as they have been actively observed by at least another member of the community. In total, 96,321 tweets were subsequently conserved. Among these tweets, 776 were immediately excluded, having been diffused either by the brand or brand managers and thus deemed unhelpful in identifying experiences developed by consumers. To identify the tweets to be excluded, we assessed the personal profile of all of the users, identified which ones were related to the brand and, finally, eliminated their tweets. This resulted in 95,545 conserved tweets. These 95,545 tweets were distributed as follows: 93,936 tweets were related to communities’ catalyzer hashtags and 1,609 tweets were derived from the six local brand communities. Moving on from this division, the tweets derived from communities’ catalyzer hashtags were used to quickly analyze the network composed by runners in digital space. Indeed, this phase was necessary to understand the kinds of actors involved in digital conversations and whether the users really aggregated around the brands in an environment such as Twitter. In-depth analysis was conducted on the 1,609 tweets from the six selected local communities. It was deemed more appropriate to consider a homogeneous sample of communities from European countries (Geertz, 1973), in coherence with channel selection procedures suggested in previous studies using the same methodological approach (Arvidsson and Caliandro, 2016; Rialti et al., 2016a). In fact, when the focus of the research is on factors fostering a consumption experience or consumers’ perception, it may be relevant to consider consumers with similar cultural background to have more reliable results (Akaka et al., 2015; Carù and Cova, 2008; Ismail et al., 2011).

Data analysis

Methodologically, we combined digital methods and netnography (Arvidsson and Caliandro, 2016). This procedure has been labeled as digital investigation (Caliandro and Gandini, 2017). Our digital investigation included observing and analyzing community members’ articulations regarding activities and discourses on Twitter (Marres, 2015). Digital methods use online data “for the study of societal change and cultural conditions” (Rogers, 2015, p. 1) through IT techniques and “natural” analytics built into digital platforms, such as mentions (@s) and retweets (RTs), which can be used for sampling a dataset and for measuring the intensity of relations among users (Caliandro and Gandini, 2017). The digital methods approach was thus used to detect and study instances in which Twitter users advertised their activities and discussed running topics. Netnography is a qualitative method, specifically designed for exploring and understanding consumer cultures within digital environments (Kozinets, 2010) using interpretative text analysis and participant observation to reconstruct digital forms of sociality and webs of significance (Delgado-Ballester and Fernández-Sabiote, 2016; Geertz, 1973). Thus, the netnographic approach provided deeper understandings of runner activities and discourses (Delgado-Ballester and Fernández-Sabiote, 2016; Rialti et al., 2016a).

Our digital investigation began with the collection of all tweets related to the selected hashtags using custom-built software – a Python script programmed for interrogating the streaming API of Twitter (Russell, 2013) – which allowed us to collect tweets in real-time. As previously discussed, we focused on the ten most-used hashtags: #nrc, #nikerunclub, #nikerunningclub, #adidasrun, #whyirunchampselysees, #nrclondon, #whyirunmadrid, #werunamsterdam #nrcbcn and #whyirunfrankfurt. After the depuration of tweets from brand (i.e. advertisings) or social media managers, a sample of 95,545 tweets remained.

We then submitted the dataset, in different stages, to an automated analysis of metadata and network analysis, a netnographic analysis and a qualitative content analysis.

Automated analysis on metadata and network analysis.

When performing the automated analysis on metadata, we built an ad hoc Python script programmed to extract – from the whole dataset composed of 95,545 tweets – hashtags (#), mentions (@) and retweets (RT) and to count their occurrences. The script automatically released the lists of the most used hashtag, most mentioned users and most retweeted messages. These rankings were useful for quantitatively exploring the activities and opinions of users. The results of the automated analysis are instrumental in obtaining both the results of the network analysis and the sample of tweets for qualitative analysis. Indeed, as shown in Table I, from this preliminary analysis, it was possible to identify the number of tweets related to each hashtag that we followed.

Moving on from this, we were able to analyze the most popular hashtags (Table II).

The same approach was also used to derive from the dataset the level of activities of the members of each communities. This kind of analysis can provide useful results in terms of affective attachment of the members to the community. On average, 40.98 per cent of users within the six communities shared at least two tweets and were responsible for an average of 83.84 per cent of tweets circulating within a given community (Table III).

Finally, we conducted an exemplary network analysis which gave us a quick and accurate picture of social structures (Gruzd et al., 2011). The network analysis focused on mentions (@) and retweets (RTs) metadata. The network analysis of mentions (@s) and retweets (RTs) was based on the in-link technique, which means that we focused on the mentions (@s) and/or retweets (RTs) each user received. Users with more in-links (or in-degrees) were mentioned and/or retweeted most frequently, so we considered them to be most popular (Arvidsson et al., 2015). The networks were analyzed and visualized through Gephi (Bastian et al., 2009), an open-source software for visualizing and exploring graphs. The results showed us how effectively members gather together (see Figure 1 for the test on the Nike-related tweets).

Netnographic analysis.

The netnographic analysis allowed for exploration and interpretation of the meanings and uses of the tweets (Kozinets 2015). The interpretative analysis was complemented by an automated analysis of metadata. An ad hoc Python script, built by us, was programmed to extract hashtags (#), mentions (@), retweets (RTs), favorites (Fav) and URLs and to count their occurrences. It automatically listed the most mentioned users, most retweeted messages, most liked messages, most used hash tags and most shared URLs. The automated analysis proved useful in systematically sampling and navigating our dataset and thereby rapidly interpreting the tweets (Caliandro and Gandini, 2017).

Qualitative content analysis.

Qualitative content analysis (Johnstone, 2008) consisted of the step by step reading of the text within tweets in a bid to detect their major discussion topics (Poell and Borra, 2011). Specifically, we conducted a manual content analysis on the 1,609 tweets belonging to the members of local communities (Jabreel et al., 2016). With the aim of the research in mind, we first tried to divide the tweets according to the kind of consumption experience fostering the development of the experiences (Altheide, 1996). First, we separated the tweets describing positive experiences deriving from individual consumption from the ones describing positive experiences deriving form communal activities, thus building on both on Nuttavuthisit’s (2010) and Triantafillidou and Siomkos’ (2014) hypotheses concerning the ways in which creation of experience may derive either from the individual consumption of goods (or services) or the participation in an activity organized by the brand. Second, observation was based on whatever it was influenced by. This strategy was coherent with Carù and Cova’s (2008, 2015) assumption, which advised that both the brand and other consumers may influence the creation of experiences of members of brand communities. The establishing of these categories of analysis followed a grounded and iterative process (Glaser and Strauss, 1967; Jabreel et al., 2016). Categorization of the tweets was not considered a priority – although guided by the aim of the research – but rather emerged during the reading of the texts through constant examination by the four authors (Altheide, 1996). Finally, 572 of the original 1,609 tweets were considered individual consumption activities, while the remaining 1,037 were concerning participation in community-related activities. With regard to the other parameter, 731 tweets showed the brand somehow, while the other 878 were tweets showing, for example, photos with friends or other members of the communities.

Results of the analyses

Several insights emerged from our analysis of 1,609 tweets from the following local communities: #whyirunchampselysees, #nrclondon, #whyrunmadrid, #werunamsterdam, #nrcbcn and #whyirunfrankfurt. In a first spite, the qualitative analysis showed that the community members share several different types of tweets concerning their activities and related experiences. Indeed, it emerged that they share content concerning brand-related products. For example, a tweet from the #Nrcbcn community read: “Ran 12.27 km with Nike + Run Club Breaking in my new shoes #nrc #nrcbcn #nikeplus #nikerunning pic.twitter.com/4zUFkC4DW1.” Additionally, it revealed that the five most liked tweets (aka favourites in Twitter jargon) shared photos celebrating the linking value of the community by depicting members running, cheering and having fun

In addition to this, we noticed four main types of photos and contents:

1. photos and contents showing a recently used or acquired product of one of the brands (i.e. TRAINING [FOOTING], Du soleil, Des chaussures, Un footing # whyirunchampselysees, link: pic.twitter.com/Vz34uXFl1G - 2 Favs);

2. auto-celebrative photos or contents illustrating or describing a personal athletic achievement (i.e. Ran 12.35 kilometres with Nike + Run Club Niiice #nrc #nrcbcn #nikeplus #nikerunning, link: pic.twitter.com/cRSRNI3ZwR – 4 Favs);

3. photos or contents showing a subjectively meaningful brand related moment (i.e. It’s Wednesday, which means one thing […] #nrclondon @NikeRunning, link: pic.twitter.com/kU5GuWQFOy – 11 Favs); and

4. photos picturing a collective celebration (i.e. Another fab night at the #nrclondon. Keep pushing, keep running…#nrc #eatsleepeunrepeat @Nike, link: www.instagram.com/p/BK87BI3jMVc – 8 Favs).

As it is possible to observe from the four types of photos and contents, members shared particularly emotional moments through their tweets and photos. The most interesting and salient observation is that the individual and the collective celebration photos are always “branded moments”. The brand is almost always visible in the photos: in the background, on the clothes and in hashtags. Interestingly, both the qualitative and quantitative analyses of tweet content indicated this brandisation of sporting practices. The branded hashtags always appeared in the top-ten most-used hashtags. This tendency is also evident in the campaigns that brands launch on Twitter; for example, #HeretoCreate by Adidas: a platform that invites consumers to “Check out the personal stories of athletes who use creativity” (www.adidas.com/us/heretocreate) in their everyday training.

In addition to this, with regard to the analysis of content, coherently with the initial results from the qualitative observation, it emerged that 572 of the 1,609 tweets were about individual consumption activities, while 1,037 were related to participation in communal consumption activities. Instead, in terms of the actors influencing the brand, 731 of the 1,609 tweets showed how the brand and brand strategists somehow influenced consumption activity and the related experience, while 878 showed how other members contributed in shaping the final experience of the consumer.

With regard to network analysis, we observed slight differences and a peculiar phenomenon amongst the global communities (#nrc, #nikerunclub, #nikerunningclub and #adidasrunner) in respect of the local communities. Fundamentally, hashtags seemed more conducive in forming brand publics (Arvidsson and Caliandro, 2016) rather than proper brand communities (Muniz and O’Guinn, 2001). A brand public is a loose online association among strangers, united by digital devices (e.g. #nikerunclub) rather than direct interaction. For example, users convening around #nrc (the hashtag attracting the most tweets) had very little interaction. On an average, they exchanged 1.14 mentions (@s) and/or retweets (RTs). In addition, out of 3,534 users, 2,774 (78.49 per cent) sent just one or even zero mentions (@s) and/or retweets (RTs) to other users, showing some signs of disconnection within the network (Figure 1). These high levels of modularity are indicative of a fragmented network (Brandes et al., 2008). This suggests that users mainly use mentions and RTs to associate their tweets with brands or other popular accounts rather than to initiate or sustain interactions. In fact, little or no between-cluster communication occurs – the network has mainly scattered and disconnected nodes. However, as it is possible to see in the cluster, some touch-points between the members of such communities do exist (Arvidsson, 2013; Arvidsson and Caliandro, 2016). The most favored tweets seem to respect the same “cultural grammar” we saw for local communities: “branded moments” of collective and/or individual celebration. In conclusion, the communities effectively act as mediators between brands and consumers; they embed the brands into emotional memory and everyday practices (Arvidsson, 2006).

Framework development

The qualitative analysis of content related to communal and individual activities of members of the six European brand communities allowed us to identify several insights concerning experience co-creation. In particular, it was possible to identify the principal community-related activities from which consumers’ experiences may originate. Specifically, it emerged that co-created experience may arise both from individual and communal consumption activities. Second, it showed the importance of the brands’ interaction with consumers or the brand–consumer relationship and the consumer-consumer interactions as elements shaping possible co-created experiences (Carù and Cova, 2015). Consequently, we developed the following framework (Table IV). On the x-axis, we consider that co-created experiences differ in terms of whether they are co-created after an individual or communal consumption activity (Carù and Cova, 2008, Ismail et al., 2011; Prahalad and Ramaswamy, 2004). On the y-axis, we consider that co-created experience differs depending on whether the other actor (excluding individual consumers) influencing experience co-creation is the social group or the brand (Lemke et al., 2011; Ramaswamy, 2008).

Moving on from the proposed model, the four principal types of co-created experiences are:

1. Brands’ products’ individual usage experiences, which are experiences deriving from individual consumption activities, i.e. consumers initially using one of the brand’s individual products and, only after that, sharing content related to the consumption practice. In spite of being individual, such experiences are co-created by brand interferences in consumers’ everyday life – such as encouraging consumers to share their experiences (Arvidsson, 2006). Such experiences are usually characterized by some degree of desire for individualism shown by the consumer (Carù and Cova, 2008, 2015). Coherently with literature on co-created experiences, we identify a type of experience related to product use but influenced by the brand’s invitation to share content (Prahalad and Ramaswamy, 2004; Ramaswamy, 2008).

2. Auto-celebrative individual experiences, which are experiences deriving from individual consumption activities which are influenced by being a part of the community (Prahalad and Ramaswamy, 2004; Carù and Cova, 2008). This type of co-created experience was identified by analyzing the content falling within the previously discussed category: “Auto-celebrative photos illustrating a personal athletic achievement”. In line with literature on experience co-creation, such co-created experiences derive from individual experiences; however, consumers often feel the need to share their success with other community members to receive approbation from them (Ismail et al., 2011; Lemke et al., 2011).

3. Brands’ products’ communal usage experiences, which are similar to the first category but involve greater consumption activities involving other community members – i.e. consumers’ use one of the products of the brand during a brand sponsored event. From this perspective, such experiences are related to products being used to participate in a communal activity. Specific types of co-created experiences have been identified by analyzing content concerning collective usage of products. Therefore, similar to the basic assumption underlying the motivation for consumers’ participation in a community (Algesheimer et al., 2005), such experiences are co-created by the intention to participate. Brand influence and other consumers’ influence, therefore, are maximum in such co-created experience (Carù and Cova, 2008; Cova and Dalli, 2009).

4. Collective celebration experiences, which are experiences co-created as a result of participation in communal activities – i.e. consumers use one of the products of the brand in the company of other members of the community. The influence of other consumers, then, is fundamental to co-creating the experience (Carù and Cova, 2008, 2015; Triantafillidou and Siomkos, 2015). Indeed, the negative behavior of one of the other members may negatively influence the memory that will form in the consumers’ mind and, as a consequence, it may influence the memory related to the products’ use.

Discussion and implications

This research contributes to the stream of literature exploring experience co-creation in brand communities (Prahalad and Ramaswamy, 2004; Ramaswamy, 2008). In particular, the developed framework offers some interesting insights on the types of co-created experiences related to consumers’ participation in a brand community (Cova and Dalli, 2009). On the one hand, we attempt to identify the two main actors capable of shaping the experience of members of brand communities (Triantafillidou and Siomkos, 2015). As such, we also try to identify the specific role of brand and of other consumers in influencing experience co-creation. On the other hand, we explore how co-created experiences differ when resulting from two different consumption activities (Carù and Cova, 2015).

In summary, we unveil that four principal types of co-created experiences exist:

1. brands’ products’ individual usage experiences;

2. auto-celebrative individual experiences;

3. brands’ products’ communal usage experiences; and

4. collective celebration experiences.

Therefore, we assess the importance of other actors, both the brand and other consumers, in co-creating experiences related to consumption activities (Carù and Cova, 2008, 2015). The results are consistent with the original conceptualization of experience co-creation phenomenon (Prahalad and Ramaswamy, 2004; Carù and Cova, 2008). However, the framework systematizes the roles of the brand and other consumers in experience co-creation with regard to the type of consumption activity. Specifically, our framework shows the ways in which other consumers may influence the final experience of an individual consumer by increasing or reducing the quality of the experience (Carù and Cova, 2015), thereby co-creating or co-destructing the experience of individual consumers. On the other hand, brand incentives to share experiences online, such as encouraging consumers to diffuse their experience on brand-initiated platforms, may amplify consumers’ satisfactory experiences, offering a resonance chamber (Ramaswamy, 2008). Brands may thus increase a consumer’s intention to replicate the experience.

Moving on from these considerations, although some touch-points with existing literature emerged, we believe that these results enrich the literature on the experience co-creation phenomenon. First, the research tries to identify and systematize the kind of co-created experiences in relation to both the actors and the kind of consumption activity. Indeed, previous researchers, such as Carù and Cova (2008, 2015) or Triantafillidou and Siomkos (2015), have identified the role of actors and contexts in the co-creation of experience (Akaka et al., 2015), but a clear systematization of the simultaneous roles of both was still needed. While some research, such as that of Nuttavuthisit (2010), developed framework exploring the motivation of co-creation, these seminal pieces of research were less focused on the actors influencing co-creation stricto sensu and more on the outcomes of the process of experience co-creation.

In addition to these results, some implications may be provided. In particular, first, it emerged that marketing strategists should never neglect the importance of brand community membership in the formation of a consumer’s overall experience – consumers may influence each other. A negative comment related to a photo from another member may influence the co-created experience (Carù and Cova, 2008). Thus, to prevent consumers from associating a brand with a negative experience, marketing strategists should monitor the status of the relationships within the community – which is particularly true in the current era of social media (Rialti et al., 2016a). Next, brands should encourage consumers to share digital contents concerning their experiences (Arvidsson, 2006). In fact, positive feedback from both brand and other consumers may increase quality of experience and, in turn, foster a positive relation between brand and consumers (Ismail et al., 2011). Finally, the majority of content concerning positive experiences share the common feature of being branded. In particular, positive content showing the satisfactory experience of a consumer with a brand may encourage other consumers to buy that brand’s products (Ismail et al., 2011; Lemke et al., 2011).

Conclusion, limitations and suggestions for future research

This research highlights some characteristics of experience co-creation in brand communities (Lemke, et al., 2011). In particular, it identifies some kinds of co-created experiences, moving from the assumption that experience co-creation occurs depending on whether the experience from a consumer is influenced by other consumers or by the brand (Carù and Cova, 2008). The principal results deal with the developed framework and the implications for marketing strategists on how to reap the benefits of experience co-creation. Moreover, implications on the importance of monitoring the phenomenon in digital era have been enunciated (Rialti et al., 2016a).

In spite of the relevant findings, this research still presents some limitations. First of all, the methodology allowed the authors only to extract the tweets containing at least one of the hashtags identified at the start of the research. As it is possible to observe, this approach does not ensure the collection of all of the possible tweets related to all of the considered social media brand communities (Arvidsson and Caliandro, 2016). Apart from the limitations related to the selected methodology, the other limitations of this research are related to the low number of communities considered and the observation period. In fact, the framework has been developed only moving on from the contents of the selected six European local communities. Hence, because of these limitations, results are not fully generalizable in a worldwide sense. From this perspective, we would suggest that scholars replicate this research on a larger scale, with a greater number of communities. In addition, another challenge lies in exploring, in more depth, the differences regarding content generated and shared by members of social media brand communities and by members of brand publics. Indeed, members of brand communities and members of brand publics tend to behave differently. Similarly, it may be interesting to investigate the impact on experience co-creation related to participation in a brand public.

Figures

Figure 1.

An exemplary social network analysis

Table I.

Tweets and users for each community

Community	Tweets	No. of Users
#Nrc	59,361	4,233
#Adidasrun	20,755	1,026
#Nikerunclub	13,299	1,919
#Whyirunchampselysees	1,039	32
#Nikerunningclub	521	205
#Nrclondon	167	31
#Whyirunmadrid	125	22
#Werunamsterdam	115	21
#Nrcbcn	99	4
#Whyirunfrankfurt	64	8
Tot	9,5545	7,501

Source: Authors’ elaboration

Table II.

Top ten hashtags per community

#Whyirunchampselysees	Fr.	#Nrclondon	Fr.	#Whyirunmadrid	Fr.	#Werunamsterdam	Fr.	#Nrcbcn	Fr.	#Whyirunfrankfurt	Fr.
whyirunchampselysees	113	nrclondon	68	whyirunmadrid	59	werunamsterdam	95	nrcbcn	48	adidasrunnersfrankfurt	25
whyirunchampselys	87	nrc	41	adidasrunners	12	nrc	38	nrc	45	tracktuesday	2
whyirunparis	36	justdoit	16	madridmarathon	6	nikerunclub	9	nikeplus	30	runchat	2
whyirunparispic	12	nikeplus	10	running	5	morningrun	3	nikerunningpic	19	adidasrunning	1
running	8	blessed	10	adidasrunning	4	run365miles	3	nikepluspic	16	nrc	1
adidasrunning	7	eveningcardio	9	madrid	4	getoutthere	2	nikerunning	5	runfra	1
heretocreate	6	nikerunning	5	Standardmarathon	3	amsterdammarathon	1	justdoit	2	irunnation	1
semiparis	5	run	3	madridmarathon2016	3	healtylife	1	alwaysbar	1	sweatyselfie	1
adidasrunners	5	nike	3	adidas	2	instarun	1	runwithhart	1	intervaltraining	1
whyirunparishttps	1	nikerunclub	2	ultraboost	2	runningzuidoost	1	bar	1	sweatbetter	1

Notes: LEGENDA:

Fr. = Frequency

Source: Authors’ elaboration

Table III.

Users’ level of activity

Community	% of atleast two tweets	% of all tweets
#Whyirunchampselysees	46.9	94.0
#Nrclondon	38.7	70.6
#Whyirunmadrid	31.8	72.9
#Werunamsterdam	48.1	73.4
#Nrcbcn	25.0	93.7
#Whyirunfrankfurt	62.5	88.0
Average % total	40.98	83.84

Source: Authors’ elaboration

Table IV.

The types of co-created experiences

		Individual or Communal Consumption Activity
		Individual Consumption Activity (Consumers use a product or benefit from a service while alone) 572 tweets (35.55%)	Communal Consumption Activity (Consumers use a product or benefit from a service with other consumers) 1037 tweets (64.45%)
Other actor/s influencing co-created experience	Brand (by interacting with member of brand community) 731 tweets (45.43%)	a) Brand’s Products Individual Usage Experiences 234 tweets (14.54%)	c) Brand’s Products Communal Usage Experiences 497 tweets (30.89%)	Strictly brand related consumption activities
	Other Members (by interacting with the individual consumers) 878 tweets (54.57%)	b) Auto-Celebrative Experiences 338 tweets (21.01%)	d) Collective Celebration Experiences 540 tweets (33.56%)	Social consumption activities
		Hedonism/ Individualism	Socialization

Source: Authors’ analysis from the sample of 1609 tweets