The Role of Online Brand Community Engagement on the Consumer–Brand Relationship

Abstract

The purpose of this study is to analyze the effects of online brand community (OBC) engagement on strategic aspects for the brand supporting the community. A total of 628 valid responses were collected through a self-administered questionnaire. The authors tested the proposed model using structural equation modeling. The results show that OBC engagement directly favors participation in the community, willingness to co-create with the brand, and positive sord-of-mouth (WOM); it also has an indirect positive influence on brand loyalty. At the same time, OBC engagement is directly influenced by OBC identification and, through this, indirectly influenced by brand identification. Likewise, brand trust directly influences brand co-creation, loyalty, and positive WOM. However, OBC participation has been shown to have no significant effect on brand co-creation and positive WOM.

1. Introduction

An online brand community (OBC) is a user group created online around a brand. OBCs have become a site in which users can interact with brands and participation in OBCs can generate strong, lasting emotional ties between consumers and brands [1,2]. Thus, analyzing the effects of OBCs on brand-related variables is one of the most useful research topics in online marketing [3].

In line with this, different OBC-related variables and motivating factors lead to positive behaviors among members of brand communities which, in turn, may affect other brand-related constructs, even when they may not be specifically related to the OBC [3,4]. Therefore, OBC management is key for organizations intent on creating and consolidating online customer relations [2,5,6,7].

Engagement is essential to the understanding of OBC dynamics. Previous studies have found that engagement remains key to explaining both participation in an online community and its effects on the brand it supports it [5,7,8,9], and it is key to the survival of OBCs [10]. In spite of the importance of brand community engagement, this concept remains poorly understood in brand community research [11]. Moreover, the Marketing Science Institute (MSI) has repeatedly called for further research to be carried out into the factors influencing consumer engagement through online processes [12,13], including OBCs [11,14]. Similarly, motivations of a more personal nature, in which customers may have to engage with OBCs, merit further research [15].

The identification of users with a brand and its OBC, as well as their trust in these brands and OBCs, remain key variables to capitalize on the benefits of the use of OBCs from a brand perspective. These variables require further research with regard to their role in influencing consumer behavior in the context of OBCs [14,16,17].

Therefore, existing research on consumer behavior in OBCs needs to overcome some important challenges. Here, this study highlights some concrete issues and current research gaps in this context which require attention [11,14,17]. First, there is a need to establish more integrated models [18] specifying the consequences of interactions in the communities [6,14,19]. More specifically, there is no consensus on the sequence of relations arising between OBC-related variables and brand-related variables [20]. Second, the context that produces a co-creative attitude among customers and affects their post-purchase behavior has still not been sufficiently clarified [21,22,23]. Third, the role that both OBC engagement and participation have in explaining strategic aspects for the brand supporting the community (i.e., brand co-creation, brand loyalty, and positive word-of-mouth (WOM)) still have to be discussed and analyzed [17,24].

In line with the above, this study seeks to address the following research objectives:

• Developing an integrated model able to explain and predict how engagement and participation in OBC develop leading to support of the brand beyond customer loyalty.
• Defining the role of variables, such as identification and trust, in the way that OBC-engaged users develop loyalty, participate in value co-creation, and deliver positive WOM.

In order to attain these objectives, this study establishes and empirically verifies a comprehensive model dealing with both OBC-related variables (i.e., OBC identification, OBC engagement, and OBC participation) and brand-related variables (i.e., brand identification and brand trust) that have usually been considered in previous studies focusing on concrete relationships between variables within the OBC context. Thus, the proposed model is thought to ponder, in greater detail, the influence of OBCs on the consumer–brand relationship than previous, more partial, and thus limited contributions. In order to achieve this, the theoretical model posited here was evaluated using partial least squares structural equation modeling (PLS-SEM) with data from the survey carried out as part of this study, which secured 628 responses from active users of OBCs.

The paper is structured as follows: First, it explains the theoretical framework adopted and the development of the hypotheses used. Next, the methodology and results are explained. Finally, the conclusions and theoretical as well as practical implications of the findings of this study are discussed.

2. Background and Hypotheses

2.1. Identification with the Community

Identification with the OBC is defined as the comparison users make between their own identity and that of the community [25]. Thus, it reflects the strength of the relationship between the consumer and the community [26].

This identification process has been analyzed in various studies [26,27] mainly using the social identity theory approach. This theory defines identification as an individual’s perception of real or symbolic belonging to a specific group [28].

Brands facilitate consumer integration into groups with common characteristics [29]. To a certain extent, a consumer’s identity is projected through the brands they consume [30,31], so their brand identification will be greater when the brand image more closely matches their own self-concept [32].

The process of identification with the brand is related to identification with the OBC. If a consumer identifies with a brand, the process of identification with the OBC will be easier [33]. Previous studies show the positive influence of brand identification as a precursor to OBC identification [17,26]. Shared brand experiences are the basis for generating identification with the OBC [34,35], and are even considered the main source of such identification [36]. In light of the above, the following hypothesis is proposed:

Hypothesis 1 (H1).

The greater the identification with the brand, the greater the identification with the online brand community.

2.2. Engagement with the Community

In the context of the OBC, engagement is understood as an intrinsic motivation for the consumer to interact and cooperate with members of the community [26,37]. These interactive experiences between consumers and brands [5,38] are expected to improve consumption and cooperation [1,39].

Possible factors leading to engagement include identification with the OBC, given that this influences behavioral intentions in the community [40]. Various studies show that greater identification with the group has a positive impact on motivation to participate in the group [41,42]. A user who identifies with the community will perceive his or her participation as congruent with his or her personal values [32] and will receive recognition from other similar individuals [30]; this generates engagement with the community [26,43]. The members of the OBC are likely to become involved, helping others and giving their support to achieving certain goals [44]; identification with the virtual community results in continued engagement in such behavior [7,45]. Thus, the following hypothesis is formulated:

Hypothesis 2 (H2).

The higher the consumer’s identification with the OBC, the higher his/her engagement with it.

2.3. Participation in the Community

Consumer participation is considered essential to ensuring an OBC’s continuity over time and facilitating lasting relationships [46,47]. Engagement, which can emerge at different levels of intensity over time [5], is one of the most important motivators for interactive participation in the community [48]. Further research [47] showed that engagement is not merely an attitude and that it requires the participation of OBC users. In fact, participation is a key outcome of engagement. Therefore, the degree of engagement with the OBC affects how actively members participate in it [48], i.e., high levels of engagement result in greater participation [17,49]. Additionally, parallel studies [49] assumed that OBC members participated only when they perceived that their engagement would result in their objectives being realized. As a result, engaged customers tend to contribute to the reputation of the brand and its long-term recognition by participating in OBCs and in other activities that support the brand [48]. In line with this, the following hypothesis is proposed:

Hypothesis 3 (H3).

The higher the consumer’s engagement with the OBC, the higher his/her participation in it.

2.4. Trust in the Brand

Generating trust in the brand has become an essential element in establishing long-term relationships between brands and their customers [50]. Trust is even more relevant to online communities, where consumers perceive a greater degree of risk in online relationships [51].

In this context, consumers’ identification with the brand can also impact on members’ trust in the brand around which the group revolves [1]. Identification with the brand encourages consumers to feel psychologically linked to the organization [32,52], noting that brand identification can have both a direct and indirect impact on brand trust. The authors of [53] consider brand identification to be the best way of developing trust as the parties do not have to calculate the likelihood of the outcomes from common interactions. Thus, the following hypothesis is formulated:

Hypothesis 4 (H4).

The higher the consumer’s identification with the brand, the higher his/her trust in it.

2.5. Consumer Willingness to Co-Create with the Brand

The rise in social media channels has facilitated the shared control of value creation processes between consumers and companies [54], arising from so-called value co-creation or co-production [55]; i.e., consumers participate with companies in creating market value associated with their offers [56].

OBCs are now positioned as key platforms for facilitating the co-creation process between brands and consumers [56,57]. In this context, OBCs have facilitated co-creation through, for instance, active debate among members or collaboration in creating and developing new products [58]. Online communities can also generate ideas for new products or changes to existing ones and become a source for brand product innovation [59].

There are various motives for members of online communities to collaborate and co-create with the brand. Engagement [22] is key here as it includes co-creation of value with customers, both in its relational aspect [5] and in terms of the service-dominant logic [60]. Consumer engagement is an essential variable for understanding the process of co-creating experiences in the digital context [5]; it allows individuals to broaden their traditional role in brand relations, actively participating in the creation and development of products in the community [61]. The co-creation process feeds off co-creative interactions and real commitment experiences for individuals, elements that can be obtained through engagement with the online community [22,62]. Thus, the following hypothesis is proposed:

Hypothesis 5 (H5).

The higher the consumer’s engagement with the OBC, the higher his/her predisposition to brand co-creation.

Brand trust is important in building beneficial exchange relationships [63]. Therefore, trust toward the brand can lead the consumers to cooperate with it in developing new products [50]. In this light, it is understood that a consumer has to reach a sufficient level of trust to be willing to co-create value with the brand; indeed, it is an essential requisite [64]. Specifically, in OBCs, this variable helps reduce perceived risks and opportunistic behavior in interactions, positively impacting on cooperative behavior among group members [50]. Thus, brand trust encourages the consumer, a community member, to participate in developing new products and sharing personal information with the company [50,65]. Thus, the following hypothesis is formulated:

Hypothesis 6 (H6).

The higher the consumer’s trust in the brand, the higher the consumer’s predisposition to co-create with the brand.

Co-creative behaviors might also be the result of the degree of members’ interaction, and, thus, participation in the brand community [41]. OBCs enable customers to interact with the company [66], facilitating the creation of collaborative company–customer groups that interact to co-create more interesting products and services [67]. For instance, member participation might result in the introduction of improvements, the development of new products [68], and in the generation of innovations [23]. Thus, the following hypothesis is suggested:

Hypothesis 7 (H7).

The degree of participation in the OBC positively influences the consumer’s willingness to co-create with the brand.

2.6. Brand Loyalty

Achieving customer brand loyalty is one of the major challenges for companies [69]. It is explained largely by their identification with the brand [40,70]; i.e., brand identification produces loyal behavior toward the brand [1,19,52,53] and facilitates the development of strong, lasting relationships with the company [33] which can even develop into passion for the brand [71]. Thus, the following hypothesis is formulated:

Hypothesis 8 (H8).

The higher the consumer’s brand identification, the higher his/her brand loyalty.

Willingness to co-create value with the brand can also be a key influence on consumer loyalty. Interaction between customers and the brand to co-create different products and services gives rise to bonds that affect loyalty [72]. Although this relationship seems logical, it has barely been researched in earlier studies.

As said above, OBCs are a suitable space for individuals to co-create experiences [57] and contribute to creating and developing new products [65]. Such content provision by OBC users has a direct influence on participants’ purchasing frequency [73]. In this context, it has been shown that value co-creation between customer and company becomes an important precursor to loyalty, demonstrating a significant relationship between both concepts [74]. This is understood to mean that user participation in co-creation processes in the community can lead to an increase in brand knowledge and, therefore, greater acceptance of subsequent innovations in the brand [68]. Consequently, maintaining these interactions over time can produce affective relationships between the parties, creating greater brand loyalty [66]. Thus, the following hypothesis is proposed:

Hypothesis 9 (H9).

The higher the community member’s willingness to engage in brand co-creation, the higher his/her loyalty toward the brand.

Trust is a key factor in facilitating long-term brand–customer relations [75]. Trust provides security between the parties involved in the exchange [76]. Hence, it would seem reasonable to posit that high levels of brand trust will result in favorable and loyal attitudes toward it [77]. In the online context, previous studies have identified the importance of brand trust as one of the most important direct conditions for loyalty [52,53]. Particularly in that context, where consumers are more uncertain in their interactions, trust in the brand becomes an essential concept for understanding long-term relations [78] and, therefore, loyalty [65,79] between the parties involved in the exchange processes. Consequently, the following hypothesis is proposed:

Hypothesis 10 (H10).

The higher the community member’s trust in the brand, the higher his/her brand loyalty.

2.7. Positive WOM

Word-of-mouth (WOM) is a key part of individuals’ interactions in online communities [80]. In online communities, which facilitate free, real-time information distribution to everyone, word-of-mouth (WOM) is a key part of individuals’ interactions [81]. Thus, it is of particular interest to explore the extent to which the OBC influences individuals’ willingness to transmit and disseminate positive brand information [82]. Research highlights the following important antecedents: trust [79,83], loyalty [83,84], engagement [85], and participation [2], among others.

Trust has a significant influence on people’s willingness to exchange information with others [34]. The greater the trust people have in the group atmosphere, the more likely they are to help others, share collective activities, and exchange information [86]. Thus, it may be assumed that community members’ trust toward the brand should motivate them to exchange information [50]. Various studies in the field of OBCs show the positive relationship between the generation of brand trust and members’ willingness to propagate positive WOM about the brand [3,79,87]. Thus, the following hypothesis is proposed:

Hypothesis 11 (H11).

The higher the community member’s trust in the brand, the higher his/her positive WOM.

In general, brand loyalty positively affects the dissemination of related favorable information. In the online environment, various studies suggest that positive WOM is one of the expected results of loyal behavior toward brands [83,84].

In the specific context of OBCs, the authors of [34] suggest that the fact that the community is made up of individuals loyal to a brand increases their likelihood of actively providing positive WOM for it. Loyalty to the brand sponsoring the community is strongly related to the intention to provide WOM [6,83]. Thus, the most loyal and motivated members of the community are most likely to participate in obtaining and disseminating positive information on the brand, both in and outside the community. Therefore, the following hypothesis is formulated:

Hypothesis 12 (H12).

The higher the community member’s loyalty to the brand, the higher his/her positive WOM.

Engagement with the community manifests itself as members’ interest in helping others or their voluntary support for the community, increasing value for all the parties involved [47,85] shows that consumer engagement covers value activities such as disseminating information on the company and its products. In this context, the authors of [88] suggest that the essential value elements that consumer engagement can provide for a brand include their influence on the behavior of other individuals through WOM. Consumer engagement increasingly influences non-transactional behavior such as WOM [48]. A consumer engaged with a brand community should show more positive reactions toward the brand [84] and propagate brand messages to other groups and consumers [5]. Thus, the following hypothesis is suggested:

Hypothesis 13 (H13).

The higher the consumer’s engagement with the OBC, the higher his/her positive WOM.

The community benefits from the exchange of knowledge, rapid transmission of ideas, and emotional support [89]. Consumers who participate in an OBC can develop greater ability, competence, and productivity in creating positive information about the brand and in disseminating it to other users [90]. Consumer participation in brand communities thus ought to influence their brand-related WOM [21,23,91] and intention to recommend the brand [2,92]. Therefore, the following hypothesis is formulated:

Hypothesis 14 (H14).

The higher the consumer’s participation in the OBC, the higher his/her positive WOM.

Figure 1 below provides a visual representation of the research model formulated for this study and includes the hypotheses outlined above.

3. Methodology

3.1. Sample

The data were collected by surveying OBC members. Participants in the study had to be internet users and members of at least one OBC, which they had accessed during the previous 10 weeks, irrespective of their level of activity. To increase the rate of participation in the survey, interviewees were entered into a draw for a tablet. The total number of valid questionnaires was n = 628, with a rather balanced gender distribution between men (49.4%) and women (50.6%). Questionnaires were completed mainly by university students; the main age ranges were 17–20 (47.5%) and 21–30 (50.9%).

Table 1. Sample characteristics (n = 628).

Characteristic	Count	%
Gender
Male	310	49.4
Female	318	50.6
Age
≤20	298	47.5
21–30	320	50.9
≥31	10	1.6
Net annual household income (€)
<1200	251	40
1201–1800	161	25.5
1801–3000	119	19
3001–5000	64	10.2
>5000	33	5.3
Community membership sector
Food and beverages	35	5.6
Culture	17	2.7
Sport	112	17.8
Retail	35	5.6
Electronics	55	8.8
Fashion and shoes	170	27.1
Motor	24	3.8
Leisure	68	10.8
Tourism and hospitality	24	3.8
Other	88	14

below shows a detailed outline of the characteristics of this study’s sample.

Respondents were asked to answer the questionnaire while thinking about the OBC in which they participated and, if there was more than one, then that in which they participated most. The average degree of participation was 4.16 (SD = 1.557) on a Likert scale from 1 (“I haven’t participated at all”) to 7 (“Very often”). The number of communities indicated by respondents exceeded 260 and belonged to a wide variety of sectors (e.g., sport, fashion, technology, etc.) and brands.

3.2. Questionnaire

The measurement scales of the variables were operationalized by adapting scales previously validated in the literature. Brand identification, OBC identification, OBC participation, and brand loyalty were measured adapting the scale of [26]. OBC engagement and brand trust were operationalized by adapting the scales of [4]. Willingness to co-create with the brand was measured using the scale of [50], and positive WOM was adapted from [84]. All items were measured on a 7-point Likert-type scales (from 1 “completely disagree” to 7 “completely agree”).

4. Results

4.1. Analysis of the Measurement Model

The model was assessed with a confirmatory factor analysis using structural equations (LISREL). The overall goodness of fit of the model and the quality of the used measurements were evaluated with verification of their one-dimensionality, reliability, and convergent and discriminant validity. The estimation method used is robust weighted least squares (RWLS) given the model conditions (e.g., non-multinormal distribution of data, rating scales, and the use of a polychoric correlation matrix). This estimation method is recommended as the most appropriate means of dealing with the relevant shortcomings and providing proper solutions (see [93]).

Prior to analyzing the measurement model, a Harman’s single-factor test was performed to assess the impact of common method bias (CMB) [94]. The total variance of the first factor was 25.6%. As a single factor could not be found responsible for most of the variance in the model, the Harman test would appear to suggest that CMB is not an issue as regards the analysis of these data.

It was verified that the model was correctly identified, that its degrees of freedom were above zero, that error variances were significant and positive in every case, and that the (standardized) parameter estimations all gave values of over 0.5 [95]. Joint confirmatory factor analysis (CFA) of all measurement scales provided satisfactory results, suggesting a good fit for the model (χ²/df = 2.363; GFI = 0.932; RMSEA = 0.047; CFI = 0.972; TLI = 0.966; NFI = 0.953; IFI = 0.973).

The reliability and convergent and discriminant validity of the considered measurement scales were also verified. Their reliability was assessed by analyzing the internal consistency of each construct. The values obtained for each proposed scale were adequate with over 0.7 for Cronbach’s alpha and acceptable item–total correlation values greater than 0.3 [96].

In accordance with [97], to analyze the convergent validity, it was verified that the indicator loads with latent variables were significant and above 0.5. However, a potential problem of loadings of items 1 and 2 of the OBC identification was identified, so these two items were dropped.

The average variance extracted (AVE) was also analyzed to confirm the convergence of the model scales [82], obtaining satisfactory results for all of them (see

Table 2. Measurement scales. Lambda loadings and reliability.

Construct and Items	Factor Loadings
Brand identification (CA: 0.837; CR: 0.841; AVE: 0.639)
This brand says a lot about the kind of person I am.	0.825
This brand’s image and my self-image are similar in many respects.	0.814
This brand plays an important role in my life.	0.758
OBC identification (CA: 0.838; CR: 0.798; AVE: 0.570)
The friendships I have with other brand community members mean a lot to me.	0.707
If brand community members planned something, I’d think of it as something “we” would do rather than something “they” would do.	0.717
I see myself as a part of the brand community.	0.835
OBC engagement (CA: 0.805; CR: 0.857; AVE: 0.599)
I benefit from following the community’s rules.	0.764
I am motivated to participate in the activities because I feel good afterwards or because I like it.	0.789
I am motivated to participate in the community’s activities because I am able to support other members.	0.782
I am motivated to participate in the community’s activities because I am able to reach personal goals.	0.761
OBC participation
How often did you participate in activities of your online brand community within the last ten weeks? 1: I haven’t participated at all; 7: Very often.
Willingness to co-create with the brand (CA: 0.968; CR: 0.970; AVE: 0.890)
I am willing to work with this brand to design new products.	0.921
I am willing to co-develop products/services with this brand.	0.955
I am willing to co-design products/services with this brand.	0.956
Overall, I am willing to cooperate with this brand in developing new products/services.	0.941
Brand loyalty (CA: 0.893; CR: 0.896; AVE: 0.742)
I intend to buy this brand in the near future.	0.877
I would actively search for this brand in order to buy it.	0.896
I intend to buy other products of this brand.	0.809
Brand trust (CA: 0.835; CR: 0.842; AVE: 0.641)
My brand gives me everything that I expect out of the product.	0.769
I rely on my brand.	0.871
My brand never disappoints me.	0.757
Positive WOM (CA: 0.900; CR: 0.904; AVE: 0.760)
I am going to spread positive WOM about the brand.	0.903
I will recommend this brand to other customers.	0.896
I will point out the positive aspects of this brand if anybody criticizes it.	0.813

Note. CA: Cronbach’s alpha; CR: composite reliability; AVE: average variance extracted.

).

Furthermore, confidence intervals were used to verify the correlation between latent variable pairs. None of the intervals obtained include a value close to unity, demonstrating the discriminant validity of the scales (see

Table 3. Analyses for discriminant validity.

	BI	OBCI	OBCE	BCC	BT	BL	PWOM
BI	0.799
OBCI	(0.609–0.741) 0.675	0.755
OBCE	(0.476–0.624) 0.55	(0.678–0.802) 0.74	0.774
BCC	(0.377–0.521) 0.449	(0.235–0.403) 0.319	(0.294–0.45) 0.372	0.94
BT	(0.409–0.561) 0.485	(0.217–0.401) 0.309	(0.249–0.425) 0.337	(0.445–0.577) 0.511	0.800
BL	(0.458–0.598) 0.528	(0.216–0.392) 0.304	(0.278–0.446) 0.362	(0.475–0.599) 0.537	(0.68–0.78) 0.73	0.861
PWOM	(0.4–0.548) 0.474	(0.457–0.429) 0.343	(0.336–0.496) 0.416	(0.407–0.539) 0.473	(0.71–0.80) 0.759	(0.699–0.787) 0.743	0.872
OBCP	(0.347–0.607) 0.31	(0.416–0.688) 0.359	(0.646–0.902) 0.503	(0.159–0.407) 0.28	(0.134–0.398) 0.173	(0.169–0.425) 0.193	(0.251–0.503) 0.245
MSV	0.455	0.547	0.547	0.288	0.576	0.552	0.576
ASV	0.256	0.219	0.237	0.184	0.263	0.273	0.097

Note. Correlations between constructs and 95% confidence intervals in brackets. Diagonal values in bold are the square root of each construct’s AVE. BI: brand identification; OBCI: OBC identification; OBCE: OBC engagement; BCC: brand co-creation; BT: brand trust; BL: brand loyalty; PWOM: positive WOM; OBCP: OBC participation; MSV: maximum shared variance; ASV: average shared variance.

).

4.2. Structural Model Testing

To perform the structural analysis of the global model, the sample was randomly divided into two subsamples: one to estimate the proposed model structure and the other to then confirm its validity. This provides greater guarantee of the validity of the model.

To contrast the causal relationships of the structural model, this study began by analyzing the fit measures of the global model, followed by the structural parameters. The overall fit indices of the structural model for the estimation subsample were reasonably acceptable (χ²/df = 2.042; GFI = 0.91; RMSEA = 0.058; CFI = 0.955; TLI = 0.47; NFI = 0.915; IFI = 0.955). As for the significance of the estimated parameters, only two relations raised were non-significant: OBC participation → brand co-creation (H7) and OBC participation → positive electronic WOM (eWOM) (H14).

The study reformulated the model excluding the relationships of OBC participation with brand co-production and positive eWOM. The modification rates offered by the SEM software used were also examined previously. Though this would be a pure data-driven recommendation which, then, based on our theory-driven approach, would require subsequent theoretical support, and there was no need to suggest adding any other causal relationship into the model.

The fit indices for the new model were slightly improved compared with the previous results, showing a better fit to the data (χ²/df = 2.026; GFI = 0.917; RMSEA = 0.057; CFI = 0.955; TLI = 0.48; NFI = 0.915; IFI = 0.955). According to the hypotheses proposed, all signs of structural coefficients showed a positive relationship between exogenous and endogenous constructs.

After estimating the model, it was tested it with the confirmation subsample. In this case, the same results were obtained in terms of significance for the structural coefficients; i.e., all except the relationships of hypotheses 7 and 14 were significant (Figure 2). The model was reformulated again excluding these two relationships. In this case, the structural model fit indices for the confirmation subsample were also quite acceptable, with similar values to those in the estimate subsample (χ²/df = 1.757; GFI = 0.92; RMSEA = 0.049; CFI = 0.968; TLI = 0.63; NFI = 0.93; IFI = 0.968).

5. Discussion

5.1. Theoretical Implications

This study proposes and empirically verifies a comprehensive model which considers both OBC-related variables (i.e., OBC identification, OBC engagement, and OBC participation) and brand-related variables (i.e., brand identification and brand trust) to explain their effects on three strategic aspects of the brand (i.e., brand co-creation, brand loyalty, and positive WOM).

The present study makes several contributions to the existing literature on OBC. First, according to the literature review, just 5 out of 14 proposed hypotheses (H1, H4, H8, H10 and H14) have been discussed and tested by previous studies, though not altogether but mostly separately and without considering the others. Therefore, apart from the new hypotheses analyzed, the integrative approach of the proposed model is one of the main contributions of this study.

Second, this study shows the importance of engagement in understanding both participation in an OBC and the results in terms of strategic aspects of the brand. User engagement with online brand communities positively influences their participation in them; this corresponds with the results from previous studies [49,98]. Moreover, the willingness of community members to co-create with the brand, their brand loyalty, and their tendency to provide positive WOM are partly determined by their engagement with the online brand community; this is an important finding because it also integrates and demonstrates the joint effects of these variables. Specifically, engagement with the online brand community exerts a direct, positive influence on community users’ willingness to co-create with the brand [5,22,61] and provide positive comments (positive WOM) [85]. In addition, this study shows that an OBC member’s willingness to co-create with the brand, together with a predictable stronger brand trust effect as demonstrated in other study contexts, also determines the member’s brand loyalty. This represents an important contribution in the contexts of OBCs if one considers that there has generally been little analysis of the relationship between co-creation and loyalty to date [74].

Third, regarding engagement antecedents, the results also show that identification with an online brand community increases the consumer’s engagement with the community, which is in line with the literature [7,17,45]. This is also consistent with findings from previous research that establish engagement as a mediator in the relationship between identification with and participation in an OBC [41,47]. In addition, in line with recent studies [17,35], this research confirms that members’ identification with the brand sponsoring the community has a positive influence on their identification with the community itself.

Fourth, with regard to brand identification and trust, results confirm that these are significant variables in understanding the behavior of OBC members toward the brand; this is congruent with related conclusions from other studies [1,3,19,65,79]. Specifically, identification with the brand has a positive effect on brand trust and loyalty. Furthermore, loyalty together with willingness to co-create with the brand and positive electronic WOM are directly and positively influenced by brand trust. In addition, brand loyalty is an important factor in the propensity of the community members to provide positive WOM [6,83].

Lastly, this study was not able to corroborate the relationship between community participation and the direct consequences proposed in the conceptual model: predisposition to co-create with the brand and provide positive WOM on the brand. The logical explanation for this result, corroborated by a validation sample, could prove to be another useful contribution of this study: the degree of the member’s participation in an OBC does not, to any extent, directly or indirectly influence outcomes such as brand co-creation, loyalty, or positive WOM. On the contrary, members’ engagement with the community, regardless of their participation level, was found to be the only specific OBC-related variable with a significant effect.

5.2. Practical Implications

OBCs can contribute to the achievement of strategic benefits for brands, such as brand co-creation, loyalty, and positive WOM. It is not enough to create an OBC and promote the participation of users; it is also fundamental to foster identification and trust with the brand that sponsors the community as well as identification and engagement with the OBC.

Firstly, companies can positively influence the identification of consumers with the brand by projecting the personality traits of the brand, values or philosophy that are in tune with what consumers themselves want to project. For example, brands such as Red Bull might favor their experiential and adventure traits, while others such as Nike might focus on goal achievement. In both cases, it will be possible for consumers who possess these features, or who aspire to have them, to feel more identification with the brand. This identification must be one of the priority objectives of the brands since getting strategic results depends to a great extent both directly and indirectly on brand trust and identification with the OBC.

Secondly, managers must promote trust in the brand. Trust is a key factor in reducing perceived risks and uncertainty, which will allow the development of lasting exchange relationships. More specifically, this study showed that trust has a positive and direct influence on co-creation, brand loyalty, and positive WOM. In order to achieve trust in the brand, organizations should act honestly, credibly, reliably, and responsibly. In this sense, it is important for organizations to fulfill promises made to the consumer and to manage their brands competently and in a transparent manner.

Thirdly, organizations should encourage identification with the OBC. In order to achieve this, developing a community sense is key with a clear set of shared norms, values, behaviors, knowledge, and emotions related to the brand. Individuals will want to be part of the community if they perceive that they have things in common with others. This will facilitate a sense of belonging in the community and higher levels of identification with it. Brands should manage their communities dynamically so that their members are the ones who build those communities through the exchange of experiences, interactions, and collaboration. Hence, collective identity is reinforced by the value that users perceive from their belonging to the community.

Fourth, the results of this research indicate that OBC engagement is a key aspect that positively influences brand co-creation, loyalty, and positive WOM. Therefore, managers should support and nurture OBC engagement. This could be achieved, for instance, by promoting user participation in community activities, encouraging members to assist each other when necessary, and rewarding, recognizing, or compensating, in some way, the activity of those who support the community. For this to be possible, it is necessary to provide the appropriate communication channels (e.g., forums, chats, and virtual events).

Finally, managers should provide tools for users to develop co-creation, loyalty, and positive WOM. For example, within the online community, users could be allowed to propose new designs and give their opinion on products through an application used to centralize a contest of ideas; re-purchase other brand products through direct access to the online shop and its catalogue; and recommend products in online forums, blogs, and scoring/recommendation systems.

5.3. Limitations and Future Research

This study has a number of limitations associated with a number of potential research opportunities. For instance, future research should explore the significance of other relationships between variables used in this model, including the potential impact of OBC engagement on brand loyalty or the effect of OBC identification on positive WOM and brand loyalty. Moreover, given the relevance of engagement in determining other brand-related variables, it would be interesting to evaluate, more broadly, the influence of this factor from a multi-dimensional approach by structuring its measurement in the dimensions that have been identified in the literature (e.g., cognitive, emotional, and behavioral) [99]. Similarly, perhaps the explanatory capacity of the model, even if it is good, could be expanded by including other variables that are specific to the OBC, such as OBC experience, recently adapted to the OBC context [17], or non-specific to the OBC, such as satisfaction with the brand [3,5,79]. Additionally, the heterogeneity of the OBCs of the sample used in this study—in terms of brands, sectors, product categories, etc.—would lend itself to replication through further research using more homogeneous samples. For instance, using samples of a small number of communities and products/services (e.g., high involvement vs. low involvement) and sectors (e.g., fashion, sports, technology). In addition, a multi-group analysis working with subsamples in terms of brand/product-related information could reveal nuances in the relationships of the proposed model and lead to potentially more insightful discussions. Finally, future studies could also use longitudinal data which would introduce dynamism into the evolution of the relationships.

6. Conclusions

This study has shown the importance of engagement in understanding both participation in an OBC and the results in terms of the strategic aspects of the brand. Engagement is a key variable in gaining members’ brand loyalty and in their willingness to co-create and recommend the brand. On the contrary, participation in the OBC does not seem to contribute to these brand achievements. This result not only emphasizes the importance of engagement in OBCs but also helps differentiate two concepts that sometimes appear to be intermingled in the literature of brand communities. Along with engagement, the other relevant variable in OBC management is identification of the user with the community. The sense of belonging to the group is an essential element for the success of the community, which is reflected in its strong effect on engagement.

Furthermore, this study concludes that the effect of engagement on key aspects of brand management is reinforced by two factors external to OBCs: brand identification and brand trust. The identification of the users with the brand exerts its influence on the origin of the process, which is why the success of the brand and, by extension, the OBC depends to a great extent on this. In our model, identification with the brand is not only essential for the achievement of identification with the OBC and brand trust but also directly and indirectly influences ulterior strategic aspects related to the brand. Furthermore, brand trust is a factor of extraordinary value that transcends the effects of the offline environment by reinforcing the effects of the online community on brand loyalty, co-creation, and positive WOM. These conclusions, therefore, are significant contributions to describing the influence of OBCs on the results of consumer–brand relationships.

Finally, according to this study’s review of the literature, just 5 out of 14 hypotheses posited here (H1, H4, H8, H10 and H14) have been tested by previous studies, though never together but mostly separately and without considering the other hypotheses. Apart from the new hypotheses analyzed here, the integrative approach of the model proposed here, which considers many relevant specific- and non-specific OBC-related variables altogether, is one of the main contributions of this study to knowledge on this topic.