Introduction

Web Services 2.0, and social networks in particular, have completely changed and turned marketing around. Consumers interact directly with their favorite brands, create their own content and play a preponderant role in the Marketing strategy used on social networks. This is a new challenge for marketers who have to combine, in a balanced way, the digital world and traditional Marketing (^{Hanna, Rohm & Crittenden, 2011}). In fact, this new Web Paradigm - Web 2.0 - gave users the possibility to communicate with brands and allowed them to assume an active and participatory position. Bidirectional communication was strengthened with the emergence of social networks since any user can create and share content, as well as give his opinion. It is therefore crucial to realize how the public reacts to the brands’ communication strategy and be fully aware of the type of actions that will generate a stronger reaction.

In this context, engagement, one of the most important metrics for marketers, takes on special relevance. This concept, which requires a strong interaction with the consumer, is the embodiment of the person’s response to the brand and of the kind of involvement they establish with the brand in a digital environment. Therefore, it has become an essential element to consider when developing any Digital Marketing strategy. However, and despite the growing interest in consumer engagement towards brands and in the role played by social networks, academic research on what drives engagement is still scarce and at a very early stage of development (^{Schivinski, Christodoulides & Dabrowski, 2016}).

This type of empirical study is essential to guide Marketing professionals through social networks and assist them in developing communication strategies that have to take into account the feedback provided by their target audience. The originality of this study has to do not only with the use of a methodology based on observation and comparison of two distinct social networks (Facebook and Instagram), but also on the fact that it analyzes different brands of alcoholic drinks which are quite popular in Portugal. This project focuses on three brands of alcoholic drinks. This choice has a lot to do with the dynamism that characterizes the investment made by those brands in advertising over the last few years. This is a sector with legal limitations with regard to its "creative freedom" (^{Almeida, 2016}). However, social networks allow users to informally create content that will promote the brand anytime they share situations in which they are drinking alcohol or that depict behaviors one usually associates with drinking (^{Lobstein, Thornton, Jernigan & Landon, 2017}).

Taking all this into consideration, the research question asked was: "Which factors may influence consumers’ engagement with alcoholic drink brands on Facebook and Instagram?". Based on the Literature Review, it was possible to design some hypotheses that guided this research question and test the "key factors influencing engagement".

1. Literature review

The Web 2.0 concept began with Tim O'Reilly over a decade ago. The author defined it as a platform that contemplates applications which are constantly updated and aim to create a service people appreciate, use and to which they also contribute. While WEB 1.0 was focused on connecting computers, WEB 2.0 focuses on connecting people and creating services in which the public can participate:

Web 2.0 is the network as platform, spanning all connected devices; Web 2.0 applications are those that make the most of the intrinsic advantages of that platform: delivering software as a continually-updated service that gets better the more people use it, consuming and remixing data from multiple sources, including individual users, while providing their own data and services in a form that allows remixing by others, creating network effects through an "architecture of participation," and going beyond the page metaphor of Web 1.0 to deliver rich user experiences (^{O’Reilly, 2005}, p.13).

With the development of Web 2.0, and the ever growing participation of the user, social networks are one of the best examples of participatory culture. Social networks reflect the qualities of Web 2.0, enabling sharing, collaboration and content creation in an extremely simple and accessible way. In ^{Aggarwal’s opinion (2011}), a social network can be defined as “(…) a network of interactions or relationships, where the nodes consist of actors, and the edges consist of the relationships or interactions between these actors (…)” (p.2).

The number of users of social networks grows at a very fast pace, impelling Marketing to adapt: “Every day the influence of social media expands as more people join online communities. (...) If Facebook were a country, it would be the most populated in the world.” (^{Tuten & Solomon, 2017}, p. 5)

This more digital and more connected world brings new challenges to Marketing, forcing, for example, Marketing mix to include consumer participation (^{Sfetcu, 2017}; ^{Kotler, Kartajaya & Setiawan, 2016}). Social networks are an excellent opportunity to improve the relationships existing between the brand and its target audience and develop greater brand awareness and loyalty in current and future consumers. This new paradigm led to the emergence of the concept of Marketing in social networks, aka Social Media Marketing (SMM). ^{Felix, Rauschnabe & Hinsch (2016}) define SMM as an “(…) interdisciplinary and cross-functional concept that uses social media (often in combination with other communications channels) to achieve organizational goals by creating value for stakeholders” (p.6). These authors also defend that, on a strategic level, Social Media Marketing is related to an organization's decisions about marketing, culture, structure and governance.

In social networks, consumer behavior gains a key position within the Marketing strategy used, as claimed by ^{Kotler, et al. (2016}) in his concept of Marketing 4.0 (^{Verhoef, Kooge & Walk, 2016}). Engagement is so important that it has become a major influence on social networks. Marketers understand the importance of consumers interacting with their content, giving feedback and creating their own content. Social networks have focused on the non-transactional behavior of consumers and this context will undoubtedly favor engagement. ^{Vinerean, Cetina, Dumitrescu & Tichindelean (2013}) consider that “True customer engagement means commitment-focused, not transaction-focused. The companies that master this aspect are the ones that are truly successful” (p.77).

Marketers’ main concern should be with the loyalty of their consumers and they should no longer worry exclusively with sales, as Marketing 1.0 used to recommend. Engagement is the first step towards loyalty: a consumer who interacts with the brand is someone who is interested in a given content and who will gladly spend his/her attention and time to give feedback to the brand. Authors like ^{Brodie, Ilic, Juric & Hollebeek (2013}) strived to create a definition of engagement that best fits virtual environment and that, at the same time, would embrace and include the different dynamics that existed before digital environment.

Consumer engagement in a virtual brand community involves specific interactive experiences between consumers and the brand, and/or other members of the community. Consumer engagement is a context-dependent, psychological state characterized by fluctuating intensity levels that occur within dynamic, iterative engagement processes. Consumer engagement is a multidimensional concept comprising cognitive, emotional, and/ or behavioral dimensions, and plays a central role in the process of relational exchange where other relational concepts are engagement antecedents and/or consequences in iterative engagement processes within the brand community (^{Brodie et al., 2013}, p. 107).

Social networking engagement is reflected in the "likes", comments or shares: interactive experiences between consumers and the brand, motivated by emotional and behavioral variables. However, being present on the different social networks is not enough to generate engagement: "Engagement is not produced by simply being ' on ' a certain kind of media, digital, social, mobile or otherwise. Marketers must design specific experiences using these media to make the brand-life goal (s) connection and elaborate on it." (^{Brodie, Hollebeek & Conduit, 2016}, p. 100). Users’ behavior on social networks can include several actions: simply visualizing a given content produced by a brand, acting as content curator, contributing to social media content or even producing contents on social networks. The last two behaviors were boosted by the development of Web 2.0 in which consumer is expected to play a new role in Marketing (^{Evans, 2010}).

^{Vries, Genslera & Leeflangab (2012}) noticed that there were different preliminary studies on marketing actions on social networks, but they also realized that information about the factors that influence the popularity of each brand was scarce. Hence, they developed a conceptual framework that determines five variables that will influence the popularity of a brand: vividness, interactivity, content of the brand post (information or entertainment), position of brand post and valence of the comments (Vries et al., 2012).

^{Vignisdóttir (2017}) also used this framework to develop a research, but on Instagram. The author used makeup brands and outlined a more specific goal: to see whether or not there is a difference in the level of engagement generated by content created by the brand versus content created by users ("regrammed”).

2. Methods

To answer the research question, we chose a mixed methodology based on the observation of Facebook and Instagram pages used by alcoholic drink brands. We used a case study, with multiple cases, and followed a descriptive research. An observation grid was used to collect and analyze data that was gathered using a non-participant observation method. The study's hypotheses were:

H1: Posting frequency leads to greater engagement, both on Facebook and on Instagram.

H2: Posts on Instagram have higher engagement rates than those posted on Facebook.

H3: High level of vividness leads to greater engagement on Facebook than on Instagram.

H4: The higher the level of interactivity, the greater the engagement generated by the brand post both on Facebook and on Instagram.

3. Results

During the analysis period, 256 posts were collected from the 3 brands under analysis: 141 were obtained from Facebook and 115 from Instagram. In Table 5 it is possible to observe that during the 120 days of the analysis, the number of daily posts differs from one brand to another and from one social network to another.

The results obtained during the analysis of 256 posts related to the three brands of alcoholic drinks were quite different as far as social platforms were concerned, and there were often significant differences between Facebook and Instagram. The answers to the hypotheses set earlier are presented below.

H1: Posting frequency leads to greater engagement, both on Facebook and on Instagram.

Post frequency of the three brands of alcoholic beverages was directly related to each one of the brand’s accumulated engagement, both on Facebook and on Instagram. The Sagres Brand had a very irregular and reduced posting frequency, thus the levels of accumulated engagement associated with that brand are much lower compared with those related to Licor Beirão and SuperBock. However, Licor Beirão, having had a lower posting frequency on Facebook, obtained higher accumulated value of reactions, comments and shares on this social network. This may have happened because the brand has more followers on the page than SuperBock.

On Instagram, Licor Beirão revealed a higher posting frequency and scored higher accumulated values in all indicators: engagement rate, comments and likes.

The hypothesis was proven because the results show that posting frequency has a positive impact on engagement, both on Facebook and on Instagram.

H2: Posts on Instagram have higher engagement rates than those posted on Facebook.

Several authors argued that Instagram has better engagement results than Facebook, as people on Instagram follow the brands more frequently and are more involved and engaged (^{Phuaa et al., 2017}). In fact, this study is in accordance with these conclusions: the average engagement rate on Instagram was 4.4%, while on Facebook it was only 1%,

The higher engagement rates on Instagram are due to a greater number of likes. At the same time, evidence shows a smaller number of comments. These results are in line with those of ^{Vignisdóttir (2017}). It should also be noted that, while Instagram has better engagement rates, Facebook remains the social network with more users. In Portugal 74% of people continue to use Facebook as their favorite social media and only 8.9% of the Portuguese population choose Instagram as their favorite platform (^{Marktest Consulting, 2017}). Overall, Instagram scored higher engagement rates, but Facebook scored higher absolute engagement.

This hypothesis was supported for the three brands of alcoholic beverages which scored higher engagement rates on Instagram than on Facebook.

H3: High level of vividness leads to greater engagement on Facebook than on Instagram.

These three brands of alcoholic drinks chose to publish more photos than videos on both social networks: only 11% and 3% of the posts were considered HIGH LEVEL posts, on Facebook and Instagram, respectively. It has to be noted that, although the number of videos posted was quite low, they have received excellent feedback, and the responsible Marketing team was highly praised. On both social networks, engagement was significantly higher in high-level posts (higher average number of comments and shares on Facebook and higher average number of likes and comments on Instagram). However, there are only 3 high-level posts on Instagram. As a consequence, engagement mean values are more influenced by outlier values.

On Facebook, the results showed that high level of vividness in a post triggers more comments and more shares, but does not have the same impact on reactions.

On Instagram, photos and images obtained more consistent engagement results, with satisfactory engagement rates and high numbers of likes. On Facebook, videos obtained better engagement results as far as the average number of comments and shares are concerned, but not as far as reactions are concerned. The results are consistent with the studies conducted by Chua and Banerjee (2013) and ^{Vignisdóttir (2017}).

The results show that the general trend should be to use video posts for Facebook and photos and images for Instagram, taking into account the characteristics of both networks. This hypothesis was partially supported by the results.

H4: The higher the level of interactivity, the greater the engagement generated by the brand post both on Facebook and on Instagram.

Contrary to what was anticipated, high level of interactivity (questions or quizzes) did not lead to better public engagement. That higher engagement was provided by medium-level posts (call to act or contests).

These results are in line with the study carried out by ^{Tafesse (2015}) which also concluded that when a post uses too many interactivity tools, chances are that it might drive target audience away. This author recommends marketers to use a moderate level of interactivity. Questions cannot be answered with a reaction or a like but require the use of comments, for instance and such requirement can inhibit people from interacting with the content because this interaction would require them to spend more time and be more motivated.

In fact, both on Facebook and Instagram, the best results were found in posts with a medium level of interactivity and for virtually all activities (comments, reactions and shares).

The hypothesis developed was not supported, since the highest level of interactivity did not lead to the highest level of engagement. The best results were obtained by posts with a medium level of interactivity, both on Facebook and on Instagram. It is recommended to use posts with this level of interactivity in order to foster call to act strategies and promote users’ action by making him interact or even co-create posts.

4. Discussion

The results made it possible to identify some factors that can influence engagement on Facebook and Instagram for those three brands and to compare the differences between these two social networks. In short, the results obtained proved that, in response to the question “Which factors may influence consumers’ engagement with alcoholic drink brands on Facebook and Instagram?”, there are several factors that may influence engagement on these two social networks and some of those factors will have a different impact on Facebook and on Instagram.

First of all, one of the factors that influence engagement towards these alcoholic drink brands is the social network itself: while on Facebook results showed higher absolute engagement (greater number of reactions, comments and shares), on Instagram there seems to be a higher engagement rate (higher percentage of followers involved with the brand). Results show that both social networks are each individually important: these brands’ Facebook page has a much larger number of followers than their Instagram page and has much higher reactions and comments rate values as well. However, the percentage of followers interacting with the posts is much lower compared to Instagram. These results corroborate those obtained by ^{Vignisdóttir (2017}), who also noted that users in this social network choose to express their opinion through likes, instead of comments.

Posting frequency proved to be an influence on engagement on both social networks and exercises similar influence. On both social networks, it is recommended that communication be consistent and frequent in order to boost engagement. These results agree with other studies (Marktest Consulting, 2017; ^{Simões & Dias, 2017}), as they concluded that the frequency of publications is a fundamental point in the strategy of social media communication.

As far as the degree of vividness is concerned, the results related to each of the social networks were quite distinct. On Facebook, high-level posts (videos) got more shares and more comments, and low-level posts (image) got more reactions. On this social network, video posts got higher peripheral engagement and a more complex interaction (shares or comments), while images got more direct interaction (reactions). On the other hand, on Instagram low-level posts showed more consistent results than high-level posts. The results are consistent with those obtained by ^{Chua & Banerjee (2015}) and ^{Vignisdóttir (2017}). Generally speaking, it is recommended to use more videos on Facebook in order to increase peripheral engagement and photographs for more direct interaction. On Instagram, photos brought more consistent results than video.

As ^{Chua & Banerjee (2015}) had already concluded in their study, our investigation also shows that, in relation to the posts interactivity variable, the high-level posts (which include quiz or questions), achieved poorer success in both social networks. Medium-level posts, which include call to act, were those that achieved greater engagement on both social networks. Indeed, the user cannot answer a question or a quiz with a simple "like". This type of posts requires more complex interactions and more of the user’s time. This may inhibit the fan’s interaction with the brand. Therefore, these brands should place more importance on posts with call to act strategies than on questions.

Conclusions

The emergence of social networks has led to the renewal of some of the Marketing practices in use. With Web 2.0, the public began to have a voice, an opinion and a way of engaging directly with the brands. Bidirectional communication brought new threats and new opportunities and may bring major benefits to the brands: the creation of engagement with their followers and their target audience in a way that traditional Marketing could not provide.

As far as alcoholic drink brands are concerned, this is an interesting case study, since communication on social networks is very different from that used in other activities. On the one hand, we have limits on advertisement and creative freedom, and on the other hand we have brands which are investing more and more in advertising and claiming their rightful place among the brands with more interactions on Facebook (^{Almeida, 2016}; Marktest Consulting, 2017).

The results emphasize the need to incorporate a communication strategy on Facebook and Instagram. This should be done in an integrated way, but taking into account the potentialities and limitations of each social network. To maximize the public's response, marketers should take into account fan motivations in each social network and be able to adapt their strategy, without losing the coherence and values of the brand.

The study showed evidence of how the user plays a preponderant role in the new communicational dynamics. Fan involvement in the communication process is the key point to build creative synergies that will improve engagement and create a brand personality that will stand out on social networks.

One of the main limitations of this study is that the statistics related to each post are not susceptible to public scrutiny. Further research can possibly be used to understand the reason why people use social networks with such different purposes and what they expect brand communication to achieve on each of those networks. Another concern would be to realize how the different engagement actions (likes, shares, comments) correspond to different levels of interaction and brand engagement.

This project does not cover everything there is to say about social network engagement, since it focuses merely on a specific case study and on a fixed analysis period. Social networking engagement will continue to be one of the most important metrics in the context of social networks and should continue to be extensively investigated through the effective use of various and different approaches.