Social media marketing: The influence of specific social media platforms on consumer engagement with the content posted on brand-owned social media channels

(1)

Social media marketing: The influence of specific social

media platforms on consumer engagement with the

content posted on brand-owned social media channels

Master’s Thesis

Author: Rik Polderman Student number: 1288 3557 Graduate School of Communication

Persuasive Communication Supervisor: Peter Neijens

Date: 26-06-2020

(2)

1

Abstract

Previous research has found out that the platform context has an influence on the

engagement with the content of the advertisement. However, previous research focused on paid social media marketing, whereas this research focused on social media marketing content posted on brand-owned social media channels. Therefore, this research intended to prove whether the findings for engagement with paid social media marketing content also applied to marketing content on brand-owned social media channels. Furthermore, the research explored the marketing objectives that can be achieved per platform if the engagement did significantly differ per platform.

The research established the different levels of engagement (passive and active) and three main types of engagement that are proven to take place with content posted on brand-owned social media channels: functional, emotional, and communal engagement. The research relied on quantitative content analysis in order to determine what types and levels of engagement were most prevalent on the three platforms chosen for the research

(Facebook, Instagram and YouTube). Once this was established it was possible to compare the mean amount of every type and level of engagement that took place on each platform. Furthermore, previous research successfully linked functional engagement to stimulating brand awareness, and the other two types to stimulating brand loyalty. Therefore, once the means were compared the research was able to determine what marketing objective each platform was most suitable for.

The main findings of the research were that the engagement with the same content varied across the three platforms. The same videos that were posted on all three social media platforms managed to evoke a different type of engagement on each platform.

Instagram had the highest amount of functional engagement, whereas Facebook evoked the highest amount of emotional engagement, and YouTube the highest amount of communal engagement. These findings determined that Instagram is a platform that is best suited for awareness campaigns, whereas the other two platforms are better suited for stimulating loyalty.

(3)

2

Introduction

The one-way communication relationship on traditional mass media has evolved into a two-way communication between brand and consumer on social media (Schivinski, & Dabrowski,

2016). Due to this evolution consumers are now able to share their brand experience with

other social media users. This is important for brands to acknowledge as it means that a brand can no longer have full control over brand narrative on social media platforms. To regain some control, brands try to stimulate positive way communication. Positive two-way communication, often referred to as customer interaction or engagement, can be stimulated through marketing content specifically made for and posted on social media. The interaction with the content is meant to evoke a type of response out of the consumer that would be beneficial towards the brand (Ashley & Tuten, 2015). This type of content is always brand initiated and is an open invitation for the consumer to interact and communicate with the brand on the brand’s social media channels.

Existing literature mainly focuses on the relevance and importance of social media marketing for brands. The umbrella term of ‘social media marketing’ is often used for all marketing activities across the multitude of social media platforms. This term could have generated the misconception that any activity, even posting the same content across all platforms, will be beneficial towards the brand. However, recent studies have discovered that a different social interaction takes place per platform. This difference in social interaction means that the consumer response to the advertisement can differ based on the context (the social media platform) it was placed in (Voorveld, van Noort, Muntinga, & Bronner, 2018). Furthermore, Voorveld and colleagues (2018) found a difference in audience advertisement evaluations per platform. Their data showed that Facebook and YouTube had the highest percentage of negative advertisement evaluations. They hypothesized that the

advertisements can be perceived as intrusive on these platforms, because they interfere with the content the user intended to watch and can only be skipped after several seconds. On the other hand, they found that Instagram advertisements were commonly perceived as

(5)

4

entertaining. The difference in audience interaction and advertisement evaluations per platform indicates that the audience perception of a social media advertisement can depend on the platform.

Even though these findings are intriguing to say the least, the research done by Voorveld and colleagues (2018) focused on paid advertisements within the digital space, which limits the generalizability of the findings to other forms of social media marketing strategies. This poses the question: Does the interaction with marketing content posted on brand-owned social media channels also vary across platforms? The content posted on these channels is intended to keep existing and potential customers actively engaged with the brand. Due to the voluntary decision to follow a brand-owned social media channel, the marketing content on it can be perceived as less intrusive (Van Noort & Willemsen, 2012). Therefore, this type of marketing could provide brands with an opportunity to avoid paid advertisements and the negative advertisement evaluations on platforms such as Facebook and YouTube. This is an important motive for the research’s aims to find out whether the platform also influences the engagement with the marketing content posted on brand-owned social media channels.

The research question that will guide this research will therefore be: To what extent does a specific social media platform influence consumer engagement with marketing content posted on brand-owned social media channels? How does this difference in

engagement impact the marketing objectives that can be achieved with the content on each platform? In order to answer these questions, the research will rely on a quantitative content analysis of the engagement with advertising content on brand-owned social media channels. The brand chosen for the analysis is Formula 1 (F1) and its’ Facebook, Instagram, and YouTube channels will be subject for the analysis. Facebook and YouTube were chosen because these are the platforms that got the most negative advertisement evaluations. Instagram will be included because it was the platform where audiences evaluated the content as entertaining (Voorveld et al., 2018).

(6)

5

The answer to the first part of the question should be able to build upon the findings of Voorveld and colleagues (2018) in two ways. First of all, it will determine whether specific social media platforms also have an influence on the engagement with the content on brand-owned social media channels. If the findings support Voorveld and colleagues’ (2018), then it would be interesting to find out how the platform specific engagement would impact the content’s ability to achieve specific marketing objectives. These findings should help determine whether a specific social media platform is more suitable for a specific type of marketing objective. If this is the case, then it could provide possible explanations for why Voorveld and colleagues (2018) found a difference in advertisement evaluations per platform for paid advertisements.

The second part to the research question is intended to explore the practical

implications of the research findings. The answer to this question should be relevant towards marketing departments and advertising agencies that create content for brand-owned social media channels to achieve specific marketing objectives. The findings of this research could extend Voorveld and colleagues’ (2018) findings to content on brand-owned social media channels. Therefore, the ability of a platform to achieve specific marketing objectives could be affected. If this turns out to be the case, then these departments and agencies have a reason to rethink the role of each platform in their marketing strategies. Furthermore, it could provide them with guidance in the creation or adaptation of content to tailor it to the specific marketing objective that can be achieved best per platform.

Theory

Marketing Through Brand-Owned Social Media Channels

Kaplan and Haenlein (2010) defined social media as a group of online applications that allow the creation and exchange of user-generated content. Two aspects of this definition are important to highlight to understand the relevance of marketing through brand-owned social media channels. Firstly, due to the competition between this ‘group of online applications’

(7)

6

the platforms have had to distinguish themselves from one another by designing each platform to stimulate the creation and exchange of a specific type of user-generated content. This means that the main purpose of using Instagram, for instance, would be the exchange of pictures; whereas for Facebook this would be written self-expression referred to as

statuses and the building of a unique user profile; and for YouTube it would be the uploading and consuming of user generated videos (Dictionary.com, 2020). This could be one of the main reasons for the differing user experiences per platform that Voorveld and colleagues (2018) found. However, despite this difference in platform design, users are still able to post similar (if not the same) content formats on all three. For example, Instagram and Facebook have created a support system for video content that is similar to that of YouTube (Hutton & Fosdick, 2011). Therefore, marketing departments can decide to upload the same content on brand-owned social media channels across all three platforms.

Secondly, since the creation and exchange of user-generated content is the fundamental principle of social media, brands can transform their role by participating in these activities. According to the definition by Kaplan and Haenlein (2010), when the brand creates and exchanges content it adheres to the principle that would make it a user of the social media platform rather than a ‘visitor’. The content posted by the brand through its account will always be within the context of the brand. This would mean that the brand is no longer pushing a branded message in a context made for consumers and their

conversations. Instead, it allows for branded messages to be conveyed to the consumer within a branded context. According to previous research, branded messages are generally perceived more appropriate within a branded context (Van Noort & Willemsen, 2012). Therefore, the advertising content released through brand-owned social media channels will more likely be perceived as appropriate by the consumer, therefore increasing the relevancy of this research.

The management of this type of content has been defined as the “practice of planning for the creation, delivery, and governance of useful, usable content” (Saji,

(8)

7

however, the intention behind the content remains to be relevant to the consumer. This content is mainly intended to spark an interaction between the consumer and the brand, which is made possible by the interaction design of the platforms. Facebook, Instagram, and YouTube all share similarities in the content interaction design such as the ability to like and share content. All three platforms also allow the user to comment through written expression on the content. However, the platforms also distinguish themselves with platform specific interaction design (e.g., Facebook emoticon reactions, the ‘thumbs up/down’ buttons on YouTube to a liking or express discontent towards the content, the heart-shaped like button on Instagram). Furthermore, ‘liking’ on Instagram can also be done by double tapping on the picture or video frame, whereas the other two platforms require a click on the designated button.

The combination of posting content and the consumer interacting with it creates a brand presence on social media, which can grow naturally through the attraction of followers. The advantage of this is that a brand’s content should naturally appear on a follower’s feed through platform coding, whereas with paid advertising it is paid to appear on someone’s feed. Besides the content’s ability to trigger interaction with engaged users, it can also even extent to non-engaged consumers. This occurs when a user that follows the brand is actively interacting with the content (sharing, commenting, liking), causing it to be featured on their connection’s feed as well (Jung, Shim, Jin, & Khang, 2016). This means that the content can indirectly lead to increased consumer loyalty through the continuous engagement that it stimulates and brand awareness by exposing non-engaged consumers to the brand’s content (Ashley & Tuten, 2015).

Consumer Engagement on Social Media

According to Dessart, Veloutsou and Morgan-Thomas (2015), engagement is a highly context-specific construct. Engagement within the context of marketing refers to the

(9)

8

consumer engagement as the psychological state within the consumer that has been triggered by an interactive and/or co-creative experience, that in the case of this research would be instigated by the brand. The psychological state of an engaged consumer can be any form of an emotional and/or cognitive response to the content. The definition of

engagement by Brodie and colleagues (2014) highlights the interdependence of user interaction and engagement. Interaction triggers engagement and being engaged with the content triggers interaction. Voorveld and colleagues (2018) found out that YouTube is mainly used to seek entertainment, whereas Instagram was used for filling empty moments (pastime) and Facebook for identification and topicality. These varying types of main reasons to use a platform should trigger very different types of interaction with the content. Due to the difference in the type of interaction someone is seeking with the content, there should also be a noticeable difference in the engagement with the content.

Engagement can vary in intensity (high versus low), reflect a valence (positive versus negative) and can be divided into different levels (passive versus active) (Dessart, et al.,

2015; Hutton & Fosdick, 2011). Muntinga, Moorman, and Smit, (2011) defined behaviours

such as watching brand related videos, reading the comments, and viewing brand related pictures as passive engagement. Whereas ‘liking’, commenting, and sharing are

engagement behaviours perceived to be a more active level of engagement. Liking would be a low intensity form of active engagement across all platforms, due to the low cognitive capacity needed to perform it. Active engagement such as commenting is a high intensity form of engagement with the brand content as it demands a higher cognitive effort to write out thoughts (Hutton & Fosdick, 2011).

Furthermore, engagement can be divided into the following types: functional, emotional, and communal engagement. Firstly, functional engagement is the type of engagement that would make the brand-initiated content function as a marketing tool (Lim,

Hwang, Kim, & Biocca, 2015). Some examples of functional engagement would include the

sharing and ‘liking’ of the content, and the use of hashtags in users’ comments. These forms of engagement behaviours with the content are used by social media algorithms to drive

(10)

9

reach (Lim et al., 2015). Tagging and liking are both considered a low intensity form of active engagement. Instagram allows users to double-tap a picture or video to like it, increasing the ease at which this can be done in comparison to the other two platforms. Therefore, on Instagram liking the content would be an even lower intensity form of active engagement in comparison to liking the content on the other two platforms.

Linking this back to Voorveld and colleagues’ (2018) findings that Instagram is used mainly for filling an empty moment (pastime activity) users could be more hesitant to

participate in higher intensity forms of active engagement such as commenting. Therefore, low intensity active engagement, and especially liking the content, is expected to take place most frequently on this platform compared to the other two. This would form the basis of the assumption that functional engagement would be the most dominant type of engagement to occur on Instagram. Therefore, the following hypothesis is proposed:

H1. If the content is posted on the brand’s Instagram channel, then it will have the highest amount of functional engagement compared to the same content on Facebook and YouTube.

Secondly, emotional engagement was defined as the use of social media interaction tools (Facebook emoticon reactions, comments) for the expression of emotion towards the content (Lim et al., 2015). Following the categorization outlined by Muntinga and colleagues (2011) this would be another type of active engagement, however this requires a higher intensity involvement with the content and the opinions of others in comparison to merely liking or sharing the content. Therefore, the emotion triggered by the content must be strong enough to transform someone from a passive engager to an active engager. Commenting especially would require more cognitive thought and mental energy to express an emotion. Voorveld et al. (2018) found out that YouTube was most often used to seek entertainment. The seeking of entertainment was indicated in Voorveld and colleagues’ (2018) research by the achievement of a positive emotion. The use of the platform to seek a positive emotional

(11)

10

experience, suggests that the response to the content on YouTube will more likely be of emotional nature. Therefore, the engagement will more likely be of the emotional type. Therefore, the following hypothesis is proposed:

H2. If the content is posted on the brand’s YouTube channel, then it will have the highest amount of emotional engagement compared to the same content on Facebook and Instagram.

Lastly, communal engagement is defined as the seeking of interaction with other members of the community to experience a sense of belonging to the community. This can range from a specific belonging to the fanbase of a team, or a more general belonging to the sport itself (Lim, Hwang, Kim, & Biocca, 2015). Seeking interaction on social media can only be achieved through commenting on a post to trigger a response from another user. In a comment one would express opinions and/or views on the topic of the content to find shared value with others who think alike. Additionally, one can like or comment on another users’ comment to express a shared or differing opinion. Following Muntinga and colleagues’ (2011) criteria, this would be an active level of engagement, which requires a higher intensity involvement with the content. Someone seeking interaction with others to experience a sense of belonging would most likely turn to Facebook, because the platform is mainly used for identification and topicality (Voorveld, et al., 2018). The comments on the content posted to Facebook are expected to be an attempt at finding shared value or seeking identification with the other followers of the brand, and so they would most likely be of communal nature. Therefore, the following hypothesis is proposed:

H3. If the content is posted on the brand’s Facebook channel, then it will have the highest amount of communal engagement compared to the same content on Instagram and YouTube.

(12)

11

Marketing Objectives: Brand Awareness or Brand Loyalty

Mihart (2012) discusses the importance of Integrated Marketing Communication (IMC) and suggests that evoking the desired reaction from the consumer is an integral part of effective marketing. Furthermore, Saji, Chauhan, and Pillai’s (2013) research discusses the influence of context (social media platform) and type of content (informative, entertainment, etc.) on the effectiveness of brand-initiated content. Content posted on brand-owned social media channels has a type of engagement that would be the intended reaction. If the brand cannot achieve the desired reaction on each platform due to the difference in engagement per platform, the question then would be: how would this impact the effectiveness of achieving the marketing objective?

The consumer journey usually starts with becoming aware of the brand, hence there is a marketing objective called ‘brand awareness’. The simple fact that a consumer knows the brand exists indicates a low level of brand awareness (Macdonald & Sharp 2000). On the other hand, when a consumer immediately thinks of a brand during the decision-making phase this is indicative of a high level of brand awareness (Macdonald & Sharp 2000). This is an important distinction to make as content on brand-owned social media channels is often aimed at increasing low-level awareness to high-level awareness (Felix, Rauschnabel, & Hinsch, 2017). Achieving a high-level awareness of the brand will increase the chance to stay top-of-mind amongst the audience. Being top-of-mind is linked to the inclusion of the brand in the consumers’ purchase consideration set (Macdonald & Sharp 2000).

It is important for a brand to be in the purchase consideration set of the consumer, because the next time the consumer needs the brand’s product they will choose from the brands in the consideration set. Once the initial purchase has been made the consumer will evaluate the product quality based upon user experience. The content on brand-owned social media channels is aimed at enriching the user experience by creating relevant and usable content. A user’s experience with the product is used to determine whether to buy the brand’s product again. Therefore, in case the social media content enriched the user

(13)

12

experience with the product, it would contribute towards establishing loyalty towards the

brand (Erdoğmuş and Cicek, 2012).

Coming back to the type of engagement and how it has an impact on the achieved marketing objective, Lim, et al. (2015) successfully linked emotional and communal engagementto stimulating brand loyalty, whereas functional engagement was better at stimulating awareness (Lim et al., 2015). These findings indicate that the type of

engagement can determine which marketing objective has been achieved on the audience (i.e. raising awareness, building loyalty, stimulating interaction).Therefore, if the findings show that a specific platform has the highest amount of emotional or communal engagement then that platform would be more suitable for stimulating loyalty towards the brand. As previously discussed, Facebook and YouTube are expected to stimulate the most amount of communal or emotional engagement. Therefore, the following hypothesis is proposed:

H4. If Facebook and YouTube have the highest amount of communal and emotional engagement, then both platforms will be better at stimulating brand loyalty in comparison to Instagram.

Furthermore, based on Lim and colleagues’ (2015) findings, the social media platform that has the highest amount of functional engagement will be more suitable for stimulating brand awareness. As previously mentioned, Instagram is expected to have the highest amount of functional engagement out of the three platforms. Therefore, the following hypothesis is proposed:

H5. If Instagram has the highest amount of functional engagement then the platform would be best for stimulating brand awareness in comparison to Facebook and YouTube.

(14)

13

Method

The research will employ a quantitative content analysis due to the social desirability effect that can occur using other methods (i.e. interview, experiment settings), which would influence engagement with the content. The study is aimed at determining the difference in consumer engagement with content that they would come across in a natural setting. This would mean that the individual who has engaged with the content did so because they wanted to. To be representative of the natural online user interaction, the study will rely on existing brand-initiated organic content. The engagement that will be used for the analysis took place on Formula 1’s social media content that is intended to engage their following in between races. F1 stimulates fan engagement by regularly posting on Facebook, Instagram, and YouTube. The organization regularly posts the same content across the three platforms, which means the same content is shown in a different social media platform context. The same content in a different context makes the engagement with the content suitable for comparison, as the context would be the main influential factor for potential differences in engagement.

Sampling

In total, the analysis included 6 different videos, which were all posted on the three different platforms. This led to a total sample size of 18 videos (6 videos x 3 platforms) which were each accompanied by the platform specific engagement. This sample size means that a total of 18 different comment sections were analysed. The first 10 comments that show up after unfolding the comment section will be used for the coding procedure. This will lead to a total of 180 comments coded (60 per platform). The comment sections were all filtered according to the platform’s ‘most relevant/top comment’ filters. This will made the top 10 comments suitable for comparison because each platform has determined the comments that would be most relevant to its’ user base. Furthermore, the ‘most relevant’ filter had a mix of most liked

(15)

14

and replied to comments, combined with most recent comments. This meant that the

comments were quite representative of the general consumer engagement that took place. The content was sampled using a cluster sampling method. The clusters of videos were formed based on the type of content. The content was posted in the period between the most recent F1 season (2019) and the upcoming season (2020) as this would be most representative of the current use and engagement that takes place on social media

platforms. Additionally, the period of 4 months (December 2019 – March 2020) was chosen because it would increase the likelihood of similar audience sizes.

The video content was of similar nature, because Saji and colleagues (2013) found that engagement with brand-initiated content is not just influenced by the context but also by the type of content (i.e. informative, entertainment, etcetera). According to Dolan, Conduit, Fahy, and Goodman (2016) informative content is used to provide users with resourceful and helpful information. The authors went on to define entertaining content as content intended to “fulfil the users’ need for escapism, hedonistic pleasure, aesthetic enjoyment, and emotional release” (Dolan et al., 2016, p. 263-264).

Voorveld and colleagues’ (2018) findings state that YouTube was used for seeking entertainment, Instagram for filling empty moments and Facebook for topicality and

identification. These findings imply that the content should be suitable for all three platforms by being a mixture of informative and entertaining content. Informative content should appeal to the Facebook users due to the topicality of the content, and entertaining content should appeal to audiences on Instagram and YouTube. Therefore, the video cluster that was selected for this research consisted of videos that ranked the highlight reel moments from past races. This content is intended to inform the consumer of the history of the sport through highly entertaining moments. The consumer engagement used for the analysis occurred on the following videos: 1) Top 10: on-track battles of 2019; 2) Top 10: Biggest F1

crashes of 2019; 3) Top 10: best overtakes of 2019; 4) Top 10 races of the decade

(16)

15

Operationalization of variables

The independent variable of the research was the different social media platforms Facebook, Instagram, and YouTube. Several dependent variables were used to determine the influence of the specific platforms on the engagement. The variables included the engagement types (communal, emotional, and functional) and levels (active and passive). There are multiple engagement behaviours that determine whether the engagement was a communal,

emotional, or functional type of engagement. Therefore, these engagement behaviours were incorporated into the codebook to determine what type of engagement was the most present on the platform.

According to Lim and colleagues (2015) functional engagement would be the liking of the content and tagging of a friend. Therefore, the total amount of likes was recorded, and the amount of tags in the top ten comments was counted. The authors state that sharing an emotional response to the content would be considered emotional engagement. For

Facebook, the number of emoticon reactions to the content were recorded. Additionally, comments including an emoticon or written emotional expression would be labelled as emotional engagement on all platforms. Lastly, Lim and colleagues (2015) state that the written expression of an opinion based on factual or logical reasoning, rather than emotional reasoning, is a form of communal engagement. This means comments that consisted of this type of written expression were labelled as communal engagement. Furthermore, a

comment intended to interact with other users was perceived to be a form of communal engagement, which means that jokes were also labelled as communal engagement. Additionally, replying on another user’s comment and/or liking it would also be a form of communal engagement. Therefore, the amount of likes and replies on a comment was also recorded.

The total amount of functional engagement per video on Facebook and Instagram was calculated by adding up the total number of likes and tags. For YouTube, the number of dislikes were also added up to these other two factors. The total amount of emotional

(17)

16

engagement per video on Facebook was calculated by adding up the number of emotional reactions and the number of comments that were labelled as emotional engagement. For the other two social media platforms only the number of comments labelled as emotional

engagement made up the total amount of emotional engagement per video. The total amount of communal engagement per video was calculated by adding up the comments labelled as communal, the number of likes on the comments, and the number of replies to the comments.

Following Muntinga and colleagues’ (2011) criteria for active and passive

engagement all three types of engagement would be considered an active form of

engagement. Therefore, the total amount of likes, comments and shares were added up to determine the number of active engagers with the content. Furthermore, Muntinga and colleagues (2011) state that the mere viewing of the content would be a form of passive engagement. Therefore, the total amount of active engagement was subtracted from the total amount of views, which was also recorded to determine the number of passive

engagers with the content. Both active and passive engagement are approximate amounts as one user could comment, like and share the content.

Procedure

Due to the online presence of the content for coding, screenshots were made of the content at the time of visitation. This improved the replicability of the study as the number of views, comments, and likes are not static. The comments were then labelled from 1 to 10 to make it clearer for the coders which comments to code. After the screenshots were taken and the comments labelled, a pdf file was made to make the coding process easier and more efficient as all relevant information is compiled into one document. Once the content was collected the codebook (See Appendix B) was used to collect the data.

(18)

17

Results

Intercoder reliability test

The research made use of two coders, including the researcher (coder 1), and the reliability-coder (reliability-coder 2). Coder 2 randomly selected 1 video on Facebook, 1 on Instagram and 2 on YouTube. This led to a total sample size of 4 videos, and 40 comments which amounts to a total of 22.2% of the material being coded by the second coder. The second coder was trained for half an hour and could find further instructions for each of the codes in the

codebook. The intercoder reliability test was only done for the variables that ended up being used in the analysis. The intercoder reliability test results are visible in Table 1.

Table 1

Intercoder reliability test results for the variables included in the analysis

Variable Name % Agreement Kappa

B.1.1. Total Comments 100% 1.00 B.1.2. Total Likes 100% 1.00 B.2.1. Total Views 100% 1.00 C.1.0. Functional Comments 100% 1.00 C.2.0. Emotional Comments 93% .848 C.3.0. Communal Comments 95% .724 C.3.1. Comment Replies 100% 1.00 C.3.2. Comment Likes 100% 1.00

Note. Intercoder reliability Kappa is sufficient 0.61 – 0.80, good 0.81-0.90 and excellent .91-1.00

The intercoder reliability test consists of two elements: the percent agreement and Cohen’s Kappa. The tests determined that for codes B.1.1., B.1.2., B.2.1., C.1.0., C.3.1. and C.3.2. there was 100% agreement between the two coders, which was confirmed by a Kappa score

(19)

18

of 1.00 for each, resulting in the conclusion that these codes were perfectly executed throughout the research. Therefore, the reliability of the results accompanied to these codes is very strong and accurately represent the statistics found in the content. However, codes C.2.0. and C.3.0. were the most ambiguous codes throughout the coding process, which is also reflected in the percent agreement of 93% and 95% and the Kappa scores (.848, .724) respectively. This translates into code C.2.0. being of good reliability Kappa (.848) and code C.3.0. being of sufficient reliability Kappa (.724) to be included in the analysis. These two codes are a bit lower in terms of reliability compared to the other variables. The most logical explanation for this would be that both variables included were expressed in written text. Written text is more susceptible to multiple interpretations compared to the numerical statistics that had to be noted down for the other variables.

Types of engagement per platform

To compare the means of the variables and determine the most dominant type of engagement across the platforms the research relied on the one-way ANOVA test. The results would indicate whether there is a difference in mean effects of engagement types on Facebook, Instagram, and YouTube. In case there is a significant difference in the amount of a specific type of engagement per platform then it would indicate that the platform is better or worse at stimulating that specific type of engagement. The one-way ANOVA test was

accompanied by Tukey’s post-hoc test to determine which of the groups varied significantly from one another. The outcomes of both are reported in Table 2, which will be referred to the for the remainder of this section. All results were verified using the Kruskal Wallace H test, because the assumptions of normal distribution and equal variances were violated for some variables.

(20)

19 Table 2

The amount of engagement with the video content on Facebook, Instagram and YouTube

Measure Facebook Instagram YouTube F (2, 15) p-value

M SD M SD M SD Total Views 1,275,167 887,804 956,344 167,836 1,817,652 1,515,128 1.097 .359 Total Comments 671a ₂₇₃ ₆₂₈a ₂₀₁ _1,908b ₉₁₀ _10.076* _.002 Total Likes 19,300a _8,351 _144,076b _25,017 _33,885a _13,561 _95.247** _.000 Within comments Functional .00a _.00 _1.68b _1.51 _.00a _.00 _7.353* _.006 Emotional 5.00 1.10 5.33 1.63 5.00 1.8 .094 .911 Communal 9.33b _1.21 _6.50a _2.26 _8.06ab _.82 _5.138* _.020 Total Engagement Passive 1,255,195 880,655 811,640 144,188 1,781,859 1,501,558 1.392 .279 Active 23,520a _9,902 _144,703b _25,119 _36,522a _14,750 _84.167* _.002 Functional 19,300a _8,351 _144,078b _25,017 _34,614a _14,090 _93.227* _.001 Emotional 3,554b _1,434 _5.33a _1.63 _5.00a _1.79 _36.740* _.003 Communal 327a ₃₅₇ ₈₆₈a _1,319 _10,590b _3,378 _45.207* _.003

Note. ANOVA results are indicated * p value < .05, ** p value < .001

Note. ab_{different superscripts are used to indicate the values where the mean difference was}

statistically significant according Tukey’s Post-Hoc test, p < .05

One of the most important observations from the results in Table 2 is that the total functional engagement had the highest mean on Instagram (M = 144,078, SD = 25,017), and this was a statistically significant difference from the means of functional engagement on Facebook Mdifference = 124,778, p < .001 and YouTube Mdifference = 109,463, p < .001.

Therefore, the research fails to reject H1, which states that if the content is posted on the brand’s Instagram channel then it will have the highest amount of functional engagement compared to the same content on Facebook and YouTube. Additionally, this finding also means that the research fails to reject H5 which states that if Instagram has the highest amount of functional engagement then the platform would be best for stimulating brand awareness in comparison to Facebook and YouTube.

Secondly, YouTube turned out to have the highest mean for the total amount of communal engagement with the content (M = 10,590, SD = 3,378), which was significantly

(21)

20

different from both Facebook Mdifference = 10,263, p < .001, and Instagram Mdifference = 9,721, p

< .001. This finding leads to the rejection of H2 that if content was posted on the brand’s YouTube channel then it will have the highest amount of emotional engagement compared to the same content on Facebook and Instagram.

Lastly, Facebook was found to have the highest mean for the total amount of

emotional engagement with the content (M = 3,554, SD = 1,434), which differed significantly from both YouTube Mdifference = 3,548, p < .001 and Instagram Mdifference = 3,549, p < .001.

Therefore, the research rejects H3 that if the content was posted on the brand’s Facebook channel then it will have the highest amount of communal engagement compared to the same content on Instagram and YouTube. Despite the rejection of H2 and H3, the research fails to reject H4 that if Facebook and YouTube have the highest amount of communal or emotional engagement, then both platforms will be better at stimulating brand loyalty in comparison to Instagram.

Engagement behaviours per platform

Interesting findings that are also worth mentioning would be the fact that table 2 shows that the means of views per platform was statistically insignificant F(2, 15) = 1.10, p = .359). Therefore, the amount of impressions the content got was not dependent on the platform. Additionally, the same could be said for the mean amount of passive engagement with the content on each platform with F(2, 15) = 1.39, p = .279. Both findings would suggest that the platforms do not influence the mean number of passive engagers with the video content. However, Instagram had the highest mean of active engagers. A mean difference that was statistically significant in comparison to both Facebook (Mdifference = 121,184, p < .001) and

YouTube (Mdifference = 108,182, p < .001). Therefore, the findings are slightly contradictive of

one another.

Secondly, the results showed that the platforms had a significant effect on the total amount of comments F(2, 15) = 10.08, p = .002, and the total amount of likes F(2, 15) =

(22)

21

95.25, p = .000 that the content got. Tukey’s post-hoc test results determined that YouTube had the highest mean amount of comments the content got in comparison to the same content posted on Facebook, Mdifference = 1,237, p = .004 and on Instagram, Mdifference = 1,280,

p = .003. Furthermore, the Table 2 results show that Instagram had the highest mean amount of likes in comparison to Facebook, Mdifference = 124,776, p < .001 and YouTube

Mdifference = 110,191, p < .001. Therefore, the results indicate that the content on Instagram

would more likely trigger low intensity active engagement such as liking in comparison to the other platforms. Whereas on YouTube the engagement with the content would more likely lead to high intensity active engagement such as commenting compared to the other two platforms.

Lastly, the Table 2 results show that there was a statistically significant difference between platforms in the presence of functional engagement and communal engagement within the comment sections. Tukey’s post-hoc results showed that Instagram had a statistically significant effect on the presence of functional engagement within the comment section in comparison to Facebook, Mdifference = 1.67, p = .012 and YouTube Mdifference = 1.67,

p = .012. For communal engagement presence on the other hand, Facebook had a

significant effect on the presence of it in comparison to Instagram, Mdifference = 2.83, p = .017.

This suggests that the content of the comments on Facebook is more likely to be of communal nature than on Instagram. On the other hand, the content of the comments on Facebook and YouTube have the same chance of being of communal nature. Furthermore, no significant difference was observed between the comments across all platforms in terms of emotional engagement presence F (2, 15) = .09, p = .911, therefore meaning that the platform type did not exert an influence on emotional engagement within the comment section.

(23)

22

In the field of social media advertising previous explorative studies have pointed out the difference in engagement with the content that occurs per platform. Instagram had the highest average amount of functional engagement per video, which was in line with the prediction made based on previous research. According to Voorveld and colleagues (2018) Instagram is mainly used by users looking to fill empty moments. Therefore, the findings of this research would suggest that users seeking content as a pastime activity would be more hesitant to participate in high-intensity level active engagement.

Content on Facebook had the highest average amount of emotional

engagement. This finding rejected the hypothesis that Facebook would have the highest average amount of communal engagement. This could be explained by the fact that the platform interaction design (the emoticon reactions) has made it easier for the user to express an emotional response to the content. Furthermore, content on YouTube had the highest average amount of communal engagement. This was mainly due to the high amount of likes that users gave to the comments of others. Therefore, one explanation could be that the content of comments more often expressed shared opinions than on Facebook or Instagram. Another explanation for this would be that jokes were labelled as communal engagement. Voorveld and colleagues (2018) found that YouTube users seek entertainment, which was defined as the “need for escapism, hedonistic pleasure, aesthetic enjoyment, and emotional release” (Dolan et al., 2016, p. 263-264). Therefore, if the commenter expressed a joke that led to the fulfilment of this desire in another user, then this comment would more likely trigger engagement with it.

Lastly, even though Facebook and YouTube did not have the highest mean in the predicted type of engagement, each platform still stimulated one type of

engagement more frequently than the other two. Therefore, these results are in line with Voorveld and colleagues’ (2018) conclusion that engagement differed significantly per platform.

Another important finding of this research would be that different social media platforms seem to be more suitable for achieving a specific marketing objective. The

(24)

23

high average amount of emotional and communal engagement on the content posted on Facebook and YouTube respectively, would indicate that these platforms are better at stimulating brand loyalty (Lim et al., 2015). Furthermore, content posted on Instagram had a high average amount of functional engagement, with users predominantly leaving a like on the content. These findings are indicative that Instagram would be a platform more suitable for stimulating awareness (Lim et al., 2015). This is an interesting insight, as paid advertising content on both Facebook and YouTube is more often evaluated negatively than on Instagram (Voorveld et al., 2018). This could be because Facebook and YouTube should be used for engaging the existing customer base with content intended to build customer loyalty instead of paid advertising content to stimulate brand awareness. Furthermore, as Instagram seems to suit content intended to stimulate brand awareness better than Facebook and YouTube, it would be a possible reason for the positive paid advertising evaluations found by Voorveld and colleagues (2018).

Conclusion

Returning to the research question at hand, according to the findings of this research each social media platform has a specific type of engagement that is more prevalent than on the others. Therefore, the engagement with the content posted on a brand-owned social media channel is influenced by the specific social media platforms to a great extent. Furthermore, this platform specific engagement leads to the achievement of a different marketing objective. Facebook and YouTube are better suited for

stimulating consumer loyalty, whereas Instagram is better suited for stimulating brand awareness.

Theoretical and practical implications

The theoretical contributions of these findings are mainly aimed at extending and building upon Voorveld and colleagues (2018) findings of the difference in engagement with content

(25)

24

per platform. Most importantly, the results indicate that their findings do apply to content posted on brand-owned social media channels. Second, the research was able to identify the type of engagement most common to each platform. This is important because: 1) it provides insights into why Voorveld and colleagues (2018) might have found a difference in advertisement evaluations; and 2) this helps specify what type of engagement a platform stimulates best. This is an appropriate bridge to the main practical implication of the findings.

The practical implications of the findings are mainly for marketing departments and advertising agencies that are currently in a continuous effort to optimize social media marketing. Now that it has been confirmed that each platform stimulates a specific type of engagement, these departments and agencies can use the insights in the content creation process. The difference in engagement with the same content could lead to the marketing objective not being achieved on one or more social media platforms. This raises the question of whether it is worth posting the content when it does not have the intended effect.

According to the findings of this research content should be created for the platform or adapted to suit the platform better. Lastly, the research provides possible answers to why paid advertising gets evaluated negatively on Facebook and YouTube, which should lead these same departments to question the role of the platforms in their marketing strategies.

Limitations and future research suggestions

The research findings were based upon a relatively small sample of videos that were posted across three platforms. To be able to make more significant claims regarding the platform specific engagement it would require a larger sample, which would also lead to an improved generalizability of the claims. Furthermore, the sample only included video content, which means that the current findings have limited applicability to other forms of advertising content. Additionally, the video content was limited to a mixture of informative and

entertaining content. Lastly, the research was limited to content that was only sampled from three platforms (Facebook, Instagram, and YouTube) all belonging to one brand (F1).

(26)

25

Therefore, future research should replicate this study with a larger and more varied sample size, both in terms of types and formats of content, across more platforms. This should lead to more generalizable claims towards platform specific engagement across the different types of platforms and content combined. Or it should focus down to one specific platform, to be able to include multiple brands.

Furthermore, the content of the comments and the content itself was not thoroughly analysed due to the quantitative nature of the study. Despite the desire to keep the content as similar as possible, the format of the video could have proven to not be enough to avoid the content of the video having an influence on the type of engagement. Saji and colleagues (2013) stated that content has an influence on the engagement with the content, and

therefore this should be further explored through a qualitative content analysis of both the video and the engagement that comes with it. Throughout this research jokes were included as communal engagement as they are intended to spark interaction with other users.

However, there is a difference between the interaction sparked by jokes and opinionated statements meant to trigger discussion. Therefore, a qualitative content analysis could also further explore the potential differences in forms of communal engagement that is present on for example Facebook and YouTube. If content triggers communal engagement that is mainly made up of jokes, then one might risk the content not being taken seriously.

Therefore, if the platform has an influence on the form of communal engagement then this should be taken into consideration as well when choosing the platform.

In general, the findings of this research answer some questions related to platform specific consumer engagement, but even more questions emerge from them. Therefore, my main suggestion for future research would be to further explore consumer engagement per social media platform. Due to the ever-changing role of social media in the life of the consumer researchers are tasked with trying to keep up and help brands gain a deeper understanding of each platform. Ultimately, this should lead to more effective marketing on social media, which seems to be increasingly relevant due to the diminishing role of

(27)

26

Reference List:

Ashley, C., & Tuten, T. (2015). Creative strategies in social media marketing: An exploratory study of branded social content and consumer engagement.

Psychology & Marketing, 32(1), 15-27.

Brodie, R. J., Hollebeek, L. D., Jurić, B., & Ilić, A. (2011). Customer engagement:

Conceptual domain, fundamental propositions, and implications for research. Journal

of service research, 14(3), 252-271.

Dessart, L., Veloutsou, C., & Morgan-Thomas, A. (2015). Consumer engagement in online brand communities: a social media perspective. Journal of Product & Brand

Management.

Dolan, R., Conduit, J., Fahy, J., & Goodman, S. (2016). Social media engagement behaviour: A uses and gratifications perspective. Journal of Strategic Marketing,

24(3-4), 261-277.

Erdoğmuş, İ. E., & Cicek, M. (2012). The impact of social media marketing on brand loyalty. Procedia-Social and Behavioral Sciences, 58, 1353-1360.

Facebook. (2020). Dictionary.com. Retrieved from:

https://www.dictionary.com/browse/facebook?s=t

Felix, R., Rauschnabel, P. A., & Hinsch, C. (2017). Elements of strategic social media marketing: A holistic framework. Journal of Business Research, 70, 118-126.

an Noort, G., & Willemsen, L. M. (2012). Online damage control: The effects of proactive versus reactive webcare interventions in consumer-generated and brand-generated platforms. Journal of interactive marketing, 26(3), 131-140.

Hutton, G., & Fosdick, M. (2011). The globalization of social media: Consumer relationships with brands evolve in the digital space. Journal of advertising

(28)

27

Instagram. (2020). Dictionary.com. Retrieved from:

https://www.dictionary.com/browse/instagram?s=t

Jung, J., Shim, S. W., Jin, H. S., & Khang, H. (2016). Factors affecting attitudes and behavioural intention towards social networking advertising: a case of Facebook users in South Korea. International journal of Advertising, 35(2), 248-265. Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and

opportunities of Social Media. Business horizons, 53(1), 59-68.

Lim, J. S., Hwang, Y., Kim, S., & Biocca, F. A. (2015). How social media engagement leads to sports channel loyalty: Mediating roles of social presence and channel commitment. Computers in Human Behavior, 46, 158-167.

Macdonald, E. K., & Sharp, B. M. (2000). Brand awareness effects on consumer decision making for a common, repeat purchase product: A replication. Journal of business

research, 48(1), 5-15.

Mihart, C. (2012). Impact of integrated marketing communication on consumer behaviour: Effects on consumer decision-making process. International

Journal of Marketing Studies, 4(2), 121.

Muntinga, D. G., Moorman, M., & Smit, E. G. (2011). Introducing COBRAs: Exploring motivations for brand-related social media use. International Journal of advertising, 30(1), 13-46.

Saji, K. B., Chauhan, K., & Pillai, A. (2013). Role of content strategy in social media brand communities: a case of higher education institutes in India. Journal of

Product & Brand Management.

Schivinski, B., & Dabrowski, D. (2016). The effect of social media communication on consumer perceptions of brands. Journal of Marketing Communications, 22(2), 189-214.

Schultz, D. E., & Peltier, J. J. (2013). Social media's slippery slope: challenges, opportunities and future research directions. Journal of research in interactive

(29)

28

Voorveld, H. A., van Noort, G., Muntinga, D. G., & Bronner, F. (2018). Engagement with social media and social media advertising: The differentiating role of platform type. Journal of advertising, 47(1), 38-54.

YouTube. (2020). Dictionary.com. Retrieved from:

(30)

29

(31)

(32)

(33)

32

Appendix B: Codebook

Codebook

For the gathering of the data the research will make use of a codebook. The codebook will consist of different sections that will have differing purposes. Section A will be dedicated to gathering data that is meant to ensure the same video is coded across all three different platforms. The section starts off with determining which coder (coder 1 or coder 2) is

responsible for the coding of the content (code A.1.0.). After which the date the content was coded on (code A.2.0.) is noted down, followed by the number (1 to 6) of the video from the video list (code A.3.0.). This is the first step in verifying the 6 different videos are all coded. The section moves on to determining the platform (Instagram, YouTube or Facebook) the engagement that will be coded in the book is on (code A.4.0.). The second step in verifying process of the video content is the title of the video that will be written down completely by the coder (code A.5.0.). Then the type of content (code A.6.0.) will be noted down in order to ensure the coders perceive the content to be of similar type. After which the date of

publication (code A.7.0.) and the length of the video (code A.8.0) are noted down to complete the verification process of the video.

Section B is aimed at collecting statistics such as the number of views, likes, and comments. This section will be divided into three topics, active engagement (code B.1.0.), passive engagement (code B.2.0.) and Facebook emotional engagement (code B.3.0.). The total amount of comments and likes will give an estimate of the amount of active

engagement with the content. Therefore, both likes and comments will fall under the section of active engagement and make up codes B.1.1. and B.1.2. respectively. According to Muntinga and colleagues (2011) the mere viewing of the content is a passive level of engagement with the content. Therefore, the total amount of views of the content minus the amount of active engagement (comments, likes) would provide an estimation of the amount of passive engagement with the content on each platform. Total views will therefore fall under the theme of passive engagement and will be labelled as code B.2.1. Lastly,

(34)

33

Facebook has the unique engagement feature that emoticons can be given to the content in the same manner as someone can also like the content. Therefore, the last topic will be Facebook emotional engagement where the numbers for each of the six emoticons is gathered under the code B.3.1.

Lastly, section C of the codebook will be focused on determining what type of engagement is most present within the comment section of each platform. This section will include the three types of engagement (functional, emotional, communal) and the indicators for each will be included in the instructions of each code. Functional engagement (code C.1.0.) in the comment section is defined by the tagging of friends or connections. Emotional engagement (code C.2.0.) is the expression of an emotion that was triggered by the content. Therefore, the comments that include an emotional reaction (happiness, excitement, anger, sadness, etc.) towards the brand released content will be coded as emotional engagement. Finally, communal engagement (code C.3.0.) is defined as the sharing of an opinion on the content based on rational reasoning, agreeing with someone else by liking it, or responding to someone else’s comment (. Therefore, this part of the codebook will examine the content for rational reasoning, while also noting down the number of likes (code C.3.1.) and replies left on the comment (code C.3.2.).

Part A: General information

A.1. Who is filling in this codebook? Highlight the coder filling in this codebook

Coder 1: Rik Polderman

Coder 2: Patricija Milaseviciute

A.2. What day was the post coded on?

A.3. What is the number of the video on the list of videos for this research? Refer to

the list of videos that are used for this research. 1. One

2. Two

2020 05

(35)

34 3. Three

4. Four 5. Five 6. Six

A.4. What platform is the video content on? Highlight the platform the video is on, so it

becomes clear what platform the coded engagement was on. 1) Instragram

2) YouTube 3) Facebook

A.5. What is the title/description of the video? (on YouTube the title is underneath the

video, on Facebook the title can be found in the video section of the F1 page where it will be next to the video, on Instagram there is only the description of the video underneath the content which discusses the content one can expect to find in the video)

A.6. What type of entertainment content is it? Informational content is meant to inform the

viewer on the events/current affairs relevant at the time of publication (e.g. F1 race

highlights, crashes, etc.) and educational content is meant to teach the audience about the sport to understand it better (e.g. F1 sport rules, the history of the sport, etc.). Highlight one of the following:

I. Informational II. Educational

A.7. What date was the content posted on? (the date of publishing should be visible

underneath the video on YouTube, above the video on Facebook, and underneath the video on Instagram)

(36)

35

A.8. What is the length of the video content? (This should be visible at the top right

corner of the content on Instagram, whereas on Facebook it is visible in the lower right corner and YouTube it is visible in the lower left corner)

minutes seconds

Part B: Engagement level

B.1.0. Active engagement: active engagement is defined by the amount of comments, likes

and shares of the content in question

B.1.1. Note down the number of comments made on the content: The number of

comments made is stated underneath the content on YouTube, on the Instagram video overview page by hovering your mouse over the video, and next to the Facebook video.

I. comments

B.1.2. Note down the number of likes the content received: Likes can be found

underneath the video on all platforms, on Facebook only count the likes not the

emoticons expressing stronger emotional reactions. This can be done by clicking on the likes it got, where you will then see a breakdown of the amount of likes and the other reactions you can leave on the video.

II. likes

B.2.0. Passive engagement: Passive engagement is defined by the number of individuals

who merely consumer/viewed the content, therefore the total amount of views needs to be collected

B.2.1. Note down the number of views the content has received: The total number of

views can be uncovered on Instagram by clicking on the number of likes underneath the content, which will take you to an overview of the total amount of likes and views. The views can be found underneath the YouTube video, views can be found in the video section of the company Facebook page underneath the title.

(37)

36 I. Views

B3.0 Emotional reactions to the content: Only applicable to facebook

B.3.1. Note down the amount of emotional reactions given to the content in the table below: The breakdown of the numbers can be found by clicking on the number of

likes/emotional reactions

Emotional reaction Number of reactions Love Care Haha Wow Sad Angry

Part C: Engagement type

Comment 1

C.1.0 Functional engagement: Did the comment contain a tag of another individual

(the tagging of another user is functional engagement as it is a form of sharing the content, a form of engagement that makes the content function for marketing purposes)

Yes/No Number of tags

C.2.0. Emotional engagement: Did the comment express an opinion on the content based on emotion? (a description of involuntary internal psychological reactions to the

content e.g. excitement, joy, frustration, disappointment, etc. The emotional reaction can be verbally communicated or through emoticons, and is an opinion based on emotional

(38)

37 Yes No

C.3.0. Communal engagement: Did the comment express an opinion based on logical reasoning, or was it meant to initiate social interaction through a joke? (formulated

belief about the topic of the content that is not based on emotional reasoning but is rather guided by logical reasoning (e.g. I like Vettel because he is a race winner). A joke that is meant to initiate interaction with other users is a form of co-creation with the content (e.g. ‘Crashstappen doing what he does best’, a joke about the driver Verstappen, controversial statement made to invite likeminded people to like the comment and for people of opposing view to come to Verstappen’s defence))

Yes No

C.3.1. Communal engagement: Note the amount of likes on the comment (the likes that

are given to comments are a sign of agreement with the person’s opinion, furthermore it is an interaction with the user rather than with the brand)

Number of likes:

C.3.2. Communal engagement: Note down the number of replies on the comment. The

replies made to another user’s comments would be indicative of social interaction and the expression of a similar or differing opinion. The replies made on comments (for both an emotional or communal oriented comment) are used to interact with the other users to start a discussion, or express a shared opinion rather than to interact with the brand.

Number of replies:

Comment 2

C.1.0 Functional engagement: Did the comment contain a tag of another individual

(the tagging of another user is functional engagement as it is a form of sharing the content, a form of engagement that makes the content function for marketing purposes)

(39)

38

Yes/No Number of tags

C.2.0. Emotional engagement: Did the comment express an opinion on the content based on emotion? (a description of involuntary internal psychological reactions to the

content e.g. excitement, joy, frustration, disappointment, etc. The emotional reaction can be verbally communicated or through emoticons, and is an opinion based on emotional

reasoning (e.g. I love the sport because I find it exciting) Yes No

C.3.0. Communal engagement: Did the comment express an opinion based on logical reasoning, or was it meant to initiate social interaction through a joke? (formulated

belief about the topic of the content that is not based on emotional reasoning but is rather guided by logical reasoning (e.g. I like Vettel because he is a race winner). A joke that is meant to initiate interaction with other users is a form of co-creation with the content (e.g. ‘Crashstappen doing what he does best’, a joke about the driver Verstappen, controversial statement made to invite likeminded people to like the comment and for people of opposing view to come to Verstappen’s defence))

Yes No

C.3.1. Communal engagement: Note the amount of likes on the comment (the likes that

are given to comments are a sign of agreement with the person’s opinion, furthermore it is an interaction with the user rather than with the brand)

Number of likes:

C.3.2. Communal engagement: Note down the number of replies on the comment. The

(40)

39

expression of a similar or differing opinion. The replies made on comments (for both an emotional or communal oriented comment) are used to interact with the other users to start a discussion, or express a shared opinion rather than to interact with the brand.

Number of replies:

Comment 3

Please follow the same criteria as outlined for comments 1 & 2 for the remainder of the comments

Functional shares: Yes/No Number of tags:

Emotional engagement: Yes No

Communal engagement Yes No

Amount of likes on comment Amount of replies on comment

Comment 4

Comment 5

(41)

40

Comment 6

Comment 7

Comment 8

(42)

41

Comment 9

Comment 10

Please follow the same criteria as outlined for comments 1 & 2 for the remainder of the comments:

Social media marketing: The influence of specific social media platforms on consumer engagement with the content posted on brand-owned social media channels