Netflix’s community relationship management. An analysis of its Facebook-USA page

ABSTRACT

This paper explores how Netflix employs brand community relationship management on its United States Facebook page. Methodologically, this study uses qualitative thematic analysis to dissect 100 posts published by Netflix’s community manager on Facebook and 826 comments posted by the company and internet-users. Results suggest that Netflix keeps an active and regular posting schedule with new content and portrays the brand as a humorous and a relatable peer. However, Netflix does not encourage Facebook users to engage in the content on the platform. The brand also seems to ignore complaints received from Facebook users as well as identified internaut-to-internaut conflicts. Findings are relevant to researchers in the fields of digital media and marketing and to practitioners working in those two fields.

KEYWORDS:

• Netflix
• Facebook
• brand communities
• community relationship management
• United States

Introduction

“Netflix” is no longer just the name of a company. It is a verb used to define a beloved leisure activity, a concept used to describe an age (Barker & Wiatrowski, 2017). Netflix has also been a source of fascination due to the way the company uses social media. In fact, its innovative and engaging social media conduct has received the attention of both academics (see Fernández-Gómez & Martín-Quevedo, 2018a) and practitioners. For example, Duczeminski (2019) emphasised the unique tone of voice that Netflix uses in speaking to its audiences, Letki (2016) highlighted the company’s efforts to give “attention and care” to all of its social media followers, and Beer (2019) stressed the “ongoing” and “creative dialogue with the audience”. Analyses of how companies such as Netflix manage their marketing strategy add value not only by advancing the field of media management research but also, as Rahe et al. (2020) suggested, by inspiring media managers who are building or leading media brands.

Social media platforms have proved to be useful for brands to build relationships with their consumers and users, increase brand loyalty and brand trust (Laroche et al., 2012), and foster and encourage the dispersion of positive electronic word of mouth (Martínez-López et al., 2016). Creating effective strategies to manage social media-based brand communities has become vital (Cottica et al., 2017). This is what Ang (2011, p. 32) calls community relationship management (CoRM), a key term in the present investigation. According to Kuo and Hou (2017), the fast growth of online brand communities in recent years should be accompanied by even greater academic attention for online brand communities.

Despite the scholarly contributions made by, for example Fernández-Gómez and Martín-Quevedo (2018a, 2018b), about how Netflix uses Twitter and Instagram (Martín-Quevedo et al., 2019), there is a paucity of academic papers that explore how Netflix uses Facebook, its largest social media platform (Chamat, 2019), in the country that contributes the most to its number of subscribers—that is, the United States of America (Barker & Wiatrowski, 2017; Ramsey, 2018). Consequently, and given the role that marketing strategies play in media business, this field remains incomplete without answering the question of how Netflix employs brand community relationship management on its United States Facebook page.

To answer the above-mentioned research question, we draw on the CoRM framework developed by Ang (2011) and on the results obtained through a qualitative thematic analysis of 100 posts posted by Netflix’s community manager on Facebook and 826 comments published by Netflix and Facebook users. The findings of this exploratory study contribute to the growing body of literature on social-media management in the subscription video-on-demand (SVoD) market. Furthermore, this investigation also provides a detailed third-party review of the workings of its community management, an approach that might uncover aspects of its strategy that may be hard to spot, even by the company itself.

Literature review

Netflix’s content and, more precisely, its in-house production are of utmost importance when talking about the ‘Netflix band’. Initially, Netflix strategically used the term ‘quality’ to legitimise so-called binge-watching. Later, this mode of viewing was replaced, and “quality TV” became linked to the notion of “diversity” applied to, for example, the characters that inhabit the platform’s content (Jenner, 2018). Given the different types of subscribers for which Netflix creates content (Lotz, 2017) and the high number of streaming titles available, having a powerful recommendation system that leads to the personalisation of the ‘Netflix experience’ became key for Netflix’s success (Amatriain, 2013).

Print advertisement, live events combined with interactive storytelling (e.g. the “Crazy Pyes Food Truck” of Orange is the New Black; see DeCarvalho & Fox, 2016), and (controversial) billboards are other tools that Netflix has incorporated in its marketing strategy, a complex machinery designed to distinguish itself from its global and local competitors (Fernández-Gómez & Martín-Quevedo, 2018a). In this intricate context, social media platforms have become one of the pillars of Netflix’s branding strategy, as has also happened with other SVoD services (DeCarvalho & Fox, 2016) and traditional broadcasters in countries such as Spain (Lacalle & Castro, 2016) and Germany (Stollfuß, 2019).

The motivations to use social media as a marketing tool are multiple. For example, social media allow companies to be in direct contact with their consumers more efficiently and at a relatively low cost (Jiao et al., 2018; Kaplan & Haenlein, 2010). Furthermore, social media facilitate consumers’ engagement with the brand (Fernández-Gómez & Martín-Quevedo, 2018a) as well as the creation of online communities, understood as social aggregations that unite like-minded individuals who share certain goals and interests (J. Brown et al., 2007; Rheingold, 1993).

The popularity and related benefits of online communities have captured the attention of marketing professionals, and many companies have invested in developing online communities around their brands for years (McWilliam, 2000). Brand communities can provide companies with a clear(er) understanding of their consumers’ experiences and needs and allow them to facilitate a place where consumers can diffuse brand-related electronic word of mouth (eWOM) (Martínez-López et al., 2016). Brand communities are also considered a great tool for positively influencing the relationship between brand and consumers (Muniz & O’Guinn, 2001). They are also effective in building brand loyalty since those members who feel emotionally attached to a brand are likely to remain loyal to the community as well as to the brand (Martínez-López et al., 2016). Despite the abovementioned benefits, social media platforms may be counterproductive for some brands, as the next section illustrates.

Community relationship management

Rumours about a brand spread easily through social media-based brand communities (SMBBCs). Moreover, the open-access nature of SMBBCs allow internet users to sabotage the brand from within (Kähr et al., 2016; Muniz & O’Guinn, 2001) and the heterogeneity of the community increases the risk of internal conflicts (Dineva et al., 2017, p. 4). An active management of their SMBBCs has, thus, become crucial for brands (Cottica et al., 2017; Noble et al., 2012). Ang (2011, p. 32) calls this community relationship management (CoRM), which is built on four pillars: connectivity, conversations, content creation, and collaboration.¹

Connectivity and conversation

Connectivity and conversation are important in building a strong brand community. These two pillars comprise brand-to-consumer and consumer-to-consumer interactions, two types of communication that are essential for a brand’s success and reputation since they involve acting (or not) on questions, complains and general comments raised by consumers (Willemsen et al., 2012). On this front, there are two general approaches. Proactive webcare happens when brands publish responses that are not preceded by requests from consumers (Van Noort et al., 2015). This type of webcare can be considered as a tool for relationship building (Tsimonis & Dimitriadis, 2014). Contrarily, reactive webcare is concerned with replying to requests from consumers (Van Noort et al., 2015) and seeks to enhance their satisfaction with the brand. This includes, for example, posts published by brands to explain legitimate complaints posted by consumers but also posts where brands debunk malicious accusations (Noble et al., 2012).

As Husemann et al. (2015) noted, avoiding consumer-to-consumer conflicts (e.g., consumers who have conflicting experiences with a brand) is particularly important in large communities that consist of people with loose social ties and different levels of shared commitment to a brand. Concerning consumer-to-consumer conflict management and moderating negative comments, Dineva et al. (2017) and Noble et al. (2012) distinguished five strategies, which the following Table 1 summarises:

The non-engaging strategy is the most often used conflict management strategy, despite studies emphasising that brands should actively intervene in conflicts (Dineva et al., 2017). Furthermore, and contrary to marketing specialists’ recommendations, brands tend to also ignore or even censor negative comments (Dekay, 2012, p. 289). All in all, establishing interactions with consumers as well as moderating interactions between them is essential for a brand’s success and reputation (Schamari & Schaefers, 2015).

Content creation

The second pillar is content creation, which relates to the idea of brands fostering their relationship with consumer by encouraging them to publish user-generated content on the community page or endorsing them to respond to the content published by others (Ang, 2011; Kumar & Reinartz, 2018). A brand asking for help in creating content also establishes interdependence between brand and consumer, which in turn also positively affects their relationship (Fournier, 1998). Together with encouraging user-generated content, brands should also create their own content to add value to the community (Tsimonis & Dimitriadis, 2014). This is especially true for Netflix managing its US Facebook page, as American social media users are generally attracted to marketer-generated content (Jiao et al., 2018). With their content, brands can offer entertainment benefits (Jiao et al., 2018) by posting content that is relevant for the community and light-hearted in nature and that makes consumers smile (Gummerus et al., 2012). Moreover, consumers also look to brands to provide them with informational content that offers them additional knowledge on the brands and their services (Jiao et al., 2018). Ideally, brands should post content on a daily basis (Gummerus et al., 2012), as witnessing inactivity can turn off (potential) consumers (Waters et al., 2009).

Also important along with the creation of original content is the tone of voice. Previous studies have found that companies must humanize their brands (S. Brown et al., 2003; Kuo & Hou, 2017; Martínez-López et al., 2016) since this allows consumers to establish a self–brand connection. Self–brand connections are established when brands embody traits that are seen as favourable by consumers (Kemp et al., 2012). Brands can humanise their brands in several ways. For example, they can assign the brand human-like qualities to enliven the brand as an actual thinking and feeling entity with a human-like mind (Fournier, 1998; MacInnis & Folkes, 2017).

Some organisations even go so far as personifying and portraying a brand as a person (Aaker, 1997; Veloutsou & McAlonan, 2012). When assigning personality traits to their brands, brands should take into consideration their heritage, such as the story of their origin, and their authenticity, their uniqueness, as consumers often value these aspects (Martínez-López et al., 2016). According to Geuens et al. (2009), assigning personality to brands can be done in four different ways, which we will explore in this study: as (1) a symbol, (2) a product, (3) an organisation, and (4) a person. The brand as a symbol is the most dehumanising and the brand as a person is the most humanising. Brand dehumanisation occurs when brands are perceived as objects; this leads to others denying a brand’s humanity (Fiske, 2013; MacInnis & Folkes, 2017). Brand dehumanisation usually occurs when brand values are not in line with consumer values; consumers are then likely to reject the brand’s humanising efforts (Puzakova et al., 2009).

Hitherto, Netflix has been praised by marketing and branding professionals for being relatable and speaking the language of its audience (Duczeminski, 2019) through the usage of a unique tone of voice, as well as consistently offering entertaining content, listening to its subscribers, and giving ‘attention and care’ to all social media users who post content on its pages (Beer, 2019; Stevens, 2018).

Based on the literature discussed, and to further understand the business strategy of a key media player, this paper aims to answer the following research questions:

RQ: How is Netflix employing brand community relationship management (CoRM) on its United States Facebook page?

SQ1: How is Netflix employing ‘connectivity’ and ‘conversation’ in its posts and comments?

SQ2: How is Netflix employing ‘content creation’ in its posts and comments?

SQ3: How are Facebook users using Netflix’s United States Facebook page to communicate with Netflix?

This paper offers one of the first overviews of Netflix brand community relationship management (CoRM) on Facebook. Specifically, the study is focused on a qualitative analysis of those posts and comments published by both Netflix and Facebook users on the Netflix US Facebook page. The next section explains the methodology used in this exploratory qualitative study, which is followed by a presentation of the results obtained. Finally, the main conclusions and the limitations of this investigation are presented and discussed.

Method

To address the stated research questions, this study conducted a qualitative thematic analysis on a sample of a total of 926 texts posted from 20/03/2019 to 03/05/2019 and collected from the Netflix US Facebook page on 5 May 2019. A netnographic approach was used to assess both Netflix’s and Facebook users’ interactions on this social network and develop a holistic view of Netflix’s CoRM approach.

Data collection

To prevent any cookies, algorithms, or other systems from possibly influencing the sample, the first author cleared all history from the browser before starting the data collection process and did not log in to any account on Facebook, therefore keeping an unidentified profile. The 926 units of analysis collected in this exploratory qualitative study were divided into posts (n = 100) and comments (n = 826). We understand posts as those blocks of content published by Netflix on its Facebook page to either inform or engage the community of Facebook-users. In this case, both textual and visual elements (e.g. images and videos) were analysed (McKenna et al., 2017). Comments are those messages that were published underneath Netflix’s posts and which could be published by either Netflix or Facebook users. Our interest in exploring the interactions between Netflix and Facebook users justifies the decision to analyse 8 times more comments than posts. The usage of these two sources of data (i.e., posts and comments) has allowed us to depict a more complete overview (Stemler, 2001) of Netflix’s community relationship management. It is also worth noting that copying and pasting the units of analysis into .doc files (primary documents) tackled the short-lived nature of the online content through the creation of static copies.

Concerning the type of sampling applied, non-probability purposive sampling (Babbie, 2017) was used. The collection was based on the following requirements relevant for this study (Table 2):

Interestingly, the dataset showed a remarkable lack of comments from Netflix’s side. From all the comments collected initially, a mere 3% (n = 24) were written by Netflix. To further gain insight into Netflix’s webcare strategy, 26 more comments published by the company were gathered, bringing the total number of Netflix comments in the final dataset to 50. These extra 26 comments were gathered from within the most recent posts that were already a part of the dataset. To be included in the sample, these comments had to:

• Be posted by Netflix;

• Be posted underneath one of the sampled 100 posts;

• Not be already part of the initial dataset;

• Include at least one word of text;

• Be written in English.

The sample proved to be broad enough to allow for an in-depth analysis and the reaching of data saturation.

Data analysis

The 926 units of analysis were analysed by the first author of the paper by means of a qualitative thematic analysis. The researcher strictly followed the quality criteria for coding suggested by Braun and Clarke (2006), such as “each data item has been given equal attention in the coding process” (p. 96).

Qualitative thematic analysis allows for the systematisation of the analysis and interpretation processes and follows a three-phase iterative coding process (Boeije, 2010)—namely, open, axial, and selective coding—that builds on Strauss and Corbin’s Grounded Theory Method. Before starting, the researcher read the complete dataset multiple times to become familiar with the units of analysis. After that, and with the research questions in mind, the researcher broke down all the data. The initial codes consisted of both descriptive and sometimes in-vivo codes (see Table 3, example 1) and more interpretive codes, which already started to be connected to the literature (see Table 3, example 2).

Table 1. Strategies for consumer-to-consumer conflict management.

Strategy	Description
Non-engaging strategy	Brands choose to not intervene in conflicts and avoid resolving the conflict
Censoring	Brands remove certain content that includes bad language or attacks on other consumers
Pacifying	Brands post comment to ask consumers to adjust their behaviour
Bolstering	Brands post a comment that affirms a consumer who is defending the brand
Informing	Brands post information to rectify a comment that has led to a consumer-to-consumer conflict
Promotion	Brands promote their services by offering an explanation for the experienced shortcoming in combination with a subtle promotion

Table 2. Units of analysis and requirements.

Type of unit of analysis	Requirements
Post	Be posted on the Netflix US Facebook page; Be posted by Netflix itself; Be one of the 100 most recent published posts as of 5 May 2019 and Be written in English.
Comment	Be posted underneath one of the sampled 100 posts; Be one of the top 8 ‘most relevant’^2 comments underneath those posts; Include at least one word of text; Be written by either Netflix or Facebook users and Be written in English.

Table 3. Example of open codes.

Example of codes	Illustrative quote
● Complaint about show cancellations	“You’ll just cancel it after a few seasons”
* Humanising the brand	“Me in every meeting”

Posted by: ● Facebook user * Netflix

The entire dataset was carefully re-read again during the axial coding, and the open codes were assessed on their relevance to the research question and interpreted in relation to Ang’s (2011) pillars of CoRM. Codes formed during the open coding were merged into overarching, mutually exclusive groups, and redundant codes were removed. The result of the axial coding process was a list of 9 overarching groups and 35 codes. For example (see Table 4):

Based on these groups and codes, a final list of dominant codes, or themes, was created during the selective coding phases. The previously discussed literature was heavily taken into consideration when defining the themes and subthemes, whilst still keeping an open mind towards identifying themes that might not be explained by previous literature. As a result of this coding phase, three overarching themes with each three subthemes were defined (see Table 5).

Table 4. Example of axial codes.

Example of group	Example of codes
Disloyalty within the community	- Threatening to cancel subscription - Notifying of subscription cancellation - Leaving Netflix for competitor - Brand saboteurs

Table 5. Themes of selective coding.

Theme	Subtheme
Lopsided dedication	Prevailing content management Non-engaging passivity Scarce proactive webcare
Predominant brand humanisation	Netflix as a living, feeling peer Arbitrary distance Lacking self-brand connections
Rejection induced entitlement	One-sided relationships Facebook user entitlement Disloyalty

Table 6. Netflix’s peer language.

Netflix	Facebook user	Similarity
“tbh wish i was invited” “that sounds like a lot of fun tbh!”	“Tbh, I can understand that you might not like … ”	Use of popular abbreviation ‘tbh’. Meaning: to be honest
“LISTEN it could work” “its SO good”	“I STILL call him Fox” “put some RESPECK on her name” “I’m really PROUD of you guys”	Using all capitals to emphasise a word
“love her!!!!” “right??????!”	“OMG!!!” “Casper and The Dark Crystal?!?!?! I am so excited for May!!!!”	Using extra punctuation marks for emphasis
sameeee Enjoyyyyy	“boyeeee”	Use of extra vowels

All in all, the content was coded using a codebook that was designed both inductively (e.g. ‘FB user: questions unanswered’) and deductively (e.g. ‘Netflix: human-like’). This codebook was developed in Atlas.ti, a software designed for qualitative research that reduces the margin of error intrinsic in manual coding. Atlas.ti also allows the researcher to generate simple totals and percentages of the number of messages falling into each identified theme. In the final step of the coding process, a list of three core themes was defined (Boeije, 2010).

Ethics

All the data used are publicly available online. McKenna et al. (2017) refer to an overall academic agreement which states that researchers are free to use public data (e.g. social media comments) in their project. However, we believe that researchers should still be actively involved in protecting social media users and their privacy. Therefore, none of the (nick)names from the internauts whose comments were included in the sample were mentioned in the paper to ensure a level of anonymity. All direct quotations included in this article were attributed to either ‘(Netflix)’ or ‘(Facebook user)’. In addition, if Facebook users or Netflix tagged (other) internauts in their comments, these tagged (nick)names were replaced in the dataset with <tagpeerinternaut> or <taginternaut>.

Analysis

The analysis of the 100 posts and 826 comments collected from the Netflix-US Facebook page led to the identification of three overarching themes and nine sub-themes, as the following table illustrates.

Lopsided dedication

The analysis of Netflix’s comments and posts represents the company's willingness towards maintaining its social media-based brand community. However, this dedication seems to be lopsided and almost solely concentrated on a rather small part of community management.

Prevailing content management

The 100 unique posts analysed in this study were posted in 44 days, averaging Netflix’s posting ratio to roughly 2 posts per day. This suggests Netflix constant dedication to generating its own and new content. A closer look at the dataset, however, reveals that the brand generally posts 3 to 5 times a day, although there were 4 out of 6 weekends in which Netflix did not post any content on Facebook-US. This not only lowered the average, but it also suggests an interesting pattern within Netflix’s posting schedule—that is, a low posting activity during the weekends.

Concerning the role of the content posted, 27% of the posts aimed to entertain (n = 27), 16% played an informational role (n = 16) (e.g., explanations about creative choices, information about the casting), and a majority of them combined both (57%, n = 57). Furthermore, every post within the dataset contained a visual component such as an image or video, of which videos were most prevalent (n = 89).

Besides providing brand-related information, Netflix addressed controversial topics (e.g. minority representation) in its posts by linking them to the shows included in its library or to the actors working in those shows. For example:

1:16 Video of interview [Seen & Heard] with ‘Special’ creator and star Ryan O’Connell talking about his experience with being gay and disabled and how ‘Special’ has helped him in accepting himself more. (Post 41: 04/18/2019 03:24PM | Netflix)

Results also revealed that Netflix tended to forget to add either the title or (expected) streaming date in posts that promote streaming titles. This practice was identified in 26% of the posts (n = 26). In some cases, these details were later added in a separate comment underneath the post and, interestingly, all of these comments were posted approximately 9 hours after the initial posts, possibly indicating a certain strategy. More than once, however, fans have indicated that they find this strategy “annoying” and “unprofessional”. The hypothesis that Netflix is deliberately leaving out information to later publish it in a comment to generate social buzz seems to be plausible.

Non-engaging passivity

Netflix’s US Facebook page is filled with complaints from its subscribers, suggesting a low level of brand satisfaction. These are related to complaints on price hikes (n = 17), current content offered (n = 44), and show cancellations (n = 68). Out of all of the complaints addressing show cancellations and price hikes, none of them received a response from Netflix.

“You keep raising prices and dropping the best shows. You do have competition now. I’m one more price hike or show cancellation away from cancelling you.” (Facebook user | 04/27/2019 06:23PM)

“Wtf Netflix?!! They just said on the news that you’re raising prices AGAIN!

You just did!!! Time to shop around I guess.” (Facebook user | 04/03/2019 12:20AM)

“Netflix Why should I watch anything you put out? You’ll just cancel it after a few seasons for something new and cheaper after you raise prices again.” (Facebook user | 05/01/2019 07:09PM)

The fact that around a third of the collected subscriber comments contained a negative sentiment towards Netflix suggests that the brand does not engage in censorship. The passivity of Netflix’s moderation extends beyond the intervention of negative commenting situations. Additionally, a significant part of the negative comments posted by subscribers is directed towards their peers.

“Piss off, and whine somewhere else.” (Facebook user in response to peer internaut | 04/30/2019 12:59AM)

“<tagpeersubscriber> You’re annoying alright.” (Facebook user in response to peer internaut | 05/03/2019 12:55PM)

Despite personal attacks from subscribers to subscribers being pretty common (n = 64), the results of this study suggest that Netflix employs a non-engaging strategy, where the brand does not intervene at all in subscriber conflicts.

Scarce proactive webcare

From the 826 comments collected, Netflix wrote 6.1% (n = 50). Of these 50 comments, 16 were comments in which Netflix provided some details about its streaming titles (e.g. airdate of the show) which it seemingly forgot to include in the initial post. In only 6 comments, the company made public this type of information in response to a request from a subscriber. This leaves 28 comments from Netflix that were actually in response to subscriber comments. All in all, Netflix seems to use its scarce responses to answer to messages that do not directly require or request a response from the company, which is consistent with a proactive webcare strategy.

Predominant brand humanisation

Netflix as a living, feeling peer

In its comments, Netflix’s portrays the brand as a humorous, relatable peer and uses a human voice. Brand humanisation can be employed in several ways, ranging from a limited form of humanisation that assigns some human-like qualities to a brand, to actually fully humanising a brand by enlivening it with a human-like mind (Fournier, 1998; MacInnis & Folkes, 2017). Netflix chooses the latter, and enlivens the brand as if it is an actual person (Aaker, 1997; Veloutsou & McAlonan, 2012), by using singular terms such as ‘me’, ‘my’ and ‘I’ and transferring personal feelings in statements such as “sums up all my feelings”. Moreover, Netflix aims to give the brand a distinct, humorous personality, as it continuously uses humour in both its posts and comments.

In addition, Netflix seems to portray the brand as a living peer by mirroring the language of its subscribers. For instance, Netflix does not use capital letters at the beginning of a sentence, adds extra vowels, uses abbreviations, and capitalises words to emphasise certain ideas (see table 6).

Arbitrary distance

In some (seemingly) arbitrary moments (n = 7), Netflix stops portraying its brand as a person and breaks its ‘class clown’ character. It is in these moments when the company suddenly starts presenting its brand as an organisation (Geuens et al., 2009) and refers to itself in a more neutral and distant way by using, for example, plural pronouns instead of single pronouns.

“Here’s everything that’s coming to Netflix US in May!” (post 26: 04/24/2019 07:16AM | Netflix)

“Here’s the photo Ross and Noah Centineo sent us to accompany this news.”

(post 73: 04/01/2019 11:47AM | Netflix; Post includes a photo of both actors in a weird position with laughing faces.)

This inconsistency could, to some extent, undo the effort that Netflix is putting into characterising its brand as a person. For example, post number 26 could have been written along the following lines to keep in line with the humanising brand strategy: “Here’s everything that’s coming in May!” or “here’s everything that’s coming to Netflix US in May! I’m so excited!”

Lack of self-brand connections

Even though Netflix tends to humanise its brand as a person, this humanisation does not seem to resonate with its subscribers, who do not identify Netflix as a singular, living peer. Despite a few examples of subscribers using the second-person-singular pronoun ‘you’ (n = 18), subscribers are twice as likely to address the brand as “you guys”, “itself”, “they”, or any other plural or objective way (n = 38). They are, this way, creating a certain distance between them and Netflix, and dehumanising the brand (MacInnis & Folkes, 2017).

Rejection-induced entitlement

One-sided relationships

As stated before, Netflix hardly ever responds to comments from its subscribers, reducing, this way, the likelihood of building long-term relationships with them. Moreover, there are plenty of examples (n = 367) in which subscribers compliment either Netflix streaming titles (n = 316) or Netflix in general (n = 51), which the brand ignores. In some comments, subscribers establish a connection with their personal lives and they share it on Netflix’s Facebook page (7.9%, n = 62). For instance, subscribers connect the airdates of certain shows with their birthdays: “Yay on my birthday! Thanks Netflix” (Facebook user | 04/26/2019 07:13AM). Others get more personal and share detailed information about themselves:

“My grandmother has Alzheimer’s and it is very hard to get her to watch anything, but for a week straight like 3 or 4 times a day, when we needed her to calm down, we would just play Our Planet and she was transfixed, each time we pressed play it was new to her again. Thank you for making something that helped us with her!” (Facebook user | 4/06/2019 03:49PM)

By not responding to these types of comments Netflix is, in a sense, rejecting those subscribers who are externalising the connection they experience with the brand. The lack of action from Netflix’s side could potentially be perceived as hurtful by subscribers who, despite opening up on Netflix's Facebook page, are left in a one-sided relationship. The following comment illustrates these feelings: “Netflix you don’t care about your viewers” (Facebook user | 05/01/2019 7:11PM).

Entitlement

Subscribers seem to get agitated over Netflix ignoring them. As one subscriber stated: “Never understood why Netflix needs a commentary thread on something they’re going to show regardless of what people say” (Facebook user | 03/24/2019 11:17PM). This dismay pushes some of them to keep posting messages until they are finally heard. “I’m asking this on every post until they respond! Join the cause!” (Facebook user | 05/01/2019 04:21PM) and “We’re not going away Netflix” (Facebook user | 05/02/2019 06:09PM) are two examples that reflect this irritation. There are also a few instances (3.9%, n = 31) in which subscribers are using the content of a post to mention a different streaming title, presumably in the hopes of finally getting a response from Netflix.

Disloyalty

Within the entire database, only one comment contained a subscriber's expression of loyalty to Netflix. Conversely, subscribers threatened the company with the idea of cancelling their subscription in 27 comments; while in 6, they communicated that they had already cancelled it. Most of these subscribers were complaining about their favourite shows being cancelled “without consideration for the viewers” (Facebook user | 05/02/2019 10:07PM). In fact, more than once subscribers indicated that they were “heart-broken” and “devastated” by the cancelation of a show, an action that the company did not justify to their satisfaction. One subscriber actually argued that if Netflix listened to its subscribers more and gave them more consideration, the company would have a very loyal subscriber base.

“Hey Netflix here’s an idea maybe if you stop cancelling great shows on their third season without any consideration for the viewers you would retain a huge loyal customer base. For now I’m waiting for Disney Plus.” (Facebook user | 05/02/2019 10:07PM)

As the previous comment illustrates, Netflix’s subscribers end up comparing the brand to its competitors, such as Disney Plus, and even to linear television. In these comparisons, Netflix is generally presented in negative terms in opposition to its competitor, presented often in positive terms.

“Netflix cancelled. Too many other apps out there to put up with their BS [bullshit].” (Facebook user | 04/27/2019 08:28AM)

“Bye bye Netflix. Hello Hulu and Amazon Prime!!” (Facebook user | 04/28/2019 07:56AM)

The missing feeling of we-ness can be considered as another factor that contributes to the environment of disloyalty within the Netflix brand community. Data reveal that in 2.9% (n = 23) of comments, subscribers express a certain feeling of belonging with the community. This is, for instance, the case of those subscribers who speak for the community as a whole while addressing certain feelings: “Netflix why would we get attached to anything ever again after your broke our hearts with Santa Clarita” (Facebook user | 05/01/2019 11:04PM).

Furthermore, there is another small group of comments (0.9%, n = 7) in which internet users not only Express disloyalty but also attempt to sabotage the Netflix brand by encouraging others to cancel their Netflix subscriptions.

“DON’T WATCH NETFLIX DON’T SUPPORT NETFLIX.” (Facebook user | 03/22/2019 03:38PM)

“Everyone should ‘unfollow’ Netflix to retaliate.” (Facebook user | 04/27/2019 07:05AM)

“We should cancel Netflix too!!!” (Facebook user | 04/27/2019 08:49AM)

Conclusion

Social media-based brand communities are strong tools for building relationships with Facebook users and for brand promotion, which explains their popularity amongst SVoD services (Laroche et al., 2012; Martínez-López et al., 2016). However, for organisations, social media is not exclusively about “joining the conversation” but requires central community management skills and practices (McCosker, 2017, p. 123). Netflix, a digital brand very responsive to new technological developments (Lobato, 2018), is often applauded by field professionals for its innovative and engaging social media conduct (see Beer, 2019; Duczeminski, 2019). Consequently, one could assume that Netflix is also revolutionary, or at the very least adequate, in managing its brand communities. However, the analysis of 100 posts published by Netflix, and 826 comments published by Netflix and Facebook users on the company’s Facebook-USA page suggests that Netflix’s community relationship management (CoRM) hardly exists (RQ1), even on the Facebook page of its leading market. For instance, results from the empirical study suggest that Netflix only rarely answers comments from its subscribers. At the base of this analysis lies Ang’s (2011) pillars of community relationship management, namely connectivity and conversation (SQ1) and content creation (SQ2). This study has shown that activity on all interacting pillars is necessary to reach an adequate social media management strategy effective for the brand and its followers.

Results from this study suggest that Netflix employs little of any CoRM on its US Facebook page, as its efforts are almost solely focused on only half of one of the abovementioned three pillars (RQ1). The first pillar of CoRM, which is connectivity, contains two important concepts: interaction facilitation and interaction management. Both of these concepts seem strangers to Netflix. The brand does not seem to be taking any steps towards stimulating positive interactions between Facebook users (Gummerus et al., 2012; Sung et al., 2010) or managing interactions that turn vicious (Berry & Seiders, 2008; Dineva et al., 2017; Sibai et al., 2015) (SQ1). By not engaging in connectivity measures, Netflix’s conduct reflects a lack of care for internet users who express themselves on its Facebook page. The brand seemingly does not have any interest in helping Facebook users fulfil their interaction needs or in protecting Facebook users who are being attacked by other internauts (SQ1). In a community as heterogeneous as Netflix’s brand community, the brand should be providing much more guidance and management (Dineva et al., 2017).

The second pillar of CoRM, which is content creation, relates to two important concepts: the encouragement of user-generated content creation and marketer-generated content creation. As has been established, Netflix is strongly focussed on its own content creation, and data show that the brand is employing many recommendable strategies to make its content relevant for Facebook users (SQ2). The brand is keeping an active posting schedule, providing informational and entertainment benefits (Gummerus et al., 2012; Waters et al., 2009), and it uses a predominantly humanising approach to give its posts character and make them appealing to Facebook users (Fournier, 1998; MacInnis & Folkes, 2017). Netflix, however, does not encourage sharing of user-generated content to be shared on the platform or any other engaging behaviours such as commenting, sharing, and liking, therefore only focussing on half of the content creation pillar (SQ2).

The third pillar of CoRM, which is conversation, is also very much neglected in Netflix’s CoRM. The brand hardly ever seizes opportunities to indulge in positive interactions with Facebook users, therefore missing numerous chances that could be very beneficial for the brand. Moreover, the brand seems to blatantly ignore negative messages, therefore in all likeliness only increasing the number of negative messages that it receives (Schamari & Schaefers, 2015; Van Noort et al., 2015) (SQ3).

The findings of this study also corroborate what several other researchers (e.g. Dineva et al., 2017) have previously found, which is that brands generally tend to ignore negative interactions within communities, such as internaut-to-internaut conflicts and complaints from Facebook users. Despite scholars and marketing specialists encouraging brands to consider negative interactions as community opportunities, Dekay (2012, p. 289) has found that brands generally tend to ignore these negative interactions, which is also a characteristic of Netflix’s passive approach to CoRM.

In sum, and despite Netflix’s being a strong content creator and using a human voice to engage its audience, data suggest that the brand is sometimes lacking some consistency in its content and is not taking full advantage of its marketing and branding skills when it comes to managing its community on its US Facebook page.

Limitations and future research

This article presents the results obtained by means of a thematic analysis of more than 900 messages published on the Netflix-US Facebook page within a time frame of 44 days. Despite the accuracy of this exploratory analysis, a longitudinal study would shed more nuanced light on Netflix’s CoRM, allowing, for example, to identify potential changes over an extended period of time. It would be also interesting for future research to focus on a comparative analysis on different Netflix Facebook pages (e.g. the US, the Netherlands, and Spain) and platforms (e.g. Facebook, Twitter, and Instagram) and to complement the netnographic approach taken in this study with focus groups and/or interviews with Netflix subscribers. This would allow researchers to go deeper into subscribers’ thoughts and feelings on this subject. Furthermore, accessibility to the inner workings of the company (e.g. interviews with the community managers) and to the direct messages that Netflix potentially receives from its subscribers would provide a deeper understanding of the brand’s strategy, as well as the possibility of identifying whether some comments have been deleted or not.

Despite its limitations, this project has provided new insights into a body of literature on social media-based brand community management, with a special emphasis on how the biggest SVoD platform of the moment, Netflix, manages its social media marketing strategy in its leading market. The investigation has also helped identify new lines of research, and its findings can be used in future studies to inform the creation of, for example, an interview guide used to interview Netflix brand community managers about their conduct on the US Facebook page.

Disclosure of potential conflicts of interest

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Alana Hendrikx

Alana Hendrikx finished her master's programme in media studies in 2019 and is currently employed as a copywriter at a social media agency. She is particularly interested in audience studies.

Deborah Castro

Dr. Deborah Castro is a lecturer at the Department of Media & Communication at Erasmus University Rotterdam, The Netherlands. Her research interests lie in the fields of television and audience studies.

Notes

1. As the fourth pillar, collaboration, is mostly used in project-related communities such as Wikipedia (Ang, 2011, p. 35), this pillar is not taken into consideration for this study. Furthermore, and due to the intersections between pillar one (connectivity) and pillar three (conversation), we discuss these together.

2. The comments were sampled using Facebook’s ‘most relevant’ setting, which sorts comments to show those that received many interaction or that have been replied to by a verified account, such as Netflix’s account, first (“Wat betekent meest relevant bij een paginabericht?”, n.d.).