People want to see people? Personalization on Facebook as a tool for corporate communications

Nora Denner (Department of Communication, Johannes Gutenberg University of Mainz, Mainz, Germany)
Hannah Schneider (University of Konstanz, Konstanz, Germany)

Corporate Communications: An International Journal

ISSN: 1356-3289

Article publication date: 4 October 2022

Issue publication date: 3 January 2023

171

Abstract

Purpose

Social networks were created to connect with friends. Therefore, communication in social networks allows addressing individual contacts of each user and is often rather private by nature. Organizations can use this to communicate frequently and personally with their stakeholders. Therefore, this study investigates how organizations use personalization in their social media communication.

Design/methodology/approach

This study uses a quantitative content analysis of Facebook posts from ten large companies (N = 500). The posts were analyzed regarding the sub dimensions of personalization, individualization and privatization, as well as type of post, tonality, emotions, user reactions and topics.

Findings

Companies at least partially personalize their communication on Facebook. Overall, 28% of all posts were personalized. Personalized posts were almost always positive regarding tonality and mostly positive regarding emotions. While the personalized posts in this study have fewer user reactions than non-personalized posts, they still have a rather high absolute number of reactions. Regarding personal characteristics, results show that professional competence and appearance were mainly addressed. Concerning privatization, only very few posts showed an individual in a private setting.

Originality/value

The present work gives an overview of how personalized posts differ from non-personalized posts and looks at depicted individuals as well as the use of private elements. By applying the conceptualization of Van Aelst et al. (2012) and by looking at organizations’ use of personalization on the social media platform Facebook, it sheds light on an area that has not been the center of attention so far and helps to expand the current state of personalization research.

Keywords

Citation

Denner, N. and Schneider, H. (2023), "People want to see people? Personalization on Facebook as a tool for corporate communications", Corporate Communications: An International Journal, Vol. 28 No. 1, pp. 30-47. https://doi.org/10.1108/CCIJ-05-2022-0055

Publisher

:

Emerald Publishing Limited

Copyright © 2022, Emerald Publishing Limited


For corporate communications, social media is a main contact point for communicating with stakeholders (Lock, 2019) and can be an essential tool to enhance, maintain, and protect the reputation of organizations (Ji et al., 2017). In social media, organizations do not have to rely on journalistic gatekeepers anymore and can provide their own content directly accessible to the public (Metz et al., 2019), engage in symmetrical interactions (Macnamara and Zerfaß, 2012), and facilitate dialog with their stakeholders (Solis and Breakenridge, 2009). Especially the commenting and sharing functions of social network sites (SNS) such as Facebook or Twitter offer the possibility to directly interact with stakeholders and build a relationship with them (Manetti et al., 2016).

One communicative strategy that companies can apply in social media is personalization. Personalization can be broadly defined as a shift in focus from issues to people (Adam and Maier, 2010) and is concerned with the role and place of individuals and the implications of this for organizations (Pruysers et al., 2018). There are various studies looking at personalization in corporate coverage (e.g. Brettschneider and Vollbracht, 2011; Denner et al., 2021, 2018; Park and Berger, 2004), but only few studies investigate how organizations use personalization in their external communication (e.g. Talanow, 2015; Thummes and Malik, 2015). Especially in SNS, which were developed as networks for friends, communication is often very personal because users share private details. Moreover, social networks provide individuals and organizations with their own communication channel where they can address their followers and contacts (Metz et al., 2019). Studies looking at personalization in SNS often analyze self-personalization of, e.g. politicians (Lee and Oh, 2012; McGregor, 2018; Metz et al., 2019). It is therefore reasonable to expect that organizations as well apply strategic personalization in their social media presence.

Thummes and Malik (2015) show that organizations use personalization on social media to increase their credibility, attract more attention, and persist longer in the memory of the audience. However, it is not clear how organizations implement personalization on social media. This paper therefore aims to contribute to this research gap by taking a closer look at how organizations use personalization in their social media communication and at the same time meets the demand for more research on organizational personalization (Denner et al., 2019; Eisenegger and Schranz, 2013). We aim to give an overview of how personalized posts differ from non-personalized posts regarding type of post, topics, and user reactions and look at depicted individuals as well as the use of private elements. To achieve that goal, we conducted a quantitative content analysis of posts from ten of the companies with the most followers on Facebook (N = 500). Consequently, this study is the first, to our knowledge, to quantitatively investigate personalization in the social media communication of companies.

Personalized public relations

The concept of personalization is mainly rooted in political communication research (e.g. Adam and Maier, 2010; Cross et al., 2018; Rahat and Sheafer, 2007; Van Aelst et al., 2012) and addresses the focus on individuals at the expense of organizations and institutions. It is often discussed in relation to media coverage (Van Aelst et al., 2012) and from this perspective, it can be defined as a process in which news stories increasingly feature single actors to make a story more newsworthy (Holtz-Bacha et al., 2014; Strömbäck, 2008). This understanding of personalization is not to be confused with the understanding in the context of customization or tailoring (Teeny et al., 2021). Here, personalization means that the display of content on social media, online market places, and websites in general differs regarding the preferences and/or previous behaviors of users (e.g. Shanahan et al., 2019). Hence, the content is personalized for every single user.

Well-applied personalization is an appropriate way of reducing complexity and attracting the attention of recipients (Brettschneider and Vollbracht, 2011; Lee and Oh, 2012). Moreover, a person's image acts as a mental link or shortcut (Gaines-Ross, 2000) to perceptions about the organization. This shortcut serves as an orientation for stakeholders and helps them to quickly form an opinion about the company (Gaines-Ross, 2000). A sympathetic and charismatic representative can thus help to strengthen the company's reputation (Denner et al., 2018). Additionally, like in politics where personalization allows the public to identify with issues and candidates more easily, personalization can help stakeholders to identify with the organization (Meeks, 2012; McGregor, 2018). At the same time, the reputation and image of a corporate representative (e.g. the CEO) can influence the perceptions and evaluations of the organization (Pollach and Kerbler, 2011).

Nevertheless, personalization as a concept has not been in the center of PR research so far (Eisenegger and Schranz, 2013). The few studies that exist mostly analyze how companies and their leaders meaningfully communicate with their stakeholders by using personalization (e.g. Hwang, 2012; Motion, 1999; Treadway et al., 2009) as well as how CEOs are portrayed in the media (e.g. Denner et al., 2018; Engelen, 2012; Park and Berger, 2004; Westphal and Deephouse, 2011). A lot of studies focusing on how companies should communicate tend to center around best practices (e.g. in the context of crises, Loretta, 2002; Bernstein, 2002) or focus on the celebrity status of CEOs (e.g. Ranft et al., 2006; Treadway et al., 2009). However, there is hardly any distinction between different forms and modes of personalized communication.

For this study, we therefore adapt the conceptualization of Van Aelst et al. (2012) rooted in political communication research and distinguish into two dimensions: individualization and privatization (p. 204–205). Individualization describes the focus on a person in general whereas privatization means the focus on the private life of a person and/or their personal characteristics.

Personalization on social media

PR professionals consider social networks to be the most important tool for online communication (Wright and Hinson, 2011, 2015). However, studies on the strategic use of SNS show that many companies do not exploit all their potential. They usually use SNS only for product information or self-promotion and not as collaborative platforms that can serve to build relationships with the stakeholders through its bidirectional communication (Thummes and Malik, 2015; Valentini, 2015). While it was debated whether the expansion of television has increased personalized reporting (Adam and Maier, 2010), communication on social media is assumed to be “intrinsic” (Enli and Skogerbø, 2013, p. 759) and “by nature” (Metz et al., 2019, p. 3) driving personalization (see also Graham et al., 2018; McGregor, 2018). Social media allow every individual and every organization to have their own personalized communication channel (Metz et al., 2019). Furthermore, the interactive nature of social media fosters engagement through commenting, liking, or sharing posts (Cho et al., 2016; Zhang et al., 2022).

Moreover, according to news value theory, companies can benefit from personalization because the media use personalization as a selection criterion (e.g. Eilders, 2006; Galtung and Ruge, 1965; Staab, 1990; Wayne and Murray, 2009). Since social media serve as a source of information for traditional mass media (Lee and Oh, 2012), personalization on social media could be a useful strategy for organizations to place their issues and viewpoints in the media.

Drawing conclusions from political communication research on the use of personalization in social media we see that personalized tweets can generate more engagement from the public (Lawrence et al., 2016). Personalization on SNS can also evoke higher levels of social presence and parasocial interaction for individuals (Lee and Oh, 2012). Social presence in turn can create feelings of connection with others, even though the communication is virtual (Baym, 2010). Parasocial interaction means that people can form an interpersonal relationship with a figure and feel close to that figure (Horton and Wohl, 1956). Therefore, personalizing communication on social media can help to create a relationship between an organization and its stakeholders more easily (McGregor, 2018). Furthermore, Lee and Oh (2012) also showed that personalized posts perceived more attention and could be remembered better than non-personalized posts. Metz et al. (2019) conducted a content analysis of politicians' Facebook posts and found that self-personalization is used quite often as a communication style and is most often present in visual communication.

In PR research, personalization on social media has played a minor role so far. There is some research on how CEOs use personal accounts for communicating via SNS that found positive effects on the CEO's perception (Hwang, 2012) and on how publics engage with corporate CEOs on social media (Men and Tsai, 2016). However, a focus on organization's SNS-accounts and their communication is, to our knowledge, even more scarce. One exception is Thummes and Malik (2015) who analyzed the Facebook pages of 39 German companies. Just a quarter of the posts (24%) showed a personalized portrayal of the company and addressed the credibility or truthfulness of the company.

Put together, research suggests that personalization on social media is certainly relevant for corporate communications. However, we do not know how companies strategically use personalization as a tool on SNS. Therefore, we ask the following research question:

RQ1.

How many corporate posts are personalized?

We know from previous studies that personalized posts often include pictures or videos, because visuals can easily transmit personalized communication (Metz et al., 2019). A visual representation of people leads to a better impression of them and illustrates the communication and consequently, makes it easier for the audience to perceive the information (McAllister, 2007). Additionally, Metz et al. (2019) showed that politicians frequently employ emotional (self-)personalization on Facebook and Thummes and Malik (2015) emphasize a positive tonality of personalized posts. Therefore, our second research question asks:

RQ2.

Are there differences in terms of type (text, pictures, or videos) and valence (tonality and emotions) between personalized and non-personalized posts of organizations?

Previous studies have found that personalized posts tend to have more user reactions than non-personalized posts (Bar-Ilan et al., 2015; Metz et al., 2019). Since these studies are from the political arena, we aim to investigate user reactions in a corporate setting and ask:

RQ3.

Are there any differences between personalized and non-personalized posts regarding user reactions (likes, shares, and comments)?

In addition, it is also meaningful to explore potential differences regarding topics of corporate posts. In personalization research, it is assumed that especially crises lead to personalized communication. It is easier to blame a single person than a complex organization (Denner et al., 2019, 2021). Moreover, a content analysis of media reporting on large German companies shows that news related to the CEO often included topics like corporate strategy as well as corporate acquisitions and mergers (Brettschneider and Vollbracht, 2011). This leads to the fourth research question:

RQ4.

Which topics are salient in personalized and non-personalized posts on Facebook?

Depending on the goal of a post, it can be useful to display different individuals in the communication on SNS. For product related posts, customers might be ideal, while employees might be suited well for employer branding purposes. In particular, the role of CEOs is often discussed in PR research in the context of personalization (e.g. Denner et al., 2018; Sohn et al., 2009; Park and Berger, 2004; Zerfaß et al., 2016). A lot of this research suggests that leaders are well-suited to the personalization strategy, because they represent the organization both internally and externally and sometimes personify the organization (Park and Berger, 2004). Studies have shown that the image of the CEO is closely linked to that of the company (e.g. Gaines-Ross, 2000; Sauerhaft and Atkins, 1989). Therefore, we are interested in the following question:

RQ5.

Which individuals are in the focus of personalized corporate posts on Facebook?

Lastly, we take a look at privatization as a subdimension of personalization and how it is implemented in corporate social media posts. Studies on the effects of privatization on SNS show that it leads to more liking from the audience (Colliander et al., 2017) and generates more reactions (Enli and Skogerbø, 2013). There is hardly any research looking at private information from a personalization point of view, except for some political communication studies which suggest that a lot of privatization takes place on social media (Graham et al., 2018; Kruikemeier, 2014). Therefore, we ask:

RQ6.

Do organizations use privatization in their posts and how do they implement it?

Method

Procedure and sample

Since we are interested in formal and content-related patterns and properties of texts, we conducted a quantitative content analysis to answer our research questions. Content analysis is a research method for investigating the manifest and latent content of communicated material through classifying, tabulating, and evaluating in order to access its meaning objectively and systematically (Krippendorff, 2012). Our body of communicated material consists of corporate posts on Facebook, more specifically of posts on German fan pages of the most popular companies on Facebook. As of January 2021, Facebook had around 2.74 billion active users making it the largest social network worldwide (We Are Social et al., 2021). In Germany, it is the most regularly used social media platform (Beisch et al., 2019). To select the companies, we took a list of the 100 most popular pages on Facebook (Fan Page List, 2019, as of August 31st, 2019, and Lingner Consulting, 2019, as of August 31st, 2019). We excluded all non-corporate pages (e.g. celebrities or athletes), career pages since they are exclusively for recruitment of employees, and companies that did not have a German page. This resulted in a list of 36 companies and we randomly selected ten out of the 36 (see Table 1).

For each company, we randomly selected 50 of the last 500 posts the companies published on their Facebook fan pages before 31 August 2019. A total of 500 Facebook posts from the ten companies were examined. The main coding took place between 22 and 30 October 2019. Our categories, the data, and the analysis script can be found online on OSF: https://osf.io/r5kwy/.

Operationalization

For the content analysis, we relied mostly on existing studies (e.g. Brettschneider and Vollbracht, 2011; Denner et al., 2021; Van Aelst et al., 2012). The codebook contained formal and content-related categories. As formal categories, we coded the company, a unique post ID, whether the post was text-based, picture-based, video-based, or gif-based, a link, or a combination of any of those types, as well as the date and the time the post was published. We also included the number of likes, comments, and shares. These metadata have already been used in other content analyses of Facebook posts (e.g. Haßler et al., 2014; Mhamdi et al., 2018) and were generated automatically. The freely accessible software Facepager (Jünger and Keyling, 2019) was used for this purpose. It can store data via the API interface in a database that can be exported as a CSV-file. We also downloaded the content of the posts via Facepager to code our content-based categories.

The content-based categories included personalization and features of the posts in general. We captured if at least one person was mentioned in the text and/or the picture (if the posts contained a picture), how many people were mentioned, and how often they were mentioned. We also captured the name, gender, and position of the central person of the post.

Moreover, we included character traits of the central person in the post, relying on the operationalization of privatization by Van Aelst et al. (2012). We coded if competence, leadership, credibility, morality, rhetorical skills, and appearance were mentioned and included the context (professional or private). To examine private life, we also relied on Van Aelst et al. (2012) and used the categories family life, past time, leisure time, and love life. In addition, we coded if the posts were marked with a sender (i.e. an abbreviation) and the perspective of the post (first or third person), and if the recipients were directly and formally/informally addressed in the post. Additionally, we captured tonality (positive/negative/neutral), and emotions (positive/negative/neutral/none). For videos and pictures in the post, we also included if a person is depicted and captured the expression of the person and context of the picture (professional/private). Lastly, we also coded the main topic of the post.

Reliability test

We tested intracoder reliability since the coding was carried out by a single coder and used Krippendorff's alpha (Krippendorff, 1971) as the reliability coefficient, because it takes the number of expressions of a category into account and thus matches the likelihood of purely random matches (Krippendorff, 2012). The reliability values in the present study are α = 1.00 for the formal categories and between α = 0.68 and α = 1.00 for the content categories. Additionally, we had two variables where alpha was below 0.67. A problem that often occurs in reliability tests with relatively small numbers of cases and skewed variables is that the variance is so small that it affects Krippendorff's α. This was the case with two variables [1]. Therefore, we checked the actual coding ex post. Specifically, we checked whether the characteristic occurred (in the case of variables for which the characteristic usually does not occur). The researchers agreed with the decisions of the coder, so that despite a failed reliability test or a lower reliability with Krippendorff's α, reliable coding of these variables can be assumed.

Results

As described, we coded 50 posts of 10 companies each (N = 500). 4% of the posts consisted of text only, 64% were a picture, 24% a video, 6% combined picture and text, 1% combined video and text, and 1% were a link or a GIF.

RQ1 asks how many corporate posts on Facebook are personalized. We define a post as personalized if an individual person is featured in the text, picture, and/or video. Overall, 28% of the examined Facebook posts were personalized. However, the rate differs extremely between the companies. The companies Intel (6%) and Walter Tools (8%) personalized the least. The most personalized posts were published by BASF (50%) and Bayer (40%; see Figure 1).

To address RQ2, we compared personalized and non-personalized posts regarding type and valence of the posts. We combined video with video and text as well as picture with picture and text for the type of the post and GIFs and links to the category “others”. Overall, a cross table shows that posts with videos tend to be personalized a bit more often other posts, but the differences are not significant (see Table 2).

Regarding valence, we investigated emotionality and tonality of the posts. Overall, 29% had a positive tonality, while 71% were neutral. Regarding emotions, 66% did not contain any emotions, while 34% did contain positive emotions, negative, or both. The tone of the personalized posts was almost only positive (92%), compared to 4% of non-personalized posts. Only a few personalized posts were phrased neutrally (8%), compared to 96% of non-personalized posts. Not a single post had a negative tone. Therefore, there is a significant difference between personalized and non-personalized posts regarding the tonality of the post (χ2(1) = 371.26; p < 0.001; V = 0.86).

More than half of the personalized posts showed positive emotions (62%), compared to 3% of non-personalized posts. Only a small proportion contained negative emotions (1%; 0% for non-personalized posts). 37% were neutral (containing positive and negative emotions), compared to 6% of non-personalized posts. 0% contained no emotions an at all, compared to 92% of the non-personalized posts. The differences between personalized and non-personalized posts were significant (χ2(3) = 384.83; p < 0.001; V = 0.88). Taken together, personalized posts show emotions, and they are often positive or positive and negative, whereas non-personalized posts often do not contain emotions. Referring to RQ2, the data reveals no significant differences regarding the type of post between personalized and non-personalized posts. However, personalized posts tend to be more positive regarding tonality and emotionality.

RQ3 asked if there are differences between personalized and non-personalized posts regarding user reactions (likes, shares, comments). We conducted a t-test for each user reaction. Regarding likes, non-personalized posts (M = 278.00; SD = 993.00) had significantly more likes than personalized posts (M = 86.00; SD = 164.00), t (498) = 2.26, p < 0.05, d = 0.23. Regarding comments, non-personalized posts (M = 81.70; SD = 380.00) had significantly more comments than personalized posts (M = 14.80; SD = 60.20), t (498) = 2.06, p < 0.05, d = 0.21. Regarding shares, non-personalized posts (M = 21.60; SD = 59.60) had significantly more shares than personalized posts (M = 10.3; SD = 25.30), t (498) = 2.14, p < 0.05, d = 0.21. However, especially for the likes of a post, there is a large variance in the reactions. To answer RQ3, we found that personalized posts retrieved less reactions than non-personalized posts.

The next research question (RQ4) asks for the topics that go along with personalization of corporate posts on Facebook. For clarity, we only looked at the upper categories and combined sports, politics, economics, social life, and culture to one category (“others”; see Table 3).

Most frequently, internal company topics were presented in a personalized way (28%) more often than in a non-personalized way (9%). This category consists of posts about the company in general, the history of the company, values of the company, posts about the personnel of the company, business figures and others. Of rather external topics, customers were the most frequently addressed in personalized posts (20%) than in non-personalized posts (1%). Additionally, events are more often presented in a personalized post (17%) than in a non-personalized post (12%).

To answer RQ4, it can be stated that personalization on Facebook often goes hand in hand with internal company topics. But also, issues dedicated to customers and the promotion of events are often personalized.

RQ5 asked which individuals are in the focus of personalized corporate posts. Employees (45%) and customers (25%) were the most frequent individuals in personalized company posts, followed by public figures (15%) and cooperation partners (12%). Other individuals were in the focus in 3% of all personalized posts. In terms of the position of the shown individuals, 33% of the personalized posts did not mention the position or function of the individual, while 67% did. RQ5 can therefore be answered with the notion that companies most frequently present employees and customers in their Facebook posts.

The last research question (RQ6) deals with privatization. We analyzed both mentioned character traits and private life of the depicted individuals. Overall, 23% of all personalized posts (6% of all posts) contained either character traits in a private context or the private life or both. Of all personalized posts, 16% of all personalized articles addressed character traits in a private context, and 20% of all personalized articles addressed the private lives of the individual depicted.

Looking at the five trait categories, we see that the presentation of the different character traits varied greatly in their frequency. Appearance was addressed in three-quarters (75%) of the personalized posts. Competence (58%) and rhetorical skills (53%) were featured in about half of the posts. In contrast, leadership (27%), morality (11%) and credibility (2%) were addressed less frequently. It is noticeable that the character traits, if they were shown, were predominantly shown in a professional context.

Of all posts mentioning the respective traits, leadership and credibility were never shown in a private context (0%), competence (9%) and appearance (10%) similarly rarely, and rhetorical skills (21%) and morality (33%) were most frequently accompanied with a private setting. Only a few of the personalized company posts addressed their private life. Love life (2%), family life (4%) and past life/childhood (5%) were rarely shown. Leisure time/hobbies (15%) were shown most frequently. Regarding RQ6, only about 23% of all personalized posts are privatized. Therefore, privatization is not used much, especially looking at the overall share (6%). If privatization is used, leisure time and hobbies as a form of the depiction of private life is used most frequently, as well as morality or rhetorical skills as a trait mentioned in a private context.

Discussion

Companies at least partially personalize their communication on Facebook. Overall, 28% of all posts were personalized (compared to 24% in the study of Thummes and Malik, 2015). Personalized posts were almost always positive regarding tonality and mostly positive regarding emotions. Since personalization is meant to build a positive reputation (Denner et al., 2018), it is not surprising that the individuals shown in the posts are portrayed in a positive way. In addition, other research has also demonstrated that emotions are often displayed in personalized posts (Metz et al., 2019). Since our results show that personalization is often accompanied by (positive) emotions and a positive tonality, this would be interesting and important to investigate further. However, there is no standardized operationalization for emotions in connection with personalization. Van Aelst et al. (2012) also do not consider emotions in their operationalization but mention their importance. Besides, our results regarding the number of reactions show that personalized posts have fewer likes, comments, and shares than non-personalized posts. This finding is contractionary to other studies (Metz et al., 2019; Thummes and Malik, 2015). While the personalized posts in this study have fewer reactions than non-personalized posts, they still have a rather high absolute number of reactions. Thus, it is reasonable to expect that the number of reactions is not only caused by the presence of personalization. It potentially also depends on, e.g. the time the posts were published, whether the posts were displayed in the users' feed due to algorithms, or whether friends of users had already reacted to them.

Regarding the presentation of character traits, the strategy of companies differs greatly from the perception of recipients. On social media, those who present themselves as likeable, audience-oriented, and authentic, are considered credible; professional character traits such as competence and leadership qualities are less important (Wehmeier and Winkler, 2012). Metz et al. (2019) further showed that professional personalization has no influence on audience engagement and reactions, while private personalization can positively increase audience engagement. The present results, however, show that it was mainly professional competence and appearance that were addressed. A professional presentation of individuals in social media posts might contradict the goal to achieve an authentic and likable perception in those posts. This could also be an explanation for the fewer user reactions for personalized posts.

Concerning privatization, only few posts showed an individual in a private setting (either character traits or private life). Traits were mainly shown in a professional context and less in a private one and leisure time was the only part of private life that was mentioned in more than 10% of all personalized posts. Thus, the results do not correspond to the studies (on politicians) which state that a lot of privatization takes place on social media (e.g. Graham et al., 2018; Kruikemeier, 2014). In this context, privatization leads to more sympathy among the audience (Colliander et al., 2017) and generates more reactions (Enli and Skogerbø, 2013). The companies in the present study seem to underestimate the impact of private representation and rather avoid it. The explanation for these results might be that the companies may not want to appear unprofessional, even on social media. Furthermore, it is conceivable that employees and especially the top management of the companies are not willing to disclose private information about themselves. This reluctance contradicts the usual behavior of SNS users, who share a lot of personal information on such platforms (Metz et al., 2019).

However, it seems that customers are more likely to be privatized. Since they are not directly part of the company, it may seem less unprofessional to show them privately. Additionally, customers may also identify more easily with other (privatized) customers. According to this assumption, the recipients sympathize with a privately portrayed customer who uses the company's products or services and are therefore also tending to buy the products or services. The company Bayer, for example, presents a whole range of different customers personally and even calls on people to “get to know” them in Facebook posts (see Figure 2).

In terms of the topic that is accompanied by personalization, the results suggest that companies mainly use personalization to inform their stakeholders and the public about internal matters. The spectrum of topics is broad and ranges from personnel topics to company values and history to business figures. There is not a central topic that is predominantly personalized. It is particularly interesting to observe that the topic of corporate crisis was very rarely presented. This contradicts the assumption in research that personalization is particularly strong in times of crisis (Denner et al., 2021; Eisenegger, 2010). However, this could also be because in the studied period no company we studied was in a major crisis that needed to be covered. Only one post by (the company) Facebook addressed a crisis (see Figure 3). This was a long, detailed statement by CEO Mark Zuckerberg about Facebook passing on data, especially to the company Cambridge Analytica. In this corporate crisis, the CEO personally stood up for it with his name and face. Even the news media addressed this post (e.g. Salinas, 2018; Wong, 2018). It can therefore be concluded that although crises are not mainly the subject of personalized posts, conversely, posts that deal with a crisis are personalized.

After company internals, customers were the second most frequently mentioned topic which is in line with the results of RQ2 since customers often were the subject of personalized posts. Therefore, it can be assumed that companies want to strengthen their relationship with customers through personalized posts. Other topics that were more frequently addressed in personalized posts were events and products. The focus was mainly on marketing, as the posts about these were used to promote events and products. This is consistent with the study findings that companies use Facebook to promote themselves and their products (Thummes and Malik, 2015; Valentini, 2015) and apparently personalization is a strategy for this promotion. For example, the announcement of a trade fair stand of RATIONAL AG focuses on two employees (Figure 4).

Even though the present study provides important insights, there are some limitations of our study that will be addressed here. The study only looks at cross-sectional data and does not offer a longitudinal view. Moreover, we only took ten companies into account which posting behavior varied to a great extent. Future research could take different branches of industries as well as different types of organizations into account. Smaller companies could also be included in the sample. It can be assumed that more personalization can be found among them, since other authors have already noticed that reporting is more personalized among smaller companies (Brettschneider and Vollbracht, 2011).

Regarding our codebook, some categories were only coded rather superficial, e.g. emotionality and tonality. Additionally, the analysis of visual elements in corporate posts is limited to a minimum in this paper, as it only constitutes one visual category in the overall analysis. Therefore, a more detailed categorization could certainly have provided more information in the visual material. In addition, the collected data via Facepager represents a snapshot at the time of the data export. Exported data, such as posts and their reactions, is a variable and changing content. Especially the number of reactions such as likes, comments and shares can be expected to change. It should also be noted that the posts can be edited or deleted by the authors.

Furthermore, over a quarter of the posts in the study contained a person in the text or image, but it is not always clear whether these people are representatives for the respective company. In addition, the present study only refers to strategic communication of companies on Facebook and the results are not necessarily transferable to other social media platforms. Therefore, it is important to conduct similar studies with other social networks. Networks that are of great importance in the business world, such as LinkedIn, could be examined to see whether personalization is prevalent there and how it is implemented. Since these networks are also relevant for the job market, it can be assumed that more employees are presented there in a personalized way so that potential new employees can identify with them and find out more about the company as an employer.

Conclusion

To sum up, we found some cross-organizational patterns regarding personalization on Facebook. If organizations use personalization in their posts on Facebook, they seem to emphasize them with visualizations. Moreover, these posts are often accompanied by a positive tonality and emotions. Organizations communicate internal topics in a personalized way as well as promote events and products with this strategy. Hence, they show employees and customers. Surprisingly and in contrast to previous research, the studied companies lacked in privatized communication on Facebook. When character traits were depicted, it was often in a professional context. This is without a doubt adequate for strategic communication, however, on SNS, privatization can lead to more sympathy and can generate more user reactions and therefore might be a good tool for companies.

Based on our study, some implications for both theory and practice can be derived. Theoretically, the study shows that the conceptualization of personalization by Van Aelst et al. (2012) can also be used for the analysis of corporate communications on SNS. Furthermore, the present study also expands the current state of personalization research by focusing on the communication of companies in social media. Thereby, it sheds light on an area that has not been in the center of research so far but is still timely and important. The results show the concrete implementation of personalization in corporate posts on Facebook and present cross-company patterns of this strategy.

Practically, personalization can be a useful tool for companies to attract the attention of the audience via their social media posts (Thummes and Malik, 2015) and to appear more approachable for their stakeholders. Furthermore, personalization can also help stakeholders to identify more easily with the organization and it reduces the complexity of messages (Eisenegger, 2010). Our results also offer some recommendations for practitioners. First, since all organizations we studied hardly used personalization, the first recommendation is that organizations could pay more attention to personalization in their social media strategy.

Regarding the implementation of personalization on SNS, the companies studied here tended to avoid private traits and private life. While professionalism and competence are considered untrustworthy on social media, sympathy and closeness to the audience have an authentic effect (Wehmeier and Winkler, 2012). It is therefore advisable for organizations to focus less on being professional and more on appearing transparent and authentic on SNS. One way to achieve this would be, for example, the private presentation of a person or at least characteristics with which the recipients can identify.

Furthermore, it is suggested that companies should orient themselves more strongly on the communication behavior of other social media users. For example, posts could be more informal and personal. Emojis are also part of the normal communication behavior of Facebook users, and it could be possible for companies to use emojis more frequently in their posts. All in all, companies should focus less on appearing professional and more on being authentic on SNS. Other channels, such as the company website, are better suited for presenting professionalism.

Figures

Percentage of personalization of the company posts

Figure 1

Percentage of personalization of the company posts

Personalized post from Facebook concerning a crisis

Figure 2

Personalized post from Facebook concerning a crisis

Privatization of a customer at Bayer

Figure 3

Privatization of a customer at Bayer

Personalized event promotion from RATIONAL AG

Figure 4

Personalized event promotion from RATIONAL AG

Overview of the companies in the sample

CompaniesNumber of fans (as of 31.08.2019)
Facebook214,432,289
McDonalds79,475,267
Instagram60,200,014
Intel38,166,294
Bayer AG2,803,383
ABB AG618,004
BASF452,178
Walter Tools284,182
Rational AG221,071
Bosch Rexroth AG142,818

Personalized and non-personalized posts compared regarding type of post (in %)

PersonalizationType of post
Text onlyPicture/picture and textVideo/video and textOthers (GIFs/Links)Overall (N)
No Personalization8174677572(362)
Personalization926332528(138)
Overall (N)100 (21)100 (348)100 (123)100 (8)100 (500)

Note(s): χ2(3) = 3.19; p > 0.05; V = 0.08

Topics of personalized and non-personalized posts (in %)

TopicPersonalizationOverall
Non-personalizedPersonalized
Products531141
Internal Issues92814
Events121713
Cooperations142
Daily life/off-topic12510
Science645
Customers1206
Others7128
Total100 (362)100 (138)100 (500)

Note(s): χ2(7) = 139.77; p < 0.001; V = 0.53; Chi-squared approximation may be incorrect

Note

1.

The two variables were (1) the context (professional or private) of the personal trait morality and (2) context of the personal trait leadership. Both were coded with 1 = professional, 2 = private, -99 = na.

References

Adam, S. and Maier, M. (2010), “Personalization of politics a critical review and agenda for research”, Annals of the International Communication Association, Vol. 34 No. 1, pp. 213-257, doi: 10.1080/23808985.2010.11679101.

Bar-Ilan, J., Bronstein, J. and Aharony, N. (2015), “Israeli parties and party leaders on Facebook during the 2013 election campaign”, iConference 2015 Proceedings, Newport Beach, CA, March 24-27, available at: https://www.ideals.illinois.edu/handle/2142/73671.

Baym, N.K. (2010), Personal Connections in the Digital Age, Polity Press, Cambridge.

Beisch, N., Koch, W. and Schäfer, C. (2019), ARD/ZDF-Onlinestudie 2019: Mediale Internetnutzung und Video-on-Demand Gewinnen Weiter an Bedeutung”, Media Perspektiven, Vol. 9, pp. 374-388.

Bernstein, A. (2002), “CEO communication gone bad”, Business Communicator, Vol. 3 No. 1, pp. 3-12.

Brettschneider, F. and Vollbracht, M. (2011), “Personalization of corporate coverage”, in Helm, S., Liehr-Gobbers, K. and Storck, S. (Eds), Reputation Management, Springer, Heidelberg, pp. 267-289.

Cho, M., Furey, L.M. and Mohr, T. (2016), “Communicating corporate social responsibility on social media: strategies, stakeholders, and public engagement on corporate Facebook”, Business and Professional Communcation Quarterly, Vol. 80 No. 1, pp. 52-69.

Colliander, J., Marder, B., Falkman, L.L., Madestam, J., Modig, E. and Sagfossen, S. (2017), “The social media balancing act. Testing the use of a balanced self-presentation strategy for politicians using twitter”, Computers in Human Behavior, Vol. 74, pp. 277-285, doi: 10.1016/j.chb.2017.04.042.

Cross, W.P., Katz, R.S. and Pruysers, S. (2018), The Personalization of Democratic Politics and the Challenge for Political Parties, ECPR Press, London.

Denner, N., Heitzler, N. and Koch, T. (2018), “Presentation of CEOs in the media: a framing analysis”, European Journal of Communication, Vol. 33 No. 3, pp. 271-289, doi: 10.1177/0267323118763876.

Denner, N., Viererbl, B. and Koch, T. (2019), “A matter for the boss? How personalized communication affects recipients' perceptions of an organization during a crisis”, International Journal of Communication, Vol. 13, pp. 2026-2044.

Denner, N., Koch, T. and Senger, S. (2021), “Faces of companies. Personalization of corporate coverage in crisis and non-crisis periods”, Journalism, Vol. 22 No. 11, pp. 2747-2763, doi: 10.1177/1464884920901615.

Eilders, C. (2006), “News factors and news decisions: theoretical and methodological advances in Germany”, Communications, Vol. 31 No. 1, pp. 5-24, doi: 10.1515/COMMUN.2006.002.

Eisenegger, M. (2010), “Eine phänomenologie der Personalisierung”, in Eisenegger, M. and Wehmeier, S. (Eds), Personalisierung der Organisationskommunikation: Theoretische Zugänge, Empirie und Praxis, Springer, Wiesbaden, pp. 11-26.

Eisenegger, M. and Schranz, M. (2013), “Personalisierung: ein zu Unrecht der PR-Praxis überlassenes Thema”, in Hoffjann, O. and Huck-Sandhu, S. (Eds), UnVergessene Diskurse - 20 Jahre PR- und Organisationskommunikationsforschung, Springer, Wiesbaden, pp. 339-362.

Engelen, P.-J. (2012), “What is the reputational cost of a dishonest CEO? Evidence from US illegal insider trading”, CESifo Economic Studies, Vol. 58 No. 1, pp. 140-163, doi: 10.1093/cesifo/ifr031.

Enli, G.S. and Skogerbø, E. (2013), “Personalized campaigns in party-centred politics: twitter and Facebook as arenas for political communication”, Information, Communication and Society, Vol. 16 No. 5, pp. 757-774, doi: 10.1080/1369118X.2013.782330.

Fan Page List (2019), “Top 100 Facebook fan pages”, available at: https://fanpagelist.com/category/top_users/view/list/sort/fans.

Gaines-Ross, L. (2000), “CEO reputation: a key factor in shareholder value”, Corporate Reputation Review, Vol. 3 No. 4, pp. 366-370.

Galtung, J. and Ruge, M.H. (1965), “The structure of foreign news: the presentation of the Congo, Cuba and Cyprus crises in four Norwegian newspapers”, Journal of Peace Research, Vol. 2 No. 1, pp. 64-90, available at: https://www.jstor.org/stable/423011.

Graham, T., Jackson, D., Broersma, M. and Wright, S. (2018), “The personal in the political on twitter: towards a typology of politicians' personalized tweeting behaviours”, in Schwanholz, J., Graham, T. and Stoll, P.-T. (Eds), Managing Democracy in the Digital Age, Springer International Publishing, Cham, pp. 137-157, doi: 10.1007/978-3-319-61708-4_8.

Haßler, J., Maurer, M. and Holbach, T. (2014), “Advancement through technology? The analysis of journalistic online-content by using automated tools”, SCM, Vol. 3 No. 2, pp. 205-227, doi: 10.5771/2192-4007-2014-2-180.

Holtz-Bacha, C., Langer, A.I. and Merkle, S. (2014), “The personalization of politics in comparative perspective: campaign coverage in Germany and the United Kingdom”, European Journal of Communication, Vol. 29 No. 2, pp. 153-170, doi: 10.1177/0267323113516727.

Horton, D. and Wohl, R.R. (1956), “Mass communication and para-social interaction; Observations on intimacy at a distance”, Psychiatry, Vol. 19, pp. 215-229, doi: 10.1080/00332747.1956.11023049.

Hwang, S. (2012), “The strategic use of Twitter to manage personal public relations”, Public Relations Review, Vol. 38, pp. 159-161, doi: 10.1016/j.pubrev.2011.12.004.

Ji, Y.G., Li, C., North, M. and Liu, J. (2017), “Staking reputation on stakeholders: how does stakeholders' Facebook engagement help or ruin a company's reputation?”, Public Relations Review, Vol. 43 No. 1, pp. 201-210, doi: 10.1016/j.pubrev.2016.12.004.

Jünger, J. and Keyling, T. (2019), “Facepager. An application for generic data retrieval through APIs”, available at: https://github.com/strohne/Facepager.

Krippendorff, K. (1971), “Reliability of recording instructions: multivariate agreement for nominal data”, Behavioral Science, Vol. 16 No. 3, pp. 228-235, doi: 10.1002/bs.3830160305.

Krippendorff, K. (2012), Content Analysis: An Introduction to its Methodology, Sage, Thousand Oaks, CA.

Kruikemeier, S. (2014), “How political candidates use Twitter and the impact on votes”, Computers in Human Behavior, Vol. 34, pp. 131-139, doi: 10.1016/j.chb.2014.01.025.

Lawrence, R.G., McGregor, S.C., Cardona, A. and Mourão, R.R. (2016), “Personalization and gender: 2014 gubernatorial candidates on social media”, in Hendricks, J.A. and Schill, D. (Eds), Communication and Midterm Elections, Palgrave Macmillan, pp. 191-206.

Lee, E.-J. and Oh, S.Y. (2012), “To personalize or depersonalize? When and how politicians' personalized tweets affect the public's reactions”, Journal of Communication, Vol. 62 No. 6, pp. 932-949, doi: 10.1111/j.1460-2466.2012.01681.x.

Lingner Consulting (2019), “Facebook B2B ranking”, available at: https://lingner.com/b2branking/ranking_b2b.php?sort=pta_index.

Lock, I. (2019), “Explicating communicative organization-stakeholder relationships in the digital age: a systematic review and research Agenda”, Public Relations Review, Vol. 45, pp. 1-13, doi: 10.1016/j.pubrev.2019.101829.

Loretta, U. (2002), “The CEO's, how to ‘guide to crisis communication’”, Strategy and Leadership, Vol. 30 No. 2, pp. 21-24.

Macnamara, J. and Zerfaß, A. (2012), “Social media communication in organizations: the challenges of balancing openness, strategy and management”, International Journal of Strategic Communication, Vol. 6 No. 4, pp. 287-308, doi: 10.1080/1553118X.2012.711402.

Manetti, G., Bellucci, M. and Bagnoli, L. (2016), “Stakeholder engagement and public information through social media: a study of Canadian and American public transportation agencies”, The American Review of Public Administration, Vol. 47 No. 8, pp. 991-1009, doi: 10.1177/0275074016649260.

McAllister, I. (2007), “The personalization of politics”, in Dalton, R.J. and Klingemann, H.-D. (Eds), The Oxford Handbook of Political Behavior, Oxford University Press, Oxford, pp. 571-589.

McGregor, S.C. (2018), “Personalization, social media, and voting: effects of candidate self-personalization on vote intention”, New Media and Society, Vol. 20 No. 3, pp. 1139-1160, doi: 10.1177/1461444816686103.

Meeks, L. (2012), “Is she ‘man enough’? Women candidates, executive political offices, and news coverage”, Journal of Communication, Vol. 62 No. 1, pp. 175-193, doi: 10.1111/j.1460-2466.2011.01621.x.

Men, L.R. and Tsai, W.H.S. (2016), “Public engagement with CEOs on social media: motivations and relational outcomes”, Public Relations Review, Vol. 42 No. 5, pp. 932-942, doi: 10.1016/j.pubrev.2016.08.001.

Metz, M., Kruikemeier, S. and Lecheler, S. (2019), “Personalization of politics on Facebook: examining the content and effects of professional, emotional and private self-personalization”, Information, Communication and Society, Vol. 23 No. 10, pp. 1-18, doi: 10.1080/1369118X.2019.1581244.

Mhamdi, C., Al-Emran, M. and Salloum, S.A. (2018), “Text mining and analytics: a case study from news channels posts on facebook”, in Shaalan, K., Hassanien, A. and Tolba, F. (Eds), Intelligent Natural Language Processing: Trends and Applications, Springer, Cham, doi: 10.1007/978-3-319-67056-0_19.

Motion, J. (1999), “Personal public relation: identity as a public relations commodity”, Public Relations Review, Vol. 4, pp. 465-479.

Park, D.J. and Berger, B.K. (2004), “The presentation of CEOs in the press, 1990-2000: increasing salience, positive valence, and a focus on competency and personal dimensions of image”, Journal of Public Relations Research, Vol. 16 No. 1, pp. 93-125, doi: 10.1207/s1532754xjprr1601_4.

Pollach, I. and Kerbler, E. (2011), “Appearing competent. A study of impression management in U.S. and European CEO profiles”, Journal of Business Communication, Vol. 48 No. 4, pp. 355-372, doi: 10.1177/0021943611414687.

Pruysers, S., Cross, W.P. and Katz, R.S. (2018). “Personalism, personalization and party politics”, in Cross, W.P., Katz, R.S. and Pruysers, S. (Eds), The Personalization of Democrativ Politics and the Challenge for Political Parties, ECPR Press/Rowman and Littlefield, London, pp. 1-17.

Rahat, G. and Sheafer, T. (2007), “The personalization(s) of politics: Israel, 1949-2003”, Political Communication, Vol. 24 No. 1, pp. 65-80, doi: 10.1080/10584600601128739.

Ranft, A.L., Zinko, R., Ferris, G.R. and Buckley, R. (2006), “Marketing the image of management: the costs and benefits of CEO reputation”, Organizational Dynamics, Vol. 35 No. 3, pp. 279-290, doi: 10.1016/j.orgdyn.2006.05.003.

Salinas, S. (2018), Zuckerberg on Cambridge Analytica: ‘We Have a Responsibility to Protect Your Data, and if We Can't Then We Don't Deserve to Serve You’, CNBC, available at: https://www.cnbc.com/2018/03/21/zuckerberg-statement-on-cambridge-analytica.html.

Sauerhaft, S. and Atkins, C. (1989), Image Wars: Protecting Your Company when There's No Place to Hide, Wiley, New York, NY.

Shanahan, T., Tran, T.P. and Taylor, E.C. (2019), “Getting to know you: social media personalization as a means of enhancing brand loyalty and perceived quality”, Journal of Retailing and Consumer Services, Vol. 47, pp. 57-65, doi: 10.1016/j.jretconser.2018.10.007.

Sohn, Y., Weaver Lariscy, R. and Tinkham, S.F. (2009), “The impact of CEO reputation: negative news and economic decisions”, International Journal of Strategic Communication, Vol. 3 No. 1, pp. 1-18, doi: 10.1080/15531180802606596.

Solis, B. and Breakenbridge, D. (2009), Putting the Public Back in Public Relations: How Social Media Is Reinventing the Aging Business or PR, FT Press/Pearson Education, Upper Saddle River, NJ.

Staab, J.F. (1990), “The role of news factors in news selection: a theoretical reconsideration”, European Journal of Communication, Vol. 5, pp. 423-443, doi: 10.1177/0267323190005004003.

Strömbäck, J. (2008), “Four phases of mediatization: an analysis of the mediatization of politics”, The International Journal of Press/Politics, Vol. 13, pp. 228-246, doi: 10.1177/1940161208319097.

Talanow, M. (2015), Corporate Media Relations und Personalisierung: Eine empirische Analyse der Medienarbeit ausgewählter DAX30-Unternehmen, Springer, Wiesbaden.

Teeny, J.D., Siev, J.J., Briñol, P. and Petty, R.E. (2021), “A review and conceptual framework for understanding personalized matching effects in persuasion”, Journal of Consumer Psychology, Vol. 31 No. 2, pp. 382-414, doi: 10.1002/jcpy.1198.

Thummes, K. and Malik, M. (2015), “Beteiligung und Dialog durch Facebook? Theoretische Überlegungen und empirische Befunde zur Nutzung von Facebook-Fanseiten als Dialogplattform in der Marken-PR”, in Hoffjann, O. and Pleil, T. (Eds), Strategische Onlinekommunikation: Theoretische Konzepte und empirische Befunde, Springer, Wiesbaden, pp. 105-130.

Treadway, D.C., Adams, G.L., Ranft, A.L. and Ferris, G.R. (2009), “A meso-level conceptualization of CEO celebrity effectiveness”, Leadership Quarterly, Vol. 20 No. 4, pp. 554-570, doi: 10.1016/j.leaqua.2009.04.008.

Valentini, C. (2015), “Is using social media ‘good’ for the public relations profession? A critical reflection”, Public Relations Review, Vol. 41 No. 2, pp. 170-177, doi: 10.1016/j.pubrev.2014.11.009.

Van Aelst, P., Sheafer, T. and Stanyer, J. (2012), “The personalization of mediated political communication: a review of concepts, operationalizations and key findings”, Journalism, Vol. 13 No. 2, pp. 203-220, doi: 10.1177/1464884911427802.

Wayne, M. and Murray, C. (2009), “UK Television news: monopoly politics and cynical populism”, Television and New Media, Vol. 10 No. 5, pp. 416-433, doi: 10.1177/1527476409334020.

We Are Social, Hootsuite, and DataReportal (2021), “Ranking der größten sozialen Netzwerke und Messenger nach der Anzahl der monatlich aktiven Nutzer (MAU) im Januar 2021 (in Millionen)”, available at: https://de.statista.com/statistik/daten/studie/181086/umfrage/die-weltweit-groessten-social-networks-nach-anzahl-der-user.

Wehmeier, S. and Winkler, P. (2012), “Personalisierung und Storytelling in der Online-Kommunikation”, in Zerfaß, A. and Pleil, T. (Eds), Handbuch Online-PR: Strategische Kommunikation in Internet und Social Web, UVK, Konstanz, pp. 383-394.

Westphal, J.D. and Deephouse, D.L. (2011), “Avoiding bad press: interpersonal influence in relations between CEOs and journalists and the consequences for press reporting about firms and their leaderships”, Organization Science, Vol. 22 No. 4, pp. 1061-1086, doi: 10.1287/orsc.1100.0563.

Wong, J.C. (2018), “Mark Zuckerberg apologises for Facebook's ‘mistakes’ over Cambridge Analytica”, available at: https://www.theguardian.com/technology/2018/mar/21/mark-zuckerberg-response-facebook-cambridge-analytica.

Wright, D.K. and Hinson, M.D. (2011), “A three-year longitudinal analysis of social and emerging media use in public relations practice”, Public Relations Journal, Vol. 5 No. 3, pp. 1-32.

Wright, D.K. and Hinson, M.D. (2015), “Examining social and emerging media use in public relations practice: a ten-year longitudinal analysis”, Public Relations Journal, Vol. 9 No. 2, pp. 1-26.

Zerfaß, A., Verčič, D. and Wiesenberg, M. (2016), “Managing CEO communication and positioning: a cross-national study among corporate communication leaders”, Journal of Communication Management, Vol. 20 No. 1, pp. 37-55, doi: 10.1108/JCOM-11-2014-0066.

Zhang, Y., Dong, C. and Cheng, Y. (2022), “How do nonprofit organizations (NPOs) effectively engage with the public on social media? Examining the effects of interactivity and emotion on Twitter”, Internet Research, ahead-of-print. doi: 10.1108/INTR-05-2021-0290.

Corresponding author

Nora Denner can be contacted at: nora.denner@uni-mainz.de

Related articles