Keywords

1 Introduction

Corporate communication is an interdisciplinary concept that is approached from marketing, public relations (PR), organizational communication, and linguistic perspectives. In marketing, the role of corporate communication for loyal relationships with stakeholders is central. In PR, it is the managing of dialogic relations with an organization’s publics. For organizational communication, the social co-creation of the process of organizing is in focus (Mazzei 2014). In linguistics, business communication addresses the pragmatic dimension of language, often taking an (inter-)cultural perspective (Fuoli 2018). Regarding marketing and PR, corporate communication is often regarded as strategic communication (Zerfass et al. 2018). This contribution will largely focus on content analyses from a corporate communication perspective.

One central capacity of corporate communication is supporting to build intangible resources that reduce transaction costs for organizations and are key for an organization’s long-term competitive advantage (Barney 1991, 2001). These intangible resources include concepts such as knowledge, trust, loyalty, reputation, responsibility, or identity (Cornelissen 2013; Fuoli 2018; Mazzei 2014). One major theme in corporate communication research is the role of corporate communication for explaining stakeholder attitudes and behavior, according to Zerfass and Viertmann’s (2017) meta study of research into corporate communication. Beyond the capacity of building intangible resources, corporate communication also enables operations, adjusts strategy, and ensures flexibility of firms (Zerfass and Viertmann 2017). That is, corporate communication supports strategic alignment, market positioning, innovation, or organizational change. These themes can become research topics in content analyses of corporate communication material. 

As organizations require monetary and human resources from their environment as well as seek sales markets, organizations also acquire social support, i.e., legitimacy from their environment (Palazzo and Scherer 2006; Suddaby et al. 2017). In this institutional perspective, organizations employ strategic communication to pursue their goals and to manage their legitimacy (Suchman 1995). Against this background, corporate social responsibility (CSR) has become a focus in corporate communication research. CSR is often conceptualized as a company’s capacity to conform to business, legal, ethical, and philanthropic standards (Carroll 1991, 2016). Operating profitably (business) and obeying the law (legal) comprise rather essential requirements, while to do what is just and fair (ethical) and to be a good citizen (philanthropic) is less obligatory but desired by society (Carroll 1991). Research in CSR studies has focused on perception, impact and promotion; image and reputation; performance; and generally the rhetoric of organizations (Ellerup Nielsen and Thomsen 2018). In CSR research, content analysis is used to assess the performance (Gunawan and Abadi 2017) and the credibility of CSR reports (Lock and Seele 2016), for instance.

Content analyses have gained popularity in corporate communication as well as CSR research since the availability of computer-aided text analysis (CATA) (Duriau et al. 2007; Short et al. 2010), a label used in organizational research. Cornelissen (2013) claims that most research into corporate communication uses surveys, e.g., for stakeholder evaluations of company reputation, while content analyses are often part in case studies alongside interviews and observations. Yet, content analyses are indispensable to identify “who says what,” in the terms of Lasswell (1948), and thus represent a classical method for analyzing corporate documents. Content analysis of annual reports “can be of real usefulness for understanding some issues of corporate strategy,” argues Bowman (1984, p. 70), because it can not only measure complex organizational constructs, including corporate culture, risk affinity, or CSR. Content analysis can also “show relationships [between constructs] which are otherwise difficult to obtain and which can be tested for validity” (ibid., p. 61). Similarly, Duriau et al. (2007, p. 6) emphasize that content analyses can reliably access “values, intentions, attitudes, and cognitions” that have manifested in corporate messages. Hence, content analyses are used in organizational studies to reveal attitudinal or cognitive aspects of organizations and organizing. In comparison to responsive methods such as surveys or interviews, Harris (2001, p. 195) suggests that content analyses serve as a “reality check” of managerial decision making.

The remainder of the article aims at providing an overview about the diversity of research themes and designs of content analyses in corporate communication.

2 Frequent Research Themes

To describe frequent research themes, I refer to two meta studies: Duriau et al. (2007) and Zerfass and Viertmann (2017). Duriau et al. (2007) conduct a meta study of content analyses in the field of organization studies between 1980 and 2005. Their analysis suggests that research into corporate communication differs regarding studies of corporate communication and studies using corporate communication material for researching corporate phenomena. They identify two major research themes that most frequently apply content analyses: (a) strategic management issues that address topics such as impression management, corporate reputation, or strategy reformulation and (b) the issue of managerial cognition involving corporate values and culture, sensemaking, blame attribution, or managerial attention in crises (Duriau et al. 2007).

Zerfass and Viertmann (2017, p. 69) analyze publications from the fields of “corporate communication, organizational communication, public relations, marketing, and strategic management,” independent from the application of content analyses. They identify twelve central constructs of tangible and intangible outcomes of corporate communication that are studied, i.e., relationships, trust, legitimacy, thought leadership, innovation potential, crisis resilience, reputation, brands, corporate culture, publicity, customer preferences, and employee commitment (Zerfass and Viertmann 2017). Beyond surveying stakeholder groups, for example for assessing corporate reputation (Wartick 2016), some of these concepts can on principle be measured by analyzing the content of corporate communication material and user-generated content.

The following examples provide an impression of the variety of themes studied in corporate communication and may serve as starting point for further investigation into a specific area of interest. Interactivity dimensions of corporate websites are analyzed using content analysis (Ha and James 1998), addressing stakeholder relationships. In crisis communication research, content analysis is conducted to understand which  crisis response strategies are used in corporate messages and how news coverage as well as users respond, for instance on social media (Holladay 2010). Combining document analysis and interviews, Huang-Horowitz and Evans (2020) reveal how small companies communicate their organizational identity to gain legitimacy. Regarding leadership, content analyses can reveal the degree of courage expressed by executives and related news coverage (Harris 2001). Li et al. (2018) regard innovation potential as one dimension of corporate culture, along with integrity, quality, respect, and teamwork. They measure corporate culture using a machine learning (ML) approach on a corpus of earnings calls, in which public companies discuss their financial results addressing the investor and analyst communities. The sentiment of user-generated online product reviews indicates customer preferences (Jo and Oh 2011; Tirunillai and Tellis 2014). Concerning employee commitment, Bujaki et al. (2018) reveal impression management strategies of accounting firms addressing diversity‐sensitive employees. Regarding internal communication, Darics (2020) analyses instant message conversations between employees and shows that instant messages intend to achieve complex communication goals, including fostering informality and building team identity.

The themes of CSR messages are analysed for various industries in CSR reports (Landrum and Ohsowski 2018) or on social media platforms like Instagram (Kwon and Lee 2021). Moreover, Lock and Seele (2016) quantitatively analyze the credibility of CSR reports by measuring truth of statements, accuracy, completeness, standards used, and sincerity, and reveal that CSR reports can be considered as mediocrely credible. Hoffmann et al. (2018) discursively analyze Facebook’s CEO speech revealing it surrounds self-identity, constructs user identity and the relationship between Facebook and its users. As a final example, VanDyke and Tedesco (2016) analyze responsibility frames in green advertising over time, indicating that a habitat protection issue changes into energy efficiency.

3 Frequent Research Designs

Regarding research designs, corporate communication can represent the independent, dependent, or mediating variable. Regarding the independent variable, corporate communication messages represent an antecedent to explain attitudinal outcomes (trust and reputation in customers) as well as operational outcomes (e.g., economic results, stock market performance, speed of news product releases) (see Duriau et al. 2007; Zerfass and Viertmann 2017). Here, content analyses are used to evaluate corporate content material—but also content generated by customers or followers. Moreover, research into corporate communication addresses the relation between symbolic communication, which can be assessed with content analyses, and substantive corporate action (Seiffert et al. 2011), often comparing the content of CSR communication and action (Jong and van der Meer 2017; Perez-Batres et al. 2012; Schons and Steinmeier 2016; Wickert et al. 2016). Concerning the dependent variable, corporate communication content is regarded as a manifestation of internal processes such as managerial sensemaking or cognition. In this case, content analysis is used to deduce on such internal processes (see Duriau et al. 2007). —One central limitation for the deduction is intentional bias in corporate communication for specific stakeholder groups. For instance, annual reports include a bias toward the positive (Rutherford 2005) or dramatize ideas (Jameson 2000). Methodological responses to this challenge include using multiple data sources and richer databases, triangulation, and sophisticated methods that provide more accurate measurements (Duriau et al. 2007).—Corporate communication messages can also be conceptualized as a mediating variable between internal processes and organizational outcomes. For instance, Porcu et al. (2016) regard internal corporate communication as a mediator between corporate culture and operational outcomes, however, use a survey for data collection.

Methodologically, research designs employing content analysis follow qualitative, standardized manual, quantitative-computational approaches, or combinations thereof. Which design to follow depends on the availability of data sources for a research question at hand and the production contexts of the specific material to be analyzed (Steenkamp and Northcott 2007). For instance, studies into corporate communication addressing journalists as stakeholder group often compare corporate messages and news coverage using quantitative content analysis (e.g., Jonkman et al. 2020; Lischka et al. 2017; Nijkrake et al. 2015). Qualitative approaches aim at revealing organizational narratives, for instance regarding corporate responsibility (Haack et al. 2012), strategy change (Lischka 2019c), and legitimacy (van Leeuwen and Wodak 1999).

According to Duriau et al. (2007), primary data sources of corporate communication content analyses are annual reports, followed by proxy statements, trade magazines, publicly available corporate documents, mission statements, internal company documents, and notes from interviews or answers to open-ended survey questions. Moreover, news coverage (e.g., Seiffert et al. 2011; Strycharz et al. 2017), CSR reports (e.g., Lock and Seele 2016), CEO speech (e.g., Beelitz and Merkl-Davies 2012; Hoffmann et al. 2018), social media communication and engagement (e.g., Abitbol and Lee 2017; Choy and Wu 2018; Kim et al. 2014; Macnamara and Zerfass 2012), corporate blogs (e.g., Catalano 2007; Colton and Poploski 2018), advertising (e.g., VanDyke and Tedesco 2016), and text messages (Darics 2020) represent data sources. Researchers from linguistics often build a corpus based on one corporate material genre from multiple organizations, for instance, a corpus of annual reports (Fuoli 2018; Rutherford 2005) or CRS reports (Yu and Bondi 2017). Researchers from other disciplines may also create corpora but without labelling their approach as a corpus approach (e.g., Seiffert et al. 2011).

For computational analyses, researchers have developed dictionaries, for instance, a finance- and accounting-specific dictionary in English (Loughran and McDonald 2011, 2015) and German (Bannier et al. 2019), for environmental sustainability (Deng et al. 2017), and for vagueness in corporate communication (Guo et al. 2017). Also more general dictionaries such as Linguistic Inquiry and Word Count (LIWC) are applied as in Merkl‐Davies et al. (2011) and Lee et al. (2020).

4 Trends

There is a variety of methodological trends regarding content analyses of corporate communication. Research combines content analysis with other data collection methods, applies machine learning (ML) and (deep) natural language processing (NLP) techniques, and extends data capacity, contexts, and materiality. The following list provides recent exemplary studies for trends in computational methods, design, sampling, and material, with methods of computational content analysis representing a comparatively large evolving field.

  • ML and (deep) NLP

    NLP is a computational method for analyzing naturally occurring human language by building statistical models of language, which has been applied in linguistics (Manning and Schütze 1999). With ML, algorithms are developed that should improve through training data and can be combined with human coding in supervised or semi-supervised settings. In deep ML, artificial neural networks are used for training (Deng and Liu 2018). Deep NLP can therefore use “both sentence structure and context of the text to provide a deeper understanding of the language” (Lee et al. 2020).

    • Combining human coding and ML (Park et al. 2019),

    • Applying semi-supervised ML (van Zoonen and van der Meer 2016)

    • Applying topic modeling, which is unsupervised as it uses statistical associations of words in a text to generate topics without dictionaries or interpretive rules (Hannigan et al. 2019; Jaworksa and Nanda 2016; Kobayashi et al. 2018; Schmiedel et al. 2018)

    • Specific dictionary development for corporate communication issues (Deng et al. 2017; Guo et al. 2017)

    • Comparing deep NLP (IBM Watson Explorer) with dictionary approaches and human coding to detect the level of charisma in leadership speeches (Lee et al. 2020)

  • Triangulation: Combining content analyses with surveys (Dudenhausen et al. 2020), combining qualitative and quantitative approaches (Jaworksa and Nanda 2016)

  • Comparative designs: Comparative approaches within Western countries (Köhler and Zerfass 2019; Yu and Bondi 2019; Yuan 2019), and beyond, such as in Asia (Bondi and Yu 2015) and in Americana (Loureiro and Gomes 2016)

  • Non-Western context: CSR communication in India (Jain and Moya 2016), in restrictive systems such as China (Zhang et al. 2017) and Russia (Sorokin et al. 2019)

  • Visuality: Analyzing visual rhetoric in corporate reports (Goransson and Fagerholm 2018; Greenwood et al. 2018; Ruggiero 2020) and multimodal (textual and visual) content analysis, for instance to account for the multimodality of corporate websites (Höllerer et al. 2019)

5 Research Desiderata

The trend on employing large collections of texts combined with ML, such as applying topic modelling algorithms, requires advances in methodological standards, for instance regarding procedures such as structural topic models (Roberts et al. 2019), validity comparisons across content analysis methods (van Atteveldt et al. 2021), and quality criteria for automated content analyses (Laugwitz 2021). With the ability to analyze extensive data sets, complex research designs may become better attainable. For instance, the various agents and processes that constitute organizational legitimacy as proposed in Bitektine and Haack (2015) may be tackled. In doing so, qualitative approaches, for instance to understand the dynamics of corporate narratives as in Jaworksa and Nanda (2016), can be fruitfully combined with computational analyses.

Regarding research objects, Zerfass and Viertmann (2017) suggest that the capacity of corporate communication should be assessed across various types and sizes of organizations (e.g., start-ups, small-and-medium enterprises, large corporations, non-profit organizations), across stakeholder groups (e.g., customers, employees, investors, and journalists), in various situational contexts (e.g., product launches, crises, and mergers), and industries. While organizations in any industry can become objects of analysis for corporate communication research, scholars in the field of communication and journalism studies may be especially interested in communication of organizations involved in public communication such a media organizations (Bachmann 2016; Lischka 2019b; Siegert and Hangartner 2017) or social media platforms (Gillespie 2010; Iosifidis and Nicoli 2020; Lischka 2019a). Against the background of globally acting organizations having the power, and sometimes the obligation, to assume political roles on a global scale (Scherer and Palazzo 2011), future research should focus on such global corporations to understand how they communicate their political stances and roles. There is additional need for comparative studies and, in particular, analyses of non-Western countries.

Moreover, the interaction of communication by multiple organizations can deliver relevant insights. Suchman (1995, p. 592) argues, orchestrated communication by a group of companies, such as social media platforms and search engines, can become a powerful “collective evangelism” when occupying an issue. From an institutional perspective, analyzing potentially orchestrated communication of globally acting organizations can show how new institutions in societies are negotiated.

Lastly, there has been a normative turn in management research towards the “grand” challenges of global societies, including poverty, good economic growth, health disparities, climate change, and sustainability (United Nations n.d.). Against the background that organizations should build value for societies, management researchers wish to contribute to how organizations can help to address and solve these grand problems (George et al. 2016). Corporate communication researchers, especially those focusing on CSR, are uniquely positioned to addressing grand challenges from a corporate communication perspective. Content analyses using material from companies as well as produced by various stakeholder groups can reveal links between communication and corporate goals as well as societal challenges on a broader scale.