Expressing and Responding to Customer (Dis)satisfaction Online: New Insights From Discourse and Linguistic Approaches

Abstract

In the current era of digitalization, customers are routinely invited to express their (dis)satisfaction with a product or a service and to provide recommendations for other prospective customers by writing reviews on a variety of online social media platforms. Such forms of electronic word-of-mouth have been found to strongly influence other consumers’ purchase decisions. In the case of negative reviews, the negativity expressed in a particular comment can spread to the whole community, which can damage a company’s reputation and profits. In an attempt to take consumer feedback into account, companies engage in “webcare.” This type of online service encounter has been defined by van Noort and Willemsen as “the act of engaging in online interactions with (complaining) consumers, by actively searching the web to address consumer feedback (e.g., questions, concerns, and complaints).” Following-up on these developments, scholars have started to research the communicative strategies used by companies to address consumer feedback and those used by (dis)satisfied customers to voice their (dis)satisfaction from the perspective of discourse analysis and linguistic pragmatics, paying attention to their linguistic realizations and their interactional dynamics. The aim of this Special Issue is to further expand our knowledge on the discourse-pragmatic strategies used in the interaction of (dis)satisfied customers and companies online, and on how these different strategies influence other prospective customers’ perceptions, ultimately impacting their purchase decisions. In doing so, it positions itself at the crossroads of linguistics, communication, and business studies.

In the current era of digitalization, customers are routinely invited to express their (dis)satisfaction with a product or a service and to provide recommendations for other prospective customers by writing reviews on a variety of online social media platforms. Such forms of electronic word-of-mouth have been found to strongly influence other consumers’ purchase decisions. In the case of negative reviews, the negativity expressed in a particular comment can spread to the whole community, which can damage a company’s reputation and profits. In an attempt to take consumer feedback into account, companies engage in “webcare.” This type of online service encounter has been defined by van Noort and Willemsen (2012, p. 133) as “the act of engaging in online interactions with (complaining) consumers, by actively searching the web to address consumer feedback (e.g., questions, concerns and complaints).” Following-up on these developments, scholars have started to research the communicative strategies used by companies to address consumer feedback and those used by (dis)satisfied customers to voice their (dis)satisfaction from the perspective of discourse analysis and linguistic pragmatics, paying attention to their linguistic realizations and their interactional dynamics. The aim of this Special Issue is to further expand our knowledge on the discourse-pragmatic strategies used in the interaction of (dis)satisfied customers and companies online, and on how these different strategies influence other prospective customers’ perceptions, ultimately impacting their purchase decisions. In doing so, it positions itself at the crossroads of linguistics, communication, and business studies.

The idea behind the present collection of articles stems from two conference panels. The first one was devoted to the expression and perception of customer dissatisfaction online. The editors of this Special Issue, Nicolas Ruytenbeek and Sofie Decock, organized it as part of the International Pragmatics Association conference organized by the Zurich University of Applied Sciences (ZHAW), which took place online in June–July 2021. The second panel, organized by Matteo Fuoli and Sofie Decock, was devoted to recent trends in webcare studies and took place during the online ABC regional conference organized by the WU Vienna in August 2021. Moreover, the present Special Issue builds on a Special Issue entitled Recent trends in webcare: a closer look at discursive practices, edited by Sofie Decock, Bernard De Clerck, Geert Jacobs, and Rebecca Van Herck in the journal Discourse, Context and Media (2022). As Decock (2022) states in her introduction to this Special Issue, its main goal was to offer a linguistic-oriented approach to online service encounters. The present collection of articles expands on the 2022 Special Issue edited by Decock, De Clerck, Jacobs, and Van Herck, as it seeks to focus not only on the language of webcare, but also on the online discourse of customer (dis)satisfaction and its interactional dynamics with webcare. Even though the contributions to the current Special Issue are not the first ones to address these topics from a linguistic perspective, taken together they highlight the relevance of such approaches in the field of business communication and stimulate new studies along these lines. Several studies have already been conducted that concentrate on language use and discourse in online service encounters (e.g., Bou-Franch & Garcés-Conejos Blitvich, 2019; Cenni & Goethals, 2017; De la O Hernández-López & Amaya, 2015; Depraetere et al., 2021; Fuoli et al., 2021; Lillqvist et al., 2016; Orthaber, 2019; Van Herck et al., 2020, 2023; Vásquez, 2011, 2014; Widdershoven et al., 2021). Some of these studies combined rigorous linguistic analysis with experiments to evaluate the effectiveness of particular linguistic strategies (De Clerck et al., 2019; Decock et al., 2020; Fuoli et al., 2021; Van Herck et al., 2023). Against this background, our aim is to build on this previous work by paying attention to some under-researched linguistic features and to methodological operationalizations of key theoretical concepts in the areas of customer feedback and webcare, such as emotional contagion and conversational human voice (CHV). More specifically, the Special Issue includes papers which zoom in on:

• specific linguistic features of the discourse of customer feedback, notably the use of adversative connectives (ACs) (i.e., constituents such as but, although, however, and even though) (Baker & Hashimoto, this collection) and the use of the first-person perspective (FPP) through first-person pronouns from a cross-linguistic perspective (Fastrich, this collection);

• how sentiment polarity as a discursive feature of customer feedback relates to star ratings (Baker & Hashimoto, this collection), or how it affects emotional contagion (Widdershoven, Pluymaekers, Zourrig, Sinclair, & Bloemer, this collection);

• the discourse of webcare responses adopting various approaches, including a move analysis combined with a cross-linguistic approach (Van Herck & Vangehuchten, this collection), an analysis of the use of CHV in these responses (Van Herck & Vangehuchten; Holmgreen, this collection), and a corpus linguistic approach (Lutzky, this collection);

• interactional patterns between customer and organizational discourse in online service encounters by exploring how companies practice webcare and how customers react to and perceive these webcare responses (Lutzky; Holmgreen, this collection).

As this brief overview shows, the studies presented in this Issue are complementary in the sense that, while some focus on the unfolding of (full) interactions between consumers and companies (Holmgreen; Lutzky) or among consumers only (Widdershoven et al.), others concentrate instead on the individual genres of either reviews (Baker & Hashimoto; Fastrich) or organizational complaint responses (Van Herck & Vangehuchten). Moreover, while all studies work with authentic, user-generated messages pertaining to digital communication, the individual articles cover a variety of online platforms and social media (Twitter, Facebook, Yelp.com, Booking.com). Last but not least, the complementarity and diversity of the contributions in this Issue also become visible when looking at the methodological and theoretical approaches that are adopted: the studies are informed by different research traditions (e.g., discourse analysis, pragmatics, corpus linguistics, service research) and adopt different theoretical perspectives (e.g., emotional contagion, customer satisfaction, organizational reputation, CHV).

In this introduction to our Special Issue, we first propose a short state-of-the-art on the role of language in scholarly work on customer feedback and webcare, followed by a detailed overview of the papers included in our Special Issue, integrating their findings and highlighting their contributions to the study of online customer feedback and webcare. We also contextualize the individual contributions in the light of theoretical and methodological developments in the linguistic study of consumer feedback and webcare, teasing out their innovative qualities as well as the areas of similarity and divergence between them. This will allow us to identify some managerial implications and some perspectives for future research resulting from this Special Issue.

The Language of Customer Feedback and Webcare From a Cross-Disciplinary Perspective

Online consumer feedback and webcare have attracted the interest of scholars from various fields. For instance, the question of how to approach consumer feedback has been addressed by marketing and communications scholars. However, as Decock (2022) points out, these scholars tend to study such subjects without paying specific attention to the language in service encounters. They are primarily interested in the effects of managers’ responses on customer outcomes, drawing on attribution theory (Kelley, 1967) and on equity and justice theory (Orsingher et al., 2010). In fact, the lack of attention devoted to the language used in service interactions was noted and criticized by marketing scholars themselves (e.g., Carnevale et al., 2017; Holmqvist & Grönroos, 2012) and recent efforts have been made to address this gap (e.g., Holmqvist et al.’s (2017) Special Issue on language use in services).

Researchers in the fields of communication and public relations have stressed the usefulness for organizations to use a conversational human voice (CHV) when addressing stakeholders (Kelleher, 2009). CHV is defined as “an engaging and natural style of organizational communication as perceived by an organization’s publics based on interactions between individuals in the organization and individuals in the public” (Kelleher, 2009, p. 117). Previous research in the context of webcare indicates that CHV improves customer satisfaction (e.g., Willemsen et al., 2013) and gives rise to higher purchase intentions and more favorable attitudes toward the companies (e.g., van Noort & Willemsen, 2012). However, it has long been highly unclear what the linguistic properties of CHV are and what the strategy of using CHV in communication exactly comprises. The notion of CHV has therefore not been immune to criticism from scholars (Creelman, 2022; Gretry et al., 2017; Zhang & Vásquez, 2014). In light of this criticism, recent research on companies’ use of CHV pays more attention to its linguistic features (e.g., Liebrecht et al., 2021). However, systematic linguistic and discursive analyses of the use of CHV in webcare interactions are still rare.

Online customer feedback has also been considered extensively by scholars in the field of language technology and machine learning. While studies of this kind generally pay attention to the linguistic features of consumer feedback, they do so with specific applied goals, such as the intention of quickly and automatically classifying reviews according to sentiment to monitor feedback (e.g., as a company’s crisis management strategy) and the automatic detection of fake reviews. Machine learning techniques are also applied to webcare, specifically with the goal, for instance, to develop AI-assisted webcare as a means to help humans to write webcare responses. These machine learning approaches also facilitate certain analyses: by automatically coding sentiment, for instance, it becomes possible to see how sentiment relates to the customer outcome or perceived usefulness of a review (e.g., Felbermayr & Nanopoulos, 2016; Ren & Quan, 2012).

While the present Special Issue is, to some extent, informed by these research traditions, it is different in the sense that it offers rigorous linguistic and discursive analyses of consumer feedback and webcare, combining linguistic methodologies and frameworks.

The Discourse of Online Customer Feedback

In the present collection, Fastrich proposes a comparative study of the use of the first-person perspective (FPP) in English- and German-language hotel reviews on Booking.com. In doing so, she seeks to shed light on the widely documented linguapragmatic contrast of English speakers as more personal, involved and person-oriented, and German speakers as more impersonal, detached and content-oriented (House, 2006). On top of adopting this contrastive pragmatic perspective, Fastrich also aims to interpret her results on FPP in the light of possible cognitive differences between English and German reviewers in how they tend to experience hotel stays. The results of Fastrich’s analyses indicate that English-speaking reviewers write more personalized reviews, which is in line with previous research showing that English speakers have a more person-oriented communication style in comparison with German speakers. From a cognitive perspective, this finding suggests that German speakers have a more detached construal of their hotel experience compared to English speakers. That being said, the analysis of FPP saturation (the number of instances of first-person pronoun use within a review over its word length) showed that, when FPP is present in German-language reviews, its use is similar to that in English-language reviews. This similarity may be the result of the emergence of a global genre of consumer reviews under the influence of English as a Lingua Franca. Regarding the use of FPP in different review components, Fastrich finds that, in both languages, reviewers frequently use FPP to refer to and narrate the negative aspects of the hotel experience, with English reviewers using FPP more than German reviewers when telling their complaint stories. Overall, the results of this study reveal an interesting interplay in online reviews between the dimensions of linguaculture, genre and cognition through the analysis of FPP.

Similar to Fastrich, Baker and Hashimoto (this collection) are interested in specific linguistic features of reviews. They focus on adversary connective (AC) clauses such as Drinks were good as usual however the staff’s service needs improvement, where the constituents preceding and following the AC however are of opposite valence. AC clauses are interesting linguistic constructions to study in the context of customer feedback as a type of evaluative discourse because they convey (un)met expectations, with the second constituent always carrying most weight. In a first research question, Baker & Hashimoto investigate the relationship between the negative versus positive valence of the second constituent of ACs (which is negative in the staff’s service needs improvement) and the number of stars assigned to restaurant reviews. To do so, they use a large dataset of reviews that form a subset of Keller and Kostromitina’s (2020) corpus of Yelp.com restaurant reviews. They find that ACs occur in two out of three reviews on average, indicating that a majority of reviewers explicitly communicate (un)met expectations and thus (dis)satisfaction with the restaurants they visit. Their results show that, while the polarity of ACs second constituents is well balanced in reviews with average star ratings, negative second constituents largely outnumber positive ones in reviews with a low number of stars, and vice versa in reviews with a high number of stars. In line with these findings, they observe a statistically significant association between star rating and positive second constituents, and a non-significant association between star rating and negative second constituents. Moreover, through an additional thematic concordance analysis using the same corpus, Baker and Hashimoto describe patterns between star ratings and the content of second constituents, which they divided into food-, service-, and experience-related topics. This analysis shows, for instance, that although service and food are important to restaurant visitors at all rating levels, the more visitors are satisfied with service, the more they pay attention to food quality.

Another study that devotes considerable attention to sentiment polarity in online consumer feedback is Widdershoven, Pluymaekers, Zourrig, Sinclair, and Bloemer (this collection), who address emotional contagion on social media. Emotional contagion is “a process in which a person or group influences the emotions or behavior of another person or group through the conscious or unconscious induction of emotion states and behavioral attitudes” (Schoenewolf, 1990, p. 50). Emotional contagion can be positive, that is, resulting from other people’s positive emotions, or negative, that is, induced by the exposure to other people’s negative emotions (Hatfield et al., 1994). This phenomenon has been demonstrated in online contexts such as social media firestorms (Berger, 2013; Ott & Theunissen, 2015) and consumer reviews (Herrando et al., 2022). In this context, emotional contagion would entail that social media users who are confronted with a message in which a fellow consumer expresses a negative sentiment about an organization, will adopt the negative emotion expressed in that message and may decide to share the original message or to create a new one expressing the same emotion (Widdershoven, 2019). This allows the negative emotion to “go viral” and affect more and more users as it spreads across different networks online (Berger & Milkman, 2012). Using an original experimental design, Widdershoven et al. (this collection) investigate the effect of messages that go against the dominant negative sentiment on emotional contagion in tweets from fellow users. They find that non-negative messages can elicit emotional contagion, thus temporarily dismissing the prevalent negative emotions. These results suggest that the negativity bias, according to which people pay more attention to negative versus positive/neutral stimuli, does not tell the whole story on emotional contagion online. This finding also provides support for the view that emotional contagion can also be caused by a conscious decision to adopt someone else’ emotion if that emotion is deemed appropriate.

Also in this Special Issue, Lutzky investigates customer feedback both prior to and in response to webcare, using tweets from (dis)satisfied travelers addressed to the airline company Ryanair. More specifically, she analyzes the linguistic features and communicative strategies used by customers when addressing Ryanair as well as when reacting to Ryanair’s webcare responses. Similar to Baker & Hashimoto, Lutzky analyzes recurrent topics or reasons of (dis)satisfaction mentioned by customers, but she does so by conducting analyses of keywords (words that appear unusually frequently in a corpus) followed by the study of frequently occurring word combinations in the form of clusters (recurrent sequences of 3–5 words) and collocations (words that frequently co-occur). In doing so, she finds that customers often address Ryanair because they lack information, which makes them request the company to provide an explanation or voice a complaint about the inconvenience caused by this absence of information. Interestingly, customers’ answers to the airline’s responses reflect the same concern about information that should be but is still not offered to them, and they also provide insights into customers’ attitudes toward Ryanair’s responses, revealing, for instance, their dislike of generic replies.

Webcare Discourse

In the present Special Issue, Van Herck and Vangehuchten zoom in on organizational response emails to complaints as a specific type of webcare. They do so by combining a cross-cultural genre analysis with a comparative linguistic analysis of the use of CHV in these emails. The results of the contrastive genre analysis of English- and Spanish-language emails indicate that while both data sets display a similar discourse structure in terms of move frequency, pointing toward a standardization of genre conventions across linguacultural borders (see also Fastrich, this collection), they do differ in their distribution of interpersonal and business-oriented moves: the interpersonal submoves Empathy, Gratitude, and Apology are more frequent in the English emails and the business-oriented moves, for example, Contact reason, Marketing, and Future contact in the Spanish emails, suggesting a remaining influence of linguacultural pragmatic conventions. With regard to the analysis of CHV, Van Herck and Vangehuchten build on Van Noort et al.’s (2014), van Hooijdonk and Liebrecht’s (2021), and Liebrecht et al.’s (2021) attempts to operationalize CHV linguistically in terms of the three following strategies: message personalization, informal register and invitational rhetoric. One way for company employees of personalizing their message is to address customers using second person pronouns (you) and personal names, and to depict themselves with first person pronouns (singular or plural), so as to highlight the human beings behind the professional/commercial roles. Informal speech characterizes everyday conversations, typically with a high degree of closeness and familiarity, and can be conveyed, for instance, by using emojis and interjections. Finally, invitational rhetoric is about communicating to the recipient one’s willingness to create a dialog; one obvious way to do this is to invite consumers to give feedback by means of asking them questions or giving suggestions. Van Herck and Vangehuchten’s cross-cultural analysis of the expression of CHV in organizational emails to complaints reveals, among other things, that the Spanish emails are less personal and invitational than the English ones.

The use of CHV in webcare discourse is also investigated by Holmgreen in the present issue. While Van Herck and Vangehuchten analyze CHV features in organizational email responses in the context of service failures, Holmgreen does so in organizational Facebook reactions to negative customer posts in the context of intentional crises. More specifically, her data are drawn from the Facebook page of a crisis-ridden Danish bank which faced allegations of money laundering. Her results indicate that, in their responses to criticism from customers, the bank employees achieve CHV via a small set of categories. These include, for instance, personal greetings and signatures, good wishes, personal pronouns, the speech acts of apologizing and the expression of regret. Regarding the sequencing of employees’ responses, Holmgreen identifies a predetermined set of moves, with the choice of some of these moves suggesting that the communication strategy of the bank is partly inspired from recommendations made in the literature on crisis communication. However, deviations from this prototypical set of moves occur in her data. They so do, for instance, when the conversation extends beyond the first exchange of messages or when it involves regular users. In addition, Holmgreen explores whether the use of CHV by the bank leads to a positive conversational outcome in the form of customers expressing satisfaction with the response and a positive attitude toward the bank.

While Van Herck and Vangehuchten’s and Holmgreen’s contributions investigate webcare discourse through a genre and/or CHV lense, Lutzky adopts, as already mentioned, a corpus linguistic method by conducting a keyword and cluster analysis. In her analysis of Ryanair’s responses to customer tweets, she finds, for example, that the airline often uses generic and formulaic expressions such as Please contact us DM and Sorry for the inconvenience in response to its customers’ requests for information and complaints. Such generic replies tend not to be appreciated by customers, as her analysis of customer tweets equally reveals.

Methodological Variety

When preparing this Special Issue, we paid special attention to include articles which are grounded in different research traditions so that we would be able to offer an overview of the various, and often complementary discourse and linguistic methods that can provide new insights into the study of business communication. In what follows, we will present these methods in some more detail, following the basic distinction between quantitative and qualitative research.

While Widdershoven et al. adopt an innovative methodological approach to the study of emotional contagion that is purely quantitative, two studies in this issue use quantitative methods complemented with a qualitative approach. Baker and Hashimoto carried out, on the one hand, a quantitative content analysis (Neuendorf, 2017) to identify the polarity of the second constituent connected by an AC in a corpus of about 35,000 restaurant reviews; they explain why this approach is more appropriate that an automatic method such as sentiment analysis. On the other hand, they apply a thematic analysis (Braun & Clarke, 2006), which consists in qualitatively coding the first-level themes featured in the second constituents of ACs, that is, whether they relate to food, service or experience, and, for each of these themes, the second-level themes such as portion, price, quality etc. in the case of the first-level theme food. Another example of a quantitative approach is Lutzky’s corpus linguistic contribution. Based on a corpus of 18,000 complete Twitter threads (about 1 million words), she demonstrates that a combination of corpus linguistic analyses, such as the analysis of keywords, clusters and collocations, can provide relevant insights into the communicative interactions between companies and their customers. Lutzky’s research not only has a quantitative dimension; it also offers a qualitative analysis, as the findings provided by her corpus analyses are complemented by insights gained from looking at the language of individual tweets.

The remaining articles of this Special Issue adopt qualitative methods, in most cases including a quantitative dimension. To begin with, the dataset that Fastrich examines consists of a random sampling of 320 negative hotel reviews originating from Booking.com, half of which in German and the other half in English. She then coded, in these reviews, the instances of (elided) first-person pronoun use; using the scheme developed by Ruytenbeek et al. (2021), she also coded the review components in which each instance of FPP occurred, so as to be able to assess the distribution of these pronouns in the different components of the reviews. By contrast, Holmgreen manually retrieved the messages from 40 users that were posted on the Facebook page of a Danish bank. Having identified the topics of crises in these messages, she determined the type of user based on the frequency of the person’s posts, and carried out a linguistic analysis of the bank’s responses in terms of CHV in order to assess the impact of the use of CHV on users’ (dis)satisfaction. Finally, Van Herck & Vangehuchten retrieved a dataset of 80 emails written by companies in response to customer complaints; these email responses were made publicly available on Facebook and Twitter by the customers themselves. The authors combined a genre analysis (De Clerck et al., 2019) with an analysis of CHV features.

It is interesting to note that, while the majority of articles in this volume adopt a monolingual perspective, both Fastrich and Van Herck and Vangehuchten propose a cross-linguistic approach. Fastrich adopts such an approach to the study of consumer feedback, and she compares English and German reviews only with regard to one linguistic element, FPP. In contrast, Van Herck and Vangehuchten conduct a cross-linguistic analysis of webcare discourse, comparing English and Spanish with regard to a variety of discursive and linguistic features. While both interpret similarities and differences against the background of communicative style (e.g., more or less personalized) and politeness theories, Fastrich also grounds her conceptualization of personalization in cognitive grammar (Langacker, 1999).

Contribution to Business Communication and Managerial Implications

Not only do the individual articles of this Special Issue advance our knowledge of the language of (responses to) consumer feedback online and make significant contributions to current research in the field of business communication, we also believe that they are relevant to companies.

Baker & Hashimoto’s article contributes to research in business communication, as it demonstrates the relevance of combining linguistic methods (corpus analysis) and linguistic theory for addressing the content of customer reviews and its potential impact on purchase intentions and companies’ reputation. Overall, their study nicely demonstrates the importance of AC clauses in communicating and analyzing (dis)satisfaction online. Their findings can be used by restaurant owners to focus on the most important information that is mentioned in customer reviews, that is, the clauses preceding and following adversary connectives (ACs). This is because ACs are key indicators of customer expectations and (dis)satisfaction.

The contribution made by Lutzky shows that the application of corpus linguistics to consumer feedback data can reveal customers’ attitudes toward companies and both customers’ and employees’ use of communicative strategies online. The results of such analyses are important because they can uncover frequent patterns of evaluative language, opening a window into customers’ attitudes and emotions toward (responses from) companies. Corpus linguistic tools are thus highly relevant for scholars interested in business discourse. Within the field of communication studies, this methodology is innovative: to date, corpus linguistics has not often been used to explore external business communication. This approach is in line with the recent tendency in the field of business communication (e.g., Decock, 2022) that wishes to promote the use of methodological tools from linguistics to address questions in professional (online) communication. The results of Lutzky’s research are also of special interest to customer service managers, because they can help them identify which aspects of their webcare are in need of improvement. For instance, these analyses highlight the fact that standard, formulaic answers are often badly perceived by consumers. On the other hand, corpus linguistic approaches also make it possible for companies to detect customer satisfaction—a measure of webcare efficiency—by paying attention to the use of polite expressions in customers’ replies to service managers.

Widdershoven et al. contribute to the field of business communication by proposing a new method for exploring emotional contagion on Twitter, based on an integrated tweet box and a “sliding frame” of four tweets. It is to be hoped that such an approach will be extended to other platforms, such as Facebook and accommodation platforms such as Booking.com. Regarding its relevance for managers, obviously, negative messages from consumers pose a threat to the public image of companies. However, the upshot of Widdershoven et al.’s study is that organizations should pay more attention to the monitoring of messages without a negative valence, especially when they occur amid social media firestorms. Customer service managers can even use these messages to minimize the reputational threat to the organization and integrate positive elements of this feedback in their own communication campaigns.

In general, discourse approaches such as the one proposed by Holmgreen are interesting for research on business communication, as they document the content of companies’ responses to customer feedback in a qualitative way. In a similar vein, Van Herck and Vangehuchten argue that the complexity of CHV should not be underestimated: How CHV should be realized in a context of webcare depends on the genre of business communication in which the interaction takes place. Holmgreen’s article offers key insights into webcare practices, as it demonstrates the necessity to tailor the use of CHV to the particular context in which the interaction between the customer and the company takes place. In particular, this contribution stresses the effect of the user’s profile on the use of CHV by customer service managers. The results of this type of studies can be used by companies to improve their webcare by using CHV in a more appropriate manner, leading to an increase in customer satisfaction.

Finally, Fastrich and Van Herck and Vangehuchten make a similar contribution to business studies. They show that, combined with insights from cognitive linguistics in the case of the former, and with a politeness perspective in the case of the latter, contrastive pragmatics is a helpful approach to examine discourse strategies in customer feedback and/or webcare. While Fastrich’s findings can be used to raise customer service managers’ awareness of the impact a personal construal can have on the trustworthiness of a review, the results of Van Herck and Vangehuchten’s research highlight the importance of adapting the use of CHV to a specific genre as well as to a specific linguaculture.

Perspectives for Future Research

In this Special Issue, we aimed to present linguistic and discursive studies on the topic of consumer feedback and webcare. Having discussed and highlighted the specificities, commonalities and divergences between the individual contributions of this Special Issue in this introductory article, we would now like to consider a few promising avenues for future work.

Following-up on Baker and Hashimoto’s contribution, we deem it important that future research explores a variety of linguistic constructions present in consumer reviews. To begin with, it will be interesting to investigate not only the content of second constituents of ACs, but also the topics of the first constituents of these clauses and how these relate to customers’ expectations. More research is also necessary to uncover possible effects of different ACs in that respect. In addition, Baker and Hashimoto’s study complements previous empirical research on restaurant reviews, such as Fox et al. (2018), who provided evidence for stronger psychophysiological responses in the readers of such reviews when they are negative compared to positive and neutral reviews. It would be interesting to see, in future work, to what extent readers’ emotional responses to reviews are modulated by the presence of AC clauses and the evaluative language used in these clauses.

A second direction for future research in line with the contributions made by Fastrich and Van Herck and Vangehuchten is to explore the effects that discursive strategies such as (im)personalization and CHV more generally have on the readers of reviews in different linguacultures. From the perspective of production, methodologies tapping into the cognitive processes at play in the writing of reviews would be especially helpful to unravel the particular motivations behind reviewers’ linguistic choices.

Finally, we are convinced that, more generally, systematic comparative studies with a linguistic angle of attack will benefit business communication research. For instance, while Ruytenbeek et al. (2021) compared the expression of negativity on the accommodation platforms Booking.com and TripAdvisor, cross-linguistic studies such as Cenni and Goethals (2017) compared Italian, Dutch and English negative reviews on TripAdvisor, and Cenni and Goethals (2020) analyzed the responses to Italian, Dutch and English negative reviews on the same platform. In a similar vein, Feng and Ren (2020) explored the use of impoliteness strategies in Chinese and English reviews on Amazon and Chen et al. (2011) carried out a contrastive study of consumer complaints in English and Chinese. Building on these studies, future work could extend the approaches and analyses presented in this Special Issue to other online review sites, specific types of companies and less well-studied languages. As explained by Lutzky (this issue), future contributions could also combine several of the methodologies exemplified in this collection, such as experimental work complemented with corpus analyses.