A Pragmatic Analysis of the (Im)politeness Strategies Used in Online Customer Reviews across Different Satisfaction Levels

1. Introduction

1.1 Rationale of the Study

Online reviews are clearly ubiquitous, serving as online discourse for the partially covered categories of digital discourse (Barton & Lee, 2013), computer-mediated communication (Herring et al., 2013), and social media platforms (Hoffmann & Bublitz, 2017). This communication is interactive, either directly or indirectly, and serves a variety of functions, such as personal and social ones. People who post reviews on online platforms about products and services may influence service providers. Both positive and negative reviews appear to affect service providers differently.

Reviews may contain either compliments or criticism. Reviewers are free to use both polite and impolite language in their reviews. Politeness strategies differ from impoliteness strategies in that the former aim to save the other person's face. Brown and Levinson (1987) note that politeness theory focuses on ways to protect the addressee's public self-image. On the other hand, "communicative strategies designed to attack faces and cause social conflict and disharmony" is how Culpeper (2003) describes impoliteness. These two strategies work with people's goals, situations, and cultures.

Trustpilot (https://www.trustpilot.com) is a free online service anyone worldwide can use. When experiences are shared, people can make better choices, and companies can improve their products and services. This platform suggests real-world reviews. People often write online evaluations using politeness and impoliteness strategies, rating their satisfaction from one to five stars. Therefore, it is intriguing to see which politeness and impoliteness strategies are used in composing those reviews at each level of satisfaction.

Previous studies on reviews (e.g., Trosborg, 1995; Geluykens and Kraft, 2003; Feng and Ren, 2020; Wu and Yang, 2024) utilized a taxonomy to analyze negative reviews. Olshtain and Weinbach (1987) originally developed this kind of taxonomy, which subdivides complaints or negative reviews into sub-strategies. Some of these studies have analyzed variables such as gender (Geluykens and Kraft, 2003), social status (Boxer, 1993), cultural and ethnic differences (Cohen and Olshtain, 1993), and cross-linguistic differences (Feng and Ren, 2020; Wu and Yang, 2024). Generally, these studies presume that negative reviews typically target the person accountable for a particular offense or situation. Consequently, a significant portion of this research also explores linguistic strategies for mitigating negative reviews. On the other hand, reviewers also offer positive comments or report satisfying experiences with products or services. Customers or consumers may use politeness strategies to offer compliments as a sign of solidarity or to preserve social ties (Alqarni, 2017).

The current study applies the ideas and elements of earlier studies to gather and evaluate data from Trustpilot (https://www.trustpilot.com). The methodology for this study differs in its data categorization. Previous research may fail to identify the combination of politeness coupled with impoliteness strategies among online reviewers. As a result, this study attempts to provide insight into how online reviewers use politeness and impoliteness strategies through different levels of satisfaction when composing online reviews.

1.2 Literature Review

This section covers the theoretical and empirical foundations for researching (im)politeness. Section 1.2.1 provides a summary of the key concepts of politeness and impoliteness, and Section 1.2.2 addresses three key factors that influence (im)politeness in social situations.

1.2.1 Perspectives on (im)politeness

Brown and Levinson (1987) defined face as "a public self-image that every member wants to claim for oneself". They proposed the concept of courtesy to preserve the interlocutor's face during discourse, recognizing its potential to maintain face. They also asserted that politeness is a universal norm in language use and classified politeness strategies into five categories as detailed below.

1) Bald-on record is defined as an activity in which a speaker talks directly or without redressive actions of demonstrating their direct intentions.

2) Positive politeness is used to tailor words to the hearer's preferences in order to be liked, accepted, or appreciated.

3) Negative politeness refers to a speaker's actions that modify their words to provide the hearer with freedom by trying not to impose on the hearer’s face.

4) Off-record refers to behaviors in which a speaker speaks indirectly to a hearer by employing words with multiple meanings or leaves implicature to be interpreted.

5) Don't do the FTA (face-threatening act) refers to any action that does not endanger the hearer's face by saying nothing.

In contrast to the concept of face-saving politeness, face-threatening acts (FTAs) are deliberate actions intended to affect the addressee through verbal or nonverbal communication. Brown and Levinson (1987) noted that individuals may engage in acts that jeopardize others' wants. Consequently, these acts are considered impolite and can disrupt social harmony, potentially leading to conflict. O’Keeffe et al. (2011) defined FTAs as "a communicative act performed by the speaker that does not respect either the hearer’s need for space (negative face) or their desire for their self-image to be upheld (positive face) or both." While politeness focuses on preserving interactants' faces, impoliteness attacks the hearer's face, which may cause the hearer to lose face.

Besides O’Keeffe‘s definition of FTA, Huang (2015) defined impoliteness as "any face-aggravating behavior, including verbal behavior relevant to a particular context." Culpeper (1996) introduced a model of impoliteness as a counterpart to Brown and Levinson's (1987) model of politeness. He characterized impoliteness as the parasitic counterpart of politeness, differing in terms of orientation to face. Impoliteness serves as a means of attacking faces, disregarding the interlocutor's humanity. Culpeper’s framework provides a comprehensive understanding of the various strategies employed in impoliteness, each serving as a distinct approach to threaten or attack face in communication. Culpeper identified five super strategies of impoliteness as follows:

1) Bald-on-record impoliteness is when someone performs a face-threatening act (FTA) in a direct, straightforward, and unambiguous manner while disregarding their facial appearance.              

2) Positive impoliteness seeks to undermine the addressee's favorable image, such as by ignoring, excluding, exhibiting disinterest, and using inappropriate language.                                                                

3) Negative impoliteness seeks to undermine the addressee's unpleasant manners. Examples include being threatening, condescending, scornful, failing to take things seriously, or violating personal space.                                            

4) Sarcasm, or mock politeness involves performing an FTA using politeness strategies that are insincere, leaving some inferences to be uncovered.                                                                                                                

5) Withholding politeness is the absence of required politeness expressions, resulting in a void where politeness would normally be expected.

1.2.2 Three Contributing Factors to (Im)politeness

Brown and Levinson (1987) explain that politeness strategies depend on three key factors: the social distance between the speaker and the listener (D), the listener’s power over the speaker (P), and the degree of imposition of the act (R). These factors combine to determine how polite a speaker needs to be. For instance, greater imposition or greater social distance generally calls for more politeness, while less distance or greater power on the speaker’s side may lead to less formal language. Unless urgency outweighs these factors, speakers possibly try to minimize threats to the listener’s “face.”

Brown and Levinson assert a direct link between social distance (D) and politeness. The greater the distance between the two speakers, the more polite they should be. Leech (2007, p. 189) supports their argument, stating that "when horizontal distance is reduced (e.g., in communication with familiars or intimates), the need for politeness is also reduced until we move into the zone of non-politeness or impoliteness." However, Spencer-Oatey (2005) paints a more nuanced picture of D's impact on politeness. Politeness is associated with interactional objectives and rapport management. A speaker can have many goals in an interaction depending on how she wishes to affect the long-term rapport level between herself and the listener. Spencer-Oatey (2005, p. 96) proposes that a speaker may have multiple rapport orientations within a relationship. Interlocutors may have a rapport-enhancement orientation (a desire to strengthen or enhance harmonious relations between the interlocutors), a rapport-maintenance orientation (a desire to maintain or protect harmonious relations), a rapport-neglect orientation (a lack of concern or interest in the quality of relations, possibly due to a focus on themselves), or a rapport-challenge orientation.

In terms of power (P), Brown and Levinson identify the players' relative power as the third determinant component of politeness. According to Brown and Levinson, P is culture-specific and difficult to correlate with specific social elements, as each connection has multiple layers that contribute to the perceived power imbalance between two people. P is a complex variable that encompasses several socially and individually influenced elements. According to Van Dijk (1989), when evaluating P, interlocutors must be aware of the power difference between them. Power might be domain-specific, such as teacher-student in a school environment or seller-buyer in a commercial setting.

The rank of imposition (R) refers to the level of threat associated with a given FTA in the appropriate culture. Brown and Levinson provide examples of intrinsic FTAs that demonstrate the speaker's (S) disregard for the hearer's (H) favorable facial expression. Those FTAs include "mentioning (of) taboo topics, including those that are inappropriate to the context," as well as "raising of dangerously emotional or divisive topics, e.g., politics, race, religion, women's liberation" (Brown and Levinson 1987, p. 67). Many politeness studies have found that the level of imposition, social power, and social distance all influence politeness and mitigation (Leech, 1983; Brown and Levinson, 1987). Briz (2003) uses the example of lending a car versus lending a book to demonstrate the impact of imposition on mitigation. He stated that lending a car is a more challenging circumstance in which to reach an agreement than lending a book, and that mitigation is more likely in the car situation.

Politeness varies depending on the context. The seriousness of face-threatening acts is determined by the speaker's and hearer's social distance (D), the level of familiarity and solidarity between two people, the speaker's and hearer's relative power (P), the power that the hearer has over the speaker, and the ranking (R) of imposition in the culture related to the speaker's desire to impose on the hearer and the hearer's acceptance of the imposition (Brown and Levinson, 1987). All three criteria (D, P, and R) influence the seriousness of an FTA, and the speaker will determine which politeness strategy to employ to reduce the danger. Brown and Levinson's findings have provided a theoretical framework for studying speaker-to-speaker face connections as grounded in underlying assumptions about the relationship encoded in polite deference and solidarity.

1.3 Previous Studies

(Im)politeness in online discourse is shaped by both platform norms and cultural background (Brown & Levinson, 1987; Kádár & Haugh, 2013). While different digital environments influence how online reviewers express criticism, cultural values affect preferences for direct or indirect strategies (Wu & Yang, 2024; Zhao, 2024; Zhao & Wahid, 2024). Accordingly, this section reviews studies on platform-based differences and cultural variations in (im)politeness.

1.3.1 Comparison of Impoliteness in Online Reviews and Social Media Discourse

Research on impoliteness on online platforms was conducted by Wu and Yang (2024) and Pasana et al. (2023).  While Pasana et al. (2023) investigated Facebook comments disparaging a government agency, Wu and Yang (2024) examined consumer evaluations on Amazon and Douban despite their different settings. These two studies disagreed on one thing. The main strategy was to be impolite on record, as user comments often included direct criticism, ridicule, and claims of incompetence. Chinese reviews tended to favor single-strategy phrases, suggesting a cultural preference for indirect confrontation in Chinese discourse compared to the more direct style of English-language speech, while Wu and Yang (2024) noted that English-language reviews often combined many impoliteness strategies.

Zhao (2022) investigated politeness on TikTok where people responded to video content with both on-record and off-record impoliteness. Unlike the methodical assessments examined by Wu and Yang (2024), impoliteness on TikTok seemed more natural and engaging, often serving purposes such as humor, social connection, or criticism of a creator's reputation. Pasana et al. (2023) also noted humor-based impoliteness in a different form, as sarcasm and ironic ridicule were used in Facebook comments to express discontent toward institutions rather than individual content providers or commercial products.

Vásquez (2011) examined complaints in TripAdvisor hotel evaluations and found that many negative reviews, coupled with recommendations or advice, indicated a speech act set in which users criticized and provided assistance to potential clients. Unlike the overt hostility seen on social networking sites, TripAdvisor reviews often mixed harsh criticism with positive comments. This shows that platform rules influence how people express rudeness. In other words, social media is more direct and dramatic, while review sites combine feedback with advice.

These results suggest that internet platforms shape impoliteness strategies in line with their communication norms. Customer reviews on e-commerce and travel sites focus on products and services, Facebook comments often target organizations, and TikTok interactions revolve around entertainment and social identity. This evidence seems to indicate that the (im)politeness strategies users use to express dissatisfaction depend on the interaction styles of each platform.

1.3.2 Cultural Variations in Politeness and Impoliteness Strategies

Cultural variation plays an important role in shaping how politeness and impoliteness are expressed, particularly in online discourse. A large body of research suggests that linguistic behavior is closely tied to cultural expectations, values, and norms regarding appropriate communication. Drawing on pragmatics and intercultural communication, scholars have frequently demonstrated that speakers from different cultural backgrounds employ distinct ways to manage face, express evaluation, and negotiate interpersonal relationships.

Studies comparing Eastern and Western communication styles commonly highlight differences in directness and indirectness. For instance, Wu and Yang (2024) and Zhao (2022) found that English-language reviewers tend to employ more direct, bald-on-record impoliteness, explicitly stating dissatisfaction or criticism. In contrast, Chinese reviewers rely on off-record strategies such as implication, metaphor, or sarcasm to convey negative evaluations indirectly. These findings correspond with Zhou (2024), whose analysis of movie reviews revealed that English reviewers prefer clear, explicit complaints, whereas Chinese reviewers soften criticism through indirect language or rhetorical devices.

Such patterns can be interpreted within broader cultural frameworks, particularly high- and low-context cultures (Hall, 1976). In high-context cultures (e.g., China, Japan, Thailand), meaning is often conveyed implicitly, and preserving social harmony is preferred. As a result, speakers may avoid direct confrontation and instead use indirect or mitigated forms of criticism. In contrast, low-context cultures (e.g., the United States, the United Kingdom) favor explicitness, making direct expressions of dissatisfaction more acceptable.

Similarly, politeness theory yields a useful framework for understanding these differences. Brown and Levinson (1987) argue that speakers use politeness strategies to manage face-threatening acts (FTAs). In collectivist cultures, where social unity and face-saving are highly valued, speakers are more likely to employ negative politeness and off-record strategies to minimize imposition. In contrast, individualist cultures may tolerate more bald-on-record expressions, specifically in contexts such as online reviews where individual viewpoints are expected.

Empirical studies of online reviews additionally support these cultural attributes. Vásquez (2011), in the analysis of TripAdvisor reviews, found that many reviewers adopt hybrid strategies, combining positive politeness (e.g., compliments, appreciation) with indirect criticism. This suggests that even within English-language contexts, there is a tendency to soften negative evaluations to maintain a socially acceptable tone. Likewise, Terkourafi (2008) emphasizes that politeness is not only culturally bound but also context-dependent, determined by genre conventions and expectations. In customer reviews, for example, balancing praise and criticism may be seen as a normative practice.

Another work highlights the role of discursive norms in digital environments. Kádár and Haugh (2013) argue that politeness and impoliteness are co-constructed through interaction and influenced by shared cultural understandings. In online contexts, where anonymity prevails and social constraints are reduced, speakers may deviate from traditional norms, yet cultural tendencies persist. Grainger (2020) further notes that indirectness, silence, or minimal responses can function as subtle forms of impoliteness, particularly when they violate expectations of engagement.

Overall, previous research indicates that cultural background greatly impacts the choice of politeness and impoliteness strategies. However, these strategies are not fixed and interact dynamically with contextual variables such as genre, power relations, and communicative intent. Consequently, understanding online review discourse requires an integrated approach that examines both cultural standards and situational factors in forming language use.

2. Method

This study utilized a mixed-methods approach to analyze politeness and impoliteness strategies in online reviews. The quantitative analysis investigated the occurrence and frequency of these strategies across five levels of satisfaction, while the qualitative analysis examined selected examples to provide deeper insights into how these strategies are pragmatically realized in context. The research instrument used in this study was a collection of online customer reviews obtained from Trustpilot (https://www.trustpilot.com).

2.1 Samples

Samples for the study were selected using stratified random selection to ensure representation across different levels of customer satisfaction. A total of 250 customer reviews were collected from Trustpilot.com. Each degree of satisfaction is represented by 1 star (lowest) to 5 stars (highest), and 50 reviews were selected from each level. This stratification guaranteed that all segments of consumer experience were equally represented. The sample size of 250 reviews was considered sufficient for qualitative content analysis, as previous research in discourse analysis has shown that datasets of 200 to 300 units can yield reliable and significant patterns (e.g., Krippendorff, 2018; Guest et al., 2006). Online reviews were chosen as the primary data source because they provide authentic, spontaneous language from consumers in real-life situations. Thus, the chosen samples provide a full account of the language aspects of the different levels of consumer satisfaction.

2.2 Data Collection

In this study, data were collected from five distinct sample groups based on customer satisfaction levels: one-star, two-star, three-star, four-star, and five-star reviews. All data were obtained from Trustpilot.com using the same data collection method to ensure consistency across groups. Specifically, a stratified random sampling technique was used to select 50 reviews from each satisfaction level, yielding a total of 250. The data were collected within the specified period of March 2026. Each review was extracted manually and recorded in a structured dataset, including relevant information such as star ratings and symbolic and textual content. Although the sample groups differ in satisfaction levels, the data collection procedure remains consistent across all groups to ensure comparability and the reliability of the analysis.

2.3 Data Analysis

The analysis of politeness and impoliteness strategies was based on the theoretical frameworks of Brown and Levinson (1987) and Jonathan Culpeper (1996, 2005, 2011). These frameworks provide a foundation for identifying how reviewers use face-saving (politeness) and face-threatening (impoliteness) strategies in their evaluations.

The analytical procedure consisted of both quantitative and qualitative components. First, each review was carefully examined to identify instances of politeness and impoliteness strategies. These instances were coded according to four main categories: bald on record, positive, negative, and off-record, along with their corresponding sub-strategies. The coding process focused on observable linguistic features, including directness, hedging, mitigation, sarcasm, and implicit meaning.

For the quantitative analysis, all identified instances were systematically counted, and the frequencies were converted to percentages to facilitate comparison across the five satisfaction levels. This allowed for the identification of dominant patterns and variations in strategy use at each level. The results were then presented in tables to provide a clear overview of distributional trends.

For the qualitative analysis, selected examples from each satisfaction level were examined in detail to explore how these strategies function in context. Particular attention was given to the reviewers’ (im)politeness strategy choices and the pragmatic effects of specific linguistic choices. This step was vital for understanding how strategies may serve different functions depending on context, thereby distinguishing between surface-level forms and underlying intentions.

3. Results

In line with the research objective, this section analyzes politeness and impoliteness strategies across five satisfaction levels in Trustpilot reviews, presenting their frequency, percentage, and examples.

3.1 Level of satisfaction: one star

Table 1 presents the distribution of politeness and impoliteness strategies employed in one-star customer reviews. The strategies are divided into four main types: bald on record, positive, negative, and off-record, along with their specific sub-types. The frequencies and percentages are calculated to illustrate how often each strategy is used within both politeness and impoliteness frameworks.

Table 1: Frequency of (Im)politeness Strategies in One-Star Reviews

The results show a clear difference between how politeness and impoliteness are used in one-star reviews. Within the politeness framework, bald-on-record strategies overwhelmingly dominated, indicating that even when politeness is present, it tends to be expressed in a highly direct, unmitigated manner. This suggests that dissatisfied customers prioritize clarity and intensity over maintaining social harmony.

On the other hand, impoliteness strategies were more varied. The most common impoliteness strategy was positive, where reviewers attacked the person’s self-esteem by ignoring them, acting uninterested, or using offensive language. Other impoliteness strategies, such as negative and off-record strategies, were frequently used, indicating that harsh criticism is common.

Those who rated one star showed deep discontent and rage, often expressing their frustration directly and sometimes in an impolite way. The following are the instances.

Figure 1: Customer review (A frustrating experience with a sneaker website)

One-star reviews stand out significantly because reviewers use strong, emotional language to express their dissatisfaction, making the difference in review ratings even more noticeable. In Figure 1, "This is the worst site to buy sneakers…" was a clear, face attack that showed frustration. Likewise, the review in Figure 2, “Stay away! Likewise, the review "Stay away! They are a scam" was a warning meant to damage the business's reputation. Another remark in Figure 3, "€1000 euros stolen... management not interested," criticized employees for carelessness without any effort at courtesy. These instances show that, whereas bald-on-record statements may seem structurally similar, their pragmatic purpose differs greatly depending on whether they aim to convey effectively or cause reputational harm.

It is crucial to note that customer evaluations that include the frustrated-face emoji in Figure 3 serve a significant practical and emotional purpose, amplifying the expressive power of the complaint. Although the linguistic material such as "police were their friends when they turned up," already conveyed annoyance, the emojis graphically exaggerated the feeling, conveying a significant amount of emotion and moral outrage that written language alone may not fully capture. Emojis, according to Dresner and Herring (2010), are paralinguistic signals that help explain or emphasize emotional meaning, much like the tone of voice in verbal communication. In this situation, the emojis serve as visual indicators of severity, underscoring the seriousness of the accusations, including theft, mismanagement, and potential corruption. Furthermore, from a pragmatic viewpoint, the use of emojis in this situation supported bald-on-record impoliteness (Culpeper, 1996), in which the reviewer skipped softeners and directly threatened the company's face. The emojis increased the face-threatening nature of the criticism by adding visual cues of rage to the criticism, thereby making the evaluation more forceful and emotionally loaded (Derks, Fischer, & Bos, 2008). Therefore, emojis work rhetorically to highlight discontent and guarantee the clear communication of the reviewer's emotional position.

3.2 Level of satisfaction: two stars

The findings reveal that in two-star reviews, politeness strategies are primarily characterized by a balance between bald on record and negative politeness, indicating that reviewers tend to combine direct expressions with mitigating elements such as explanations or reasons. This suggests a shift from purely direct criticism toward a more elaborated and rationalized form of complaint. In contrast, impoliteness strategies remain dominated by bald-on-record forms although their overall frequency is lower and less diverse than in one-star reviews. Positive and negative impoliteness are present but less prominent, while off-record impoliteness is minimal, indicating a reduced reliance on indirect or implied attacks.

Table 2: Frequency of (Im)politeness Strategies in Two-Star Reviews

One-star reviews express dissatisfaction, often using strong emotional language and claiming a scam, whereas two-star reviews still reflect disappointment but may mention positives, like eventual resolution or partial service, as illustrated below.

In Figure 4, this two-star review expressed disappointment not only with a defective item but also with the caliber of customer care. The reviewer's dissatisfaction stemmed from an inadequate customer service response to an incomplete item, which was a countertop bench lacking bolts. By saying things like "NO SERVICE" and "Truly awful service," the reviewer used a bald-on-record politeness strategy (Brown & Levinson, 1987) to openly voice discontent without any effort to soften the complaint. These statements immediately called into question the company's competence and concern, resulting in unmitigated face-threatening activities (FTAs). Additionally, the sarcastic rhetorical question in "How can I send a photo of something I don’t have?" was also an off-record strategy, especially a rhetorical use meant to mock rather than ask. Culpeper (1996) claimed that such use of rhetorical questions is a subtype of impoliteness, falling under both off-record and positive impoliteness strategies, since it indirectly criticizes and therefore disturbs or mocks the hearer. The tone of the two-star review was more negative and impolite, as the reviewer claimed they had to purchase the bolts themselves, so linking the business to a negative quality was useless and untrustworthy. This action records the company's debt, which is a sign of negative impoliteness (Culpeper, 1996). Furthermore, all-capitalized phrases like "SEND US A PHOTO OF THE MISSING BOLTS" underlined mocking and skepticism, hence strengthening the face-threatening tone. Recent studies by Zhang and Vásquez (2014) supported the idea that consumers frequently used sarcasm, rhetorical questions, and exaggeration in online complaints to express strong criticisms, especially in low-star ratings. Similarly, Chen et al. (2011) emphasized that digital reviewers might alternate between direct and indirect approaches to express rage while preserving some level of social etiquette.

In Figure 5, the complaint was about slow replies and inadequate communication from several hotel agents. Although the reviewer was clearly annoyed, a combination of politeness and impoliteness strategies was employed to express criticism. Hedging statements like "I believe things can be improved" exhibited optimistic conventional indirectness and helped reduce imposition, therefore reflecting negative politeness according to Brown and Levinson's (1987) approach. Acknowledging the manager's usefulness: "He did a great job." The customer also used positive politeness to preserve group affiliation and soften the tone of the complaint. However, Culpeper's (1996) classification clearly identifies some impolite traits. The reviewer mocked the company's ongoing failure to address the problem in "They said the same thing: they were unable to connect," highlighting the company's repeated inability to fix the issue. Moreover, the utterance "the other staff were not helpful" directly linked the hearers to a negative quality; "they couldn't provide compensation..." put the company's debt on record, therefore raising the face risk. Chen et al. (2011) asserted that these strategies typically emerge during prolonged delays, leading consumers to perceive their time as undervalued. Vásquez (2011) also observed that online reviews often need to balance politeness and criticism to appear fair and maintain credibility in public debate. Thus, this review showed the utilization of both politeness and impoliteness to convey discontent while maintaining the reviewer's own optimistic face.

The brief critique "Rates are ridiculously high!" in figure 6 used bald-on-record impoliteness to confront the addressee's positive face without softening or correcting the message. The word "ridiculously" heightened the judgment by implying intense emotional displeasure and possibly irony. Regarding cultural views, recent research has found that negative online customer reviews are full of direct, emotionally charged language. For example, Feng and Ren (2020) found that English-language evaluations often employ more favorable impoliteness strategies, such as sarcasm and direct criticism. This implies a cultural tendency in English evaluations toward clear statements of discontent. Moreover, Wei (2024) found that unfavorable online evaluations often use bald-on-record strategies to show extreme disapproval, particularly when consumers believe their expectations have not been met. Emphatic evaluative words like "ridiculously" worked as mock impoliteness, insulting the reader's competence or judgment. The review's lack of mitigating language in this context seems to intensify the face-threatening nature of the feedback, consistent with findings from recent studies on impoliteness in online consumer reviews.

3.3 Level of satisfaction: three stars

The results indicate that three-star reviews exhibit a more balanced distribution of politeness strategies compared to lower-rated reviews. While bald-on-record strategies remain the most frequent, their dominance is reduced, and positive politeness strategies increase notably. This suggests that reviewers at this level tend to combine direct expressions with supportive or cooperative elements, such as providing understanding, justification, or constructive feedback. Negative politeness and off-record strategies are also present but play a less prominent role.

Table 3: Frequency of (Im)politeness Strategies in Three-Star Reviews

In contrast, impoliteness strategies occur less frequently. This indicates a reduced reliance on overt face-threatening acts and a shift toward more moderate forms of evaluation. Overall, three-star reviews reflect a communicative style that balances criticism with politeness, signaling a more neutral or mixed evaluation.

Reviews scored at the three-star level usually show moderate discontent rather than severe rage. Reviewers in this category frequently use fewer harsh words, and some try to offset their grievances with neutral or somewhat positive comments. These are the examples.

In Figure 7, the use of ellipsis in the review (e.g., “The price and booking experience with…”) can be interpreted as an off-record strategy, in which the reviewer intentionally left the statement incomplete. This indirectness allows the reader to infer the full meaning without explicitly stating it, thereby softening the critique and reducing potential imposition. Then, the reviewer used a combination of positive politeness, bald-on-record, and negative politeness strategies. The review began on a polite note, acknowledging the hotel’s reasonable pricing and easy booking process, which aligns with positive politeness, particularly the strategies of attending to the hearer’s wants and showing optimism. This served to soften the criticism that follows. However, the tone quickly shifted to a bald-on-record approach, with direct statements such as “It was not a 4-star hotel at all” and “Service was bad and a lack of proper management.” These utterances reflected unmitigated criticism, consistent with Brown and Levinson’s description of face-threatening acts (FTAs) performed without redressive action. Additionally, the reviewer offered justification for their dissatisfactory politeness strategy used to provide and maintain a sense of rationality and fairness. This tendency is often observed in moderate-star reviews, where reviewers aim to combine straightforward strategies with a degree of courtesy (Chen & Lv, 2025; Wu & Yang, 2024).

In Figure 8, the reviewer expressed justified dissatisfaction with a claim process that lasted approximately six months. They lamented the prolonged delays, poor communication, and the significant amount of time wasted on phone calls and emails. Although compensation was eventually received, it only occurred after the head office intervened. The reviewer’s overt frustration was evident in comments such as “I wasted hours,” which exemplifies a bald-on-record strategy, while the expression “jumping through hoops” demonstrates an off-record strategy by indirectly alluding to the difficulties experienced. However, the reviewer also justified their complaints by contextualizing them within a sequence of events. For example, the statement “Eventually my claim was paid, but only after...” illustrates Brown and Levinson’s (1987) negative politeness through justification, legitimizing the complaint and maintaining the speaker’s credibility. Chen and Lv (2025) argued that consumer reviews expressing reasonable dissatisfaction often combine narrative structure with direct complaint to enhance the persuasiveness of criticism.

In Figure 9, the reviewer viewed the introductory videos and expressed a desire to participate in the program as an affiliate. Their website, however, lacked any contact details, such as a phone number or email, which left them doubtful. The reviewer requested that the business provide contact information and was expecting a response. Then dissatisfaction was expressed through negative politeness and bald-on-record techniques, while maintaining formal respect. The reviewer first employed positive politeness to indicate interest and common objectives, in line with Brown and Levinson's (1987) concept of attending to the hearer's needs. The change to unhappiness, however, showed hedging and indirectness, which were characteristic of negative politeness, as it read, "This raises some doubt." A bald-on-record strategy emphasizing clarity over face-saving was the direct order, "Plz provide your phone number". Often, internet evaluations use this method, in which the reviewer combined civility with aggressiveness. On digital platforms, such hybrid strategies usually reflect a compromise between face-threatening actions (FTAs) and the reviewer’s aim of eliciting a response from businesses (Zhao, 2022, p. 125). Pasana et al. (2023) found that, similarly, even three-star-rated online reviews often use bald-on-record actions when reviewers feel overlooked or need follow-up.

3.4 Level of satisfaction: four stars

The findings show that four-star reviews are strongly characterized by the use of politeness strategies, with bald on record remaining the most frequent, primarily used to express sincerity and straightforward positive evaluation. Positive politeness strategies are also evident, including expressions of appreciation, optimism, and supportive reasoning, indicating a cooperative and affiliative communicative style. Negative politeness appears to a limited extent, mainly through the use of hedges to soften minor criticisms. Off-record strategies are not evident, suggesting that reviewers prefer clarity and directness even in relatively positive evaluations.

In contrast, impoliteness strategies are used only sparingly in the four-star reviews. While some instances of bald-on-record, positive, and negative impoliteness are present, off-record impoliteness strategies are absent. This suggests that reviewers generally prefer relatively direct yet restrained expressions when conveying mild dissatisfaction or criticism. Overall, the reviews maintain a largely positive, respectful tone, emphasizing appreciation and balanced evaluation rather than strong, face-threatening attacks.

Table 4: Frequency of (Im)politeness Strategies in Four-Star Reviews

While the three-star rating reflects indifference and vague dissatisfaction, the four-star review contains engaged and positive comments that clearly express a desire to improve the service.  The examples are as follows:

In Figure 10, the reviewer started by stating they felt great, then repeated it to emphasize approval. This aligned with positive politeness strategies, especially the actions of increasing interest and demonstrating approval (Brown & Levinson, 1987). Using hedging and a conditional statement, the reviewer expressed concern about the limited service hours for individuals in the Pacific Time Zone, which reflected poor politeness and off-record strategies.

These options let the reviewer voice a worry indirectly, therefore avoiding imposition and keeping a respectful tone. Although it still expressed discontent, the use of inference aimed to encourage the business to consider a time zone change without directly requesting it, thereby demonstrating courtesy. Overall, the review was well-organized, balanced praise with a gentle suggestion, and was a great example of offering thoughtful, respectful criticism.

In Figure 11, the customer began with thanks to a helpful staff member, demonstrating high positive politeness through compliments and thanks (Brown & Levinson, 1987). But the reviewer quickly brought up a problem with the call quality, saying, "It would have been a 5, but..." which may be seen as a typical example of hedged criticism. This is in line with Chen et al.’s (2011) study of how customers often use polite language to mask complaints so they do not seem harsh. The reviewer attempts to be fair, but the usage of contrastive structures ("but") and the fact that the agent has to repeat herself over and over again show that she is still unhappy. Culpeper (2011) calls this "impoliteness behind a polite front," meaning relationship tension, but it is deliberate to avoid open conflict.

The review in Figure 12 was very short and appeared incomplete, consisting only of "what I was looking for." This brevity, especially when thorough reviews are expected, can be interpreted as an ellipsis, suggesting more is left unsaid. Terkourafi (2008) notes that breaking genre rules, such as not explaining the 4-star rating, implies something beyond neutrality. Here, the shortness may signal emotional detachment or hidden dissatisfaction. Locher and Watts (2005) assert that even brief answers carry social significance. In this case, the restrained politeness hinders genuine engagement or connection.

Overall, the analysis of 4-star reviews shows that people are mostly polite, but there are also hints of tension in their words. While reviewers often use polite language, such as thanking and praising, they also include criticism in an indirect or softened way, often mixed with their positive comments. Additionally, Figure 12 demonstrates a more implicit form of resistance. Such strategies may not appear overtly impolite but can still disrupt expected interactional norms, functioning as covert expressions of dissatisfaction (Culpeper, 2011; Locher & Watts, 2005).

3.5 Level of satisfaction: five stars

The findings demonstrate that five-star reviews are overwhelmingly characterized by the use of politeness strategies, particularly bald-on-record and positive politeness. While bald-on-record strategies remain highly frequent, they are primarily used to express sincerity, honesty, and a clear positive evaluation rather than criticism. Positive politeness strategies also play a significant role, reflecting expressions of appreciation, gratitude, optimism, and compliments, which contribute to a highly supportive and affiliative tone. Off-record strategies, such as emoticons and ellipses, further enhance the friendly, expressive nature of these reviews, while negative politeness appears only minimally. In contrast, impoliteness strategies are almost entirely absent, with only a few instances observed. This indicates avoidance of face-threatening acts and highlights the predominance of respectful, positive communication.

Table 5: Frequency of (Im)politeness Strategies in Five-Star Reviews

Five-star reviews show that the customer is completely happy and uses polite, thankful, and emotional words (Brown & Levinson, 1987). These reviews often use emoticons, intensifiers, and personal compliments to show that they really like something and to connect with the person who wrote it. They do not usually have complaints or threats of violence, which makes it ideal for studying how language strengthens social bonds (Locher & Watts, 2005). Here are some examples:

The above review started with a clear positive tone, using utterances like "Love the hoodies’ warmth and look they…." and "They are so comfortable!" These statements strengthened the bond and shared appreciation among people in the same group. The tone was friendly and excited, and positive adjectives like "love" and "comfortable" were used to make people feel more connected. The reviewer is probably trying to praise and confirm the good quality of the provider’s product.

However, the second part of the review, "Shipping takes too long to get your items!" introduced a slight shift in tone. The expression indicated a potential threat to someone's face, but it was less severe because it did not involve direct guilt or accusation. Instead, the issue was made in a straightforward way, and then a face-saving emoji             (       ) was added to lighten the tone. Chen et al. (2011) proposed that these kinds of indirect complaints in service feedback are a way for the speaker to be polite, as the reviewer would rather not confront the person but still wants to show that they are unhappy. Culpeper's (2011) theory of "politeness-impoliteness hybridity," which entails wrapping criticism in praise to maintain balance, reinforces this idea.

The reviewer followed the usual genre rule for a 5-star rating, which is to praise the work, but they also included a marker of displeasure. It is noticeable that this goes against the usual style of five-star reviews and may suggest hidden conflict. Even though the review has a five-star rating, the minor criticism introduces a polite form of resistance. Locher and Watts (2005) call this "relational work," in which disagreement is framed harmoniously.

In Figure 14, the reviewer praised the delivery and service ("the service and posting were excellent") and said they would buy from the company again ("I would buy from you again..."), which is a polite way to express their satisfaction. The use of positive strategies displays approval, builds solidarity, and shows friendliness (Brown & Levinson, 1987). The constant use of emojis (              ) and the casual "Hi" at the beginning are examples of digital-era politeness that is expressive, pleasant, and relational (Locher & Graham, 2010). The last "thanks" restored a tone of thankfulness and friendliness, helping keep relationships strong (Locher & Watts, 2005).

This review is a wonderful example of positive politeness, as it is enthusiastic and emotionally honest. The reviewer stressed that they would use the service again ("on several occasions"), that they interacted directly with workers, and that the service was excellent. These strategies reflected a desire to become closer to others, make friends, and show gratitude (Brown & Levinson, 1987). The approbation maxim (Leech, 2007) says to use phrases like "They listen, understand," and "go out of their way" to give the hearer as much praise as possible and make their positive face even better.

The reviewer made the compliments more personal by mentioning a staff member ("Brian K.") and thanking him directly. This made the relationship even closer. Locher and Watts (2005) argue that this form of language is relational work, in which positive feedback helps maintain and strengthen social ties. Using strong modifiers like "truly" and "highly recommend" shows that the reviewer is very emotionally involved, which makes politeness more emotive (Kádár & Haugh, 2013).

Figure 16: Customer review (Exceptional service quality and customer satisfaction)

This review in Figure 16  shows how to be nice in a digital setting. The reviewer used positive politeness to offer express praise ("you are really the best team ever") and to express thanks ("thanks for doing everything for us"). These actions bolstered the hearer's self-esteem and fueled their desire for love, acknowledgment, and recognition (Brown & Levinson, 1987). "Keep going, guys 😘❤️" was a mix of emotional support and digital signals (emojis) in the off-record strategy that show unity, warmth, and community alignment.  

The use of casual words like "U" and "guys" and the lack of punctuation may seem informal, but they actually add a tone of spontaneous affection rather than carelessness. Locher and Graham (2010) suggested that digital relational labor commonly uses informal syntax and emojis to convey civility and emotional support without words. The repeated praise and encouragement demonstrated strong emotional commitment, which aligns with Kádár and Haugh's (2013) focus on emotional etiquette in public digital praise.

The five-star reviews consistently express high levels of satisfaction through polite expressions such as praise, gratitude, emotional intensity, and supportive language (Brown & Levinson, 1987). Reviewers used emojis, a casual tone, and personal compliments to make people feel part of a community and feel kind. There were no direct complaints or threats, and any minor issues, if present, appear significantly less serious. These reviews aligned with what people want from the genre: strengthening social ties and keeping customers coming back (Locher & Watts, 2005; Kádár & Haugh, 2013). Overall, five-star reviews are more than just evaluations; they help establish relationships by showing significant emotional involvement without any tension.

4. Discussion

The results of this study contribute to the continuing academic discussion regarding rudeness in digital communication by redefining politeness and impoliteness as strategies that are not adversarial but rather integrated and sensitive to context. Past studies, such as those by Pasana et al. (2023) and Wu and Yang (2024), mostly examined the importance of on-record rudeness in political and consumer commentary. This study reveals that online reviews, especially those with three or four stars, often employ mixed strategies that combine politeness and impoliteness within the same text. This result aligns with Vásquez's (2011) study of TripAdvisor reviews, in which users downplayed complaints by placing them alongside recommendations and praise, thereby matching user expectations and the platform's requirements.

This study also supports the findings of Wu and Yang (2024) and Zhao (2022), which show that platform conventions and cultural discourse norms shape how people behave politely or rudely. This study does not directly examine cross-cultural data, but the presence of indirect complaints, softened FTAs, and emotionally motivated praise suggests that people are more likely to negotiate in a practical way than to face each other directly. This style differs from the more direct, confrontational tone common in Facebook political discussion (Pasana et al., 2023) or on fast-paced, socially performative sites like TikTok (Zhao, 2022), where being rude can be a way to fit in or have fun.

In short, this study adds to our understanding of digital pragmatics by showing that it is not always evident what (im)politeness means in online evaluations. Instead, user strategies are influenced by factors such as how seriously they perceive the situation, platform conventions, cultural expectations, and their own goals for the encounter. More research should examine how these hybrid expressions function across different types of user-generated content, especially in contexts where people leave both positive and negative reviews.

4.1 Severity of the situation and utilization of (im)politeness strategies

The different ways of being polite and impolite in one- to five-star evaluations are due to the interaction of three contextual characteristics that Brown and Levinson (1987) found: social distance (D), relative power (P), and ranking of imposition (R). Customers who give one-star reviews, who are the most unhappy, tend to be less courteous. If they feel they do not know the firm well or do not have any control over it, they may complain, criticize, or be rude. The perceived high imposition, which expects redress, refunds, or apologies, often justifies a more confrontational tone. On the other hand, five-star reviews, which indicate low imposition and high pleasure, favor positive politeness strategies such as praise, solidarity cues, and emotive language. This is because the connection is friendly, and the reviewer does not have to make any demands. Customers who leave four- and three-star reviews prefer a middle ground. They are courteous but use hedges or indirectness to temper their criticisms, showing they know they might hurt someone's feelings while still expressing their dissatisfaction. Customers often behave as both critics and supporters in these reviews, which shows a negotiation of power. As the scenario worsens (from five to one star), reviewers switch from methods that improve the situation to ones that worsen it. The effects of D, P, and R become increasingly clear in how they affect the language used to express (im)politeness.

The study also supports Brown and Levinson's (1987) assertion that the choice of politeness strategy is influenced by the balance of social distance (D), relative power (P), and the ranking of imposition (R). When reviews get one or two stars, people tend to feel a lot of pressure and distance from the reviewer, which makes them more likely to give direct or harsh feedback. On the other hand, 5-star reviews usually indicate that the service provider was not very demanding and more in line with the customer, leading to very positive, solidarity-based politeness. But the middle-range evaluations (three to four stars) make things more complicated: users regularly mix strategies, such as combining complaints with compliments or softening harsh comments with emojis. These evaluations show the situational ambiguity that Zhao (2022) and Locher and Watts (2005) discuss when users encounter mixed intentions such as praise, displeasure, solidarity, and criticism within the same contact.

4.2 Reconceptualizing Politeness and Impoliteness as Contextually Fluid

Brown and Levinson's (1987) traditional theories of politeness have generally seen politeness and impoliteness as separate and opposing groups. However, looking at real-world conversations, such as customer reviews, shows that this binary framing does not always capture the complexity of interactional purpose. Hybrid strategies are prevalent in numerous three- and four-star reviews. These occur when positive politeness, including praise or thanks, accompanies indirect or muted criticism, or when face-threatening activities are softened with emotional language or emojis. These instances show that reviewers often balance conflicting goals, such as expressing displeasure while still fitting in with others or giving praise while still making it clear that their expectations were not met. These mixed strategies cannot be fully understood without considering the context, including the seriousness of the issue, the social distance between the reviewer and the business (D), the perceived power relationship between them (P), and the perceived ranking of imposition of the complaint (R) (Brown & Levinson, 1987). As argued by Culpeper (2011) and Locher and Watts (2005), politeness and impoliteness should not be viewed as fixed characteristics of speech, but rather as contextually fluid and dynamically interpreted in interaction. They are evaluative and emergent. Therefore, it might be better to consider (im)politeness as a continuum, with methods changing according to intent, relationship goals, and contextual inferences, rather than putting reviews into strict categories. This new way of thinking enables a more practical, context-sensitive analysis that considers the strategic and social complexity of everyday digital conversations.

Author Contributions: The sole author designed, analyzed, interpreted and prepared the manuscript.

Funding: The author received no financial support for the research.

Conflicts of Interest: The author declares no competing interests.

Informed Consent Statement/Ethics approval: Chiang Mai University Research Ethics Committee (CMUREC) has examined documents of this research project and found that this research project does not involve human subjects. Therefore, is a Non-Human Subject Research, and does not require to be reviewed and approved research ethics by CMUREC.

Declaration of Generative AI and AI-assisted Technologies: During the preparation of this work, the author utilized ChatGPT and Grammarly to enhance language clarity, readability, grammatical accuracy, and consistency through proofreading. After using this tool/service, the author reviewed and edited the content as needed and took full responsibility for the content of the publication.