Platform-Specific Quality Dimensions in Instagram Commerce: How Social Media Features Drive Consumer Behavior

Comments and Suggestions for Author

Authors' Response

Overall, the manuscript is timely and relevant. However, there is a need to address the issue of incremental contribution and improve the methods in terms of the higher-order constructs. Not to mention the need to streamline the hypotheses development section. 

Thank you for this overall assessment and for identifying these key areas for improvement. We have addressed all three concerns as detailed below and in the subsequent specific comments.

Regarding incremental contribution: The revised manuscript more precisely delineates the theoretical gap, articulating that conventional e-service quality frameworks (E-S-QUAL, WebQual) were designed for transactional environments and are conceptually inadequate for social media commerce contexts characterized by social embeddedness, real-time interactivity, and community-driven experiences. A new subsection (Section 2.3.3, “Theoretical Integration and Model Justification”) has been added to explicitly articulate how the combination of means-end chain theory and uses and gratifications theory provides a novel and complementary theoretical architecture. Two explicit research questions are now stated in the Introduction, anchoring the study’s contribution to specific theoretical gaps. The Theoretical Implications section (5.2.1) now articulates four distinct contributions.

Regarding streamlining the hypotheses development section: Section 2.6 has been revised by removing a duplicate sentence inadvertently present at the end of Section 2.6.8, and by eliminating forward-looking references to the study’s own empirical results that had appeared within hypothesis subsections. Each subsection now follows a consistent structure: theoretical rationale grounded in means-end chain theory and uses and gratifications theory, followed by supporting prior empirical evidence, concluding with the formal hypothesis statement.

Regarding higher-order construct specification: Please see the detailed response to Comment 4 below.

Several methodological aspects should be strengthened:

1. Sampling and data collection:
   The manuscript should provide clearer details on participant recruitment, inclusion criteria, and justification of sample size.

Thank you for this valuable suggestion. We have revised Section 3.1 to provide clearer details regarding participant recruitment, inclusion criteria, sampling strategy, and the data collection period. Specifically, we now clarify that a non-probabilistic convenience and snowball sampling strategy was used, that participants were recruited through social networks (Instagram, Facebook, and WhatsApp), and that screening questions ensured respondents were at least 18 years old and had made a purchase through Instagram within the previous six months. We also specify that data collection took place between October and November 2022 and that 258 valid responses were retained after data cleaning.

Additionally, Section 3.3 now includes a formal sample size justification using G*Power. An a priori power analysis (effect size = 0.111; α = 0.05; power = 0.95; 7 predictors) indicated a minimum required sample size of 204 observations. The final sample (n = 258) exceeds this requirement, and a post hoc analysis confirmed adequate statistical power (0.986).

All modifications have been highlighted in blue in the revised manuscript.

1. Measurement transparency:
   Sources of measurement items and any adaptations made for the Instagram context should be more explicitly reported.

Thank you for this helpful suggestion. We agree that greater transparency regarding the origin of the measurement items and their contextual adaptation strengthens methodological clarity and replicability.

In the revised manuscript (Section 3.2), we now explicitly identify the source of each construct. Specifically, social media quality and its four dimensions were adapted from the SOME-Q model developed by Suryani et al. (2021). Customer service and privacy were derived from Sun et al. (2015), while brand image, electronic word of mouth (eWOM), repurchase intention, and customer satisfaction were adapted from Rita et al. (2019) and related prior research.

Additionally, we clarify that minor wording adjustments were made to contextualize the items for Instagram-based commerce (e.g., replacing general references to “online platforms” with “Instagram”), while preserving the original conceptual meaning of each construct. These modifications have been incorporated into Section 3.2 and highlighted in blue in the revised manuscript.

1. Common method bias:
   The authors should clearly state what procedural and statistical remedies were applied to assess and mitigate common method bias.

Thank you for this important comment. We agree that clarifying both procedural and statistical remedies strengthens the methodological transparency of the study.

In the revised manuscript, we have explicitly incorporated both approaches.

From a procedural perspective, several ex-ante remedies were implemented during questionnaire design and data collection. Respondent anonymity was guaranteed, participants were informed that there were no right or wrong answers to reduce evaluation apprehension, and the measurement items were carefully refined and pretested to ensure clarity and minimize ambiguity. These procedures follow established recommendations for mitigating common method variance in self-reported survey research (Podsakoff et al., 2003). This clarification has been added at the end of Section 3.1.

From a statistical perspective, we applied the full collinearity variance inflation factor (VIF) approach proposed by Kock (2015), which enables the assessment of both multicollinearity and potential common method bias in PLS-SEM models. The VIF values are now explicitly reported in Table 8. The results indicate that most VIF values are well below the conservative threshold of 3.3, while a small number slightly exceed this value (VIF = 3.406) but remain clearly below the critical threshold of 5.0. These findings suggest that common method bias does not pose a serious concern in the present study.

1. Higher-order construct specification:
   The modeling of social media quality as a higher-order construct requires clearer justification, including the type of hierarchical component model used and the estimation approach in PLS-SEM.

Addressing these points will significantly improve methodological rigor and replicability.

Thank you for this insightful and constructive comment. We agree that clarifying the specification of the higher-order construct enhances methodological rigor and replicability.

In the revised manuscript, we now explicitly state that social media quality was modeled as a reflective–reflective hierarchical component model (Type I HCM), following the classification proposed by Becker et al. (2012). This specification is consistent with the original SOME-Q conceptualization developed by Suryani et al. (2021), where the four first-order constructs (clarity, attractiveness, interactivity, and relevance) are treated as reflective dimensions forming the higher-order construct.

Furthermore, we clarify that the two-stage approach was employed in the PLS-SEM estimation procedure. In the first stage, latent variable scores of the first-order constructs were obtained, and in the second stage, these scores were used as indicators of the higher-order construct. This approach is recommended for reflective–reflective hierarchical models and improves estimation efficiency while preserving construct validity.

These clarifications have been incorporated into Section 3.3 of the revised manuscript. All modifications have been highlighted in blue to facilitate review.

Comments and Suggestions for Author

Authors' Response

Thanks for the opportunity to review this paper. Kindly note the following.

This is an interesting study, it discusses and test a trending topic of social media and its influence on different aspects of customers including satisfaction and word of mouth. While the results are significant, I don’t see a need and it is not even normal to include the statistical results in the abstract of the study. Reporting the result of the study is enough without going into too much details in the abstract. Such technical results are already reported in the analysis and discussion section.

Thank you for this observation. We agree that the abstract should convey the substantive contributions of the study without replicating the detailed statistical reporting already present in the Results and Discussion sections.

We have revised the abstract to present findings in narrative, substantive terms, replacing specific path coefficients (β) and significance values with descriptive summaries of the key results. The revised abstract now communicates the main findings (e.g., the relative importance of privacy versus customer service as antecedents of social media quality, and the role of satisfaction as a mediating mechanism) without presenting the technical output in detail.

These revisions have been incorporated and highlighted in blue in the revised manuscript.

The authors have develop a good introduction however, they still need to focus more on motives of the study? Why Chile and no other places? What have motivated them for such research? What are the specific research questions that you are trying to answer?

Thank you for this constructive observation. We have substantially revised the Introduction to address these three points.

Regarding the motivation for Chile: The revised manuscript now provides three specific justifications for the Chilean context: (1) Chile is consistently ranked among Latin America’s most digitally mature economies, with advanced telecommunications infrastructure and high social media penetration rates; (2) Instagram commerce has grown substantially in Chile, with a significant proportion of SMEs relying on the platform as their primary commercial channel; and (3) Chile’s evolving data protection regulatory landscape—particularly the 2024 update to Law 21.719, aligned with GDPR principles—makes privacy perceptions especially salient in social commerce contexts. This combination of digital maturity and growing privacy regulation positions Chile as a theoretically meaningful and timely empirical setting.

Regarding the research motivation: We now explicitly articulate the theoretical gap motivating the study: existing e-service quality frameworks were designed for transactional environments and fail to capture the social embeddedness, real-time interactivity, and community-driven characteristics that define social media commerce. This gap limits both theoretical understanding and practical guidance for SMEs operating on platforms like Instagram.

Regarding the research questions: Two explicit research questions have been added to the Introduction: (RQ1) What are the antecedent factors that shape consumer perceptions of social media quality in Instagram commerce? and (RQ2) How does social media quality influence consumer attitudes and behavioral intentions in Instagram commerce contexts?

All revisions have been highlighted in blue in the revised manuscript.

Authors are also advised to conclude the introduction section with an organization of the research to allow readers understand the flow of this research.

Thank you for this suggestion. A closing paragraph has been added to the Introduction outlining the organization of the remainder of the manuscript.

The revised Introduction now concludes with a paragraph indicating that: Section 2 presents the theoretical framework, including the evolution of service quality, the SOME-Q model, the theoretical foundations (means-end chain theory and uses and gratifications theory), and the hypotheses development; Section 3 describes the research methodology; Section 4 reports the results of the measurement model and structural model analyses; and Section 5 discusses the findings and presents the theoretical and practical implications, limitations, and directions for future research.

This addition has been highlighted in blue in the revised manuscript.

For the theoretical foundations or theory of the study, I advise combining the two theories in one section simultaneously to reduce number of sections in the article. You need also to justify the use of two theories. You may also try to combine other small sections with one big title to eliminate distractions.

Thank you for this structural recommendation. We have addressed this comment in two ways.

First: Sections 2.3.1 (Means-End Chain Theory) and 2.3.2 (Uses and Gratifications Theory) are now followed by a new integrative subsection, Section 2.3.3 (“Theoretical Integration and Model Justification”), which explicitly justifies the use of both theories in combination. This section explains that means-end chain theory provides the overarching hierarchical framework connecting antecedents (customer service, privacy) → social media quality → outcomes, while uses and gratifications theory explains why specific quality dimensions matter to users by linking each dimension to distinct gratification categories. The combined framework is shown to be theoretically complementary and mutually reinforcing rather than redundant.

Second: We have reviewed the overall section structure for conciseness. Sections 2.1 and 2.2 (service quality evolution and social media quality as a construct) have been streamlined to reduce overlap, and Sections 2.4 and 2.5 (antecedents and consequences) have been tightened to focus on the most relevant theoretical arguments supporting each hypothesis.

All revisions are highlighted in blue in the revised manuscript.

In line 303 and 304, you are not supposed to discuss your result now, your result will be shown in the discussion, and here you are still arguing the model of the study. This applies to other hypotheses.

Thank you for this important observation. The reviewer is correct that the hypotheses development section should present only the theoretical rationale and prior empirical support motivating each hypothesis, without anticipating or referencing the present study’s own findings.

We have reviewed all subsections of Section 2.6 and removed any forward-looking references to the study’s own empirical results. Each subsection now adheres strictly to the following structure: (1) theoretical argument grounded in means-end chain theory and uses and gratifications theory, (2) supporting evidence from prior literature, and (3) the formal hypothesis statement. No current study results are discussed within the hypotheses development section.

These revisions have been applied throughout Section 2.6 and are highlighted in blue in the revised manuscript.

I wonder why there was not even a single mediator in this research despite the possibility to do that or develop mediation hypotheses. Authors needs to justify that or admit this in the limitation section.

Thank you for raising this important methodological point. We acknowledge this as a genuine limitation of the study and have addressed it honestly in the revised manuscript.

In the Discussion section (Section 5.1), we clarify that while the model’s structure—with customer satisfaction positioned between social media quality and behavioral outcomes—is theoretically consistent with a mediating role, formal mediation hypotheses were not originally specified, and indirect effects were not tested via bootstrapping procedures in SmartPLS. The observed pattern of results (where satisfaction shows stronger effects on eWOM and repurchase intention than social media quality alone) is suggestive of mediation but does not constitute a formal test.

In the Limitations section (Section 5.3), we now explicitly acknowledge the absence of formal mediation testing as a limitation, noting that future research should specify and test indirect effect hypotheses using bootstrapped confidence intervals, which would provide a more rigorous examination of the mechanisms linking social media quality to behavioral outcomes through satisfaction.

These additions are highlighted in blue in the revised manuscript.

There is no need for you to keep reporting the technical result in every section. You have already reported them in table 8 and in the discussion sections. Please do not include technical results in implications section. Directly discuss your theoretical implications based on the findings and explain what values this research offers to the theories. The same thing applies to practical implications.

Thank you for this clear and valid recommendation. We agree that repeating statistical parameters (β values and significance levels) in the Implications sections is redundant and detracts from the substantive theoretical and practical discussion.

Both the Theoretical Implications (Section 5.2.1) and the Practical Implications (Section 5.2.2) have been revised to discuss findings in substantive terms without restating the path coefficients. The theoretical implications now focus on what the findings contribute to existing theory (e.g., the primacy of privacy as a quality antecedent in social commerce, the extension of MEC and U&G theories to Instagram commerce, and the confirmation of the S-O-R framework in this context). The practical implications now provide actionable guidance organized around privacy strategy, content quality investment, and post-purchase satisfaction management, without reprinting the statistical output.

All revisions are highlighted in blue in the revised manuscript.

The research methodology section needs further discussion. Please include the type of research such as quantitive or qualitative and also indicate that the sample is snowball and convenience one. Tell us the time use for collecting data, who reviewed the questionnaire, and how did you reduce the bias?

Thank you for this valuable comment. We have revised the methodology section to improve clarity and transparency regarding the research design and data collection procedures.

First, we now explicitly state that the study follows a quantitative, explanatory, cross-sectional research design (Section 3.1).

Second, we clarify that a non-probabilistic sampling approach was employed, combining convenience and snowball sampling techniques. This information has been incorporated into Section 3.1.

Third, the data collection period (October–November 2022) has been clearly specified in the revised manuscript.

Fourth, we now explicitly indicate that the questionnaire was reviewed by three academic experts in marketing and electronic commerce to ensure content validity and conceptual consistency, and that a pilot test was conducted prior to full deployment (Section 3.2).

Finally, we have elaborated on the procedural steps taken to reduce potential bias in self-reported survey research. These include guaranteeing respondent anonymity, reducing evaluation apprehension, and implementing screening questions. These clarifications have been added to Section 3.1.

All revisions have been highlighted in blue in the revised manuscript to facilitate review.

Show the CMB and VIF results.

Thank you for this suggestion. Following the reviewer’s recommendation, we now explicitly report the variance inflation factor (VIF) values within the structural model results table (Table 8). These VIF values are used to assess both multicollinearity and potential common method bias following the full collinearity approach proposed by Kock (2015).

As shown in Table 8, all VIF values remain below the critical threshold of 5.0, and most are below the conservative threshold of 3.3. These results confirm that neither multicollinearity nor common method bias threatens the validity of the structural model.

Elaborate more on the demographic of the respondents.

Thank you for this suggestion. Section 3.1 of the revised manuscript now includes an expanded discussion of the sample’s sociodemographic characteristics, presented alongside Table 1.

The narrative now highlights key features of the sample: the predominance of female respondents (85.66%), the concentration of participants in the northern region of Chile (75.58%)—which reflects the geographic distribution of the researchers’ academic networks—the predominantly young adult profile (55.04% aged 18–24 years), the high educational attainment (66.66% with university-level education), and the notable purchasing frequency, with 39.92% of respondents having made four or more purchases through Instagram in the prior six months, indicating substantial social commerce engagement.

The implications of these demographic characteristics for the generalizability of the findings are also briefly discussed, acknowledging the potential limitations of the non-probabilistic sampling strategy.

All additions are highlighted in blue in the revised manuscript.

Attach the questionnaire you have used as appendix.

Thank you for this recommendation. The complete measurement instrument, comprising all items organized by construct with their corresponding response scale and original sources, has been included as Appendix A at the end of the revised manuscript.

The appendix presents all items as administered to respondents (originally in Spanish, with an English translation provided for reporting purposes), organized by construct: Privacy (PRI, 3 items), Customer Service (CS, 3 items), Social Media Quality–Clarity (CLA, 3 items), Attractiveness (ATT, 3 items), Interactivity (INT, 4 items), and Relevance (REL, 3 items), Brand Image (BI, 4 items), Repurchase Intention (RI, 3 items), Electronic Word of Mouth (eWOM, 2 items), and Customer Satisfaction (CSAT, 3 items). All items were measured on a five-point Likert scale (1 = Strongly Disagree, 5 = Strongly Agree). Please see Appendix A in the revised manuscript.

Follow the journal guidelines for the writing styles.

Thank you for this reminder. The revised manuscript has been reviewed for compliance with the Journal of Theoretical and Applied Electronic Commerce Research (JTAER) formatting and style guidelines.

Specifically, we have verified: (1) reference formatting follows the journal’s required citation style; (2) section headings and numbering are consistent with journal conventions; (3) figure and table captions follow the prescribed format; (4) in-text citations adhere to the author-year format specified by the journal; and (5) language and academic register have been reviewed throughout for consistency and clarity.

Any deviations from the journal template identified during this review have been corrected in the revised submission.

Comments and Suggestions for Author

Authors' Response

The manuscript exhibits a sufficient degree of methodological rigor and tackles a pertinent topic within its discipline. The work offers helpful empirical insights even though it doesn't make a really original theoretical contribution. The text might satisfy publication requirements with enhancements to methodological clarity, theoretical positioning, and language quality.

1. Make the introduction stronger by outlining the research need more precisely and placing the study in the context of current literature.

Thank you for this recommendation. The Introduction has been substantially revised to more precisely articulate the research need and situate the study within current literature.

The revised Introduction now provides a sharper delineation of the theoretical gap: while e-service quality frameworks such as E-S-QUAL (Parasuraman et al., 2005) and WebQual (Loiacono et al., 2002) constitute foundational contributions to electronic commerce scholarship, they were designed for transactional environments emphasizing system reliability, fulfillment efficiency, and security protocols. These models fail to capture the social embeddedness, real-time interactivity, and community-driven experiences that define social media commerce. Despite growing scholarly attention to social commerce, platform-specific quality frameworks that account for these distinctive characteristics remain underdeveloped in the literature.

Current literature is engaged more directly through the incorporation of recent references (2023–2025) documenting the rapid growth of Instagram commerce, the increasing salience of privacy concerns in social platforms, and the inadequacy of existing quality models for social media contexts. Two explicit research questions have been added to anchor the study’s contribution to these identified gaps.

The Introduction now concludes with a paragraph outlining the structure of the remainder of the manuscript. All revisions are highlighted in blue.

2. To ensure reproducibility, enlarge the methods section to include more information on sample characteristics, data collecting, and analytical techniques.

Thank you for this important recommendation for reproducibility. The Methods section (Section 3) has been substantially expanded across all three subsections.

In Section 3.1 (Instrument Design and Data Collection), we now explicitly state: the quantitative, explanatory, cross-sectional research design; the non-probabilistic convenience and snowball sampling strategy; participant eligibility criteria (aged 18 or older, at least one Instagram purchase in the prior six months); the data collection period (October–November 2022); the distribution channels (Google Forms via Instagram, Facebook, and WhatsApp); questionnaire review by three academic experts in marketing and electronic commerce; a pilot test prior to full deployment; and procedural bias-reduction measures (respondent anonymity, evaluation apprehension reduction, screening questions).

In Section 3.2 (Measurements), we now explicitly identify the source of each construct’s scale, describe the minor wording adaptations made for the Instagram commerce context, and provide an expanded demographic description of the final sample (n = 258).

In Section 3.3 (Data Analysis), we now detail the PLS-SEM estimation approach, including the reflective–reflective Type I HCM specification for social media quality, the two-stage estimation procedure, bootstrapping with 5,000 subsamples, the a priori G*Power analysis, and the full collinearity VIF approach for common method bias assessment.

All additions are highlighted in blue in the revised manuscript.

3. Check the work for academic tone and language clarity, fixing any minor grammar mistakes and enhancing sentence structure.

Thank you for this observation. A thorough language revision has been conducted throughout the manuscript.

Particular attention has been given to the Abstract, Introduction, hypotheses development (Section 2.6), Discussion, and Implications sections, where complex sentence structures have been simplified, run-on sentences divided, passive constructions reviewed for clarity, redundant phrases eliminated, and academic register elevated throughout.

Specific issues addressed include: inconsistent use of tenses across the manuscript, ambiguous pronoun references, overly long and complex sentences in the theoretical framework, and occasional informal phrasing in the Discussion. The revised manuscript presents a more consistent and polished academic style throughout.

We note that the core meaning of the manuscript was not altered by the language revision; all edits were confined to style, clarity, and grammatical correctness. The revised sections are highlighted in blue.

4. Take into account going into further detail about the findings' theoretical and practical ramifications.

Thank you for this suggestion. Both the Theoretical Implications (Section 5.2.1) and the Practical Implications (Section 5.2.2) have been substantially expanded.

The Theoretical Implications now articulate four distinct contributions: (1) the study advances e-service quality theory by demonstrating that privacy has become the dominant antecedent of quality perceptions in social commerce contexts, surpassing customer service—a finding that calls for revision of traditional quality models; (2) the empirical validation of the SOME-Q framework (Suryani et al., 2020) in the Instagram commerce context extends its applicability and confirms the construct’s nomological validity; (3) the results provide support for the stimulus–organism–response (S-O-R) framework in social commerce, with satisfaction confirmed as the psychological mechanism linking quality perceptions to behavioral outcomes; and (4) the Chilean context contributes to the underrepresented literature on social commerce in emerging digital markets in Latin America.

The Practical Implications have been reorganized around three actionable themes: (1) privacy as a strategic priority—firms should invest in transparent data governance, visible privacy assurances, and consent mechanisms as foundational elements of their social media quality strategy; (2) content quality investment—systematic attention to clarity, attractiveness, interactivity, and relevance of Instagram content is essential for building brand image and customer satisfaction; and (3) post-purchase satisfaction management—since satisfaction is the primary driver of eWOM and repurchase intention, firms should prioritize follow-up communication, responsive after-sales support, and post-purchase engagement strategies.

All additions are highlighted in blue in the revised manuscript.

Comments on the Quality of English Language

The meaning of the document is evident throughout and is largely comprehensible. Nonetheless, there are a few places where the English may be changed to increase readability, clarity, and academic style. There are sporadic instances of unclear terminology, strange phrase structures, and grammatical errors. To enhance the overall quality of expression and guarantee that the research contributions are expressed more effectively, a thorough language revision, ideally by a fluent or native English speaker, is advised.

Thank you for this constructive feedback on the quality of the English. We acknowledge that the original submission contained instances of non-standard phrasing and grammatical inconsistencies that affected the academic quality of the presentation.

In response, a comprehensive language revision has been conducted throughout the manuscript. Key improvements include: correction of grammatical errors (subject-verb agreement, article usage, tense consistency), restructuring of complex or awkward sentence constructions, replacement of ambiguous or informal terminology with precise academic language, and improvement of paragraph cohesion and transition quality across all sections.

The revisions have been particularly thorough in the Abstract, Introduction, Theoretical Framework, and Discussion sections, which are the most visible sections for editorial and reviewer assessment. We are confident that the revised manuscript presents a substantially improved level of English language quality.

Comments and Suggestions for Author

Reviewer 3: Cut down on small repetitions of comparable arguments in the Introduction and Hypotheses sections.

Authors: Thank you for this observation. We have removed the redundant theoretical opening paragraphs from all eight hypothesis subsections (Sections 2.6.1–2.6.8). Each subsection previously began with one or two paragraphs re-explicating the mechanisms of means-end chain theory and uses and gratifications theory in relation to the specific hypothesis, even though the integrated theoretical rationale was already fully developed in Section 2.3.3. These paragraphs have been deleted, and each subsection now proceeds directly to the empirical evidence supporting the hypothesis. No argumentative content has been lost; the revision eliminates only redundancy. The affected sections are highlighted in blue in the revised manuscript.

Reviewer 3: Standardize decimal formatting (if journal style calls for it, use periods in place of commas).

Authors: Thank you. All decimal notation has been standardized to periods throughout the manuscript, including in the body text and in Tables 3, 7, and 8, where comma-formatted values appeared in the prior version.

Reviewer 3: For those who are not as familiar with hierarchical PLS modeling, provide a brief description of the two-stage HCM.

Thank you for this suggestion. We have expanded the relevant passage in Section 3.3 to make the two-stage approach accessible to non-specialist readers. The revised text now reads as follows:

"The two-stage approach to PLS-SEM was employed to estimate the hierarchical model. In the first stage, a standard PLS model was estimated using the original measurement items of the four first-order constructs (clarity, attractiveness, interactivity, and relevance), and their latent variable scores were obtained. In the second stage, these latent variable scores were used as manifest indicators of the second-order social media quality construct, effectively reducing model complexity while preserving the conceptual structure of the hierarchy. This procedure is preferred over the repeated indicators approach because it avoids artificial inflation of the higher-order construct's explanatory power and produces more stable parameter estimates (Becker et al., 2012; Hair et al., 2014)."

Reviewer 3: Discussion: Why customer service was less influential than privacy.

We found this a genuinely interesting question to engage with, because the gap between the two coefficients (β = 0.537 vs. β = 0.350) is substantial enough to warrant more than a passing comment. We have added a paragraph to Section 5.1 arguing that the asymmetry reflects a functional difference between the two variables rather than a simple difference of degree. Privacy functions as an enabling condition; it must be in place before a consumer is willing to engage meaningfully with any other aspect of the platform. When users are uncertain about how their data will be used, that concern acts as cognitive interference, suppressing the evaluation of everything else, including service quality. Customer service, in contrast, is reactive and conditional: it enhances the experience for users who are already comfortable enough to engage, but it cannot compensate for a foundational trust deficit. This logic is consistent with the hierarchical structure proposed by means-end chain theory, and it also maps onto the specific nature of social commerce, where the commercial transaction is embedded within a social environment not originally designed for it, which makes data collection feel more intrusive and privacy concerns more salient than in a conventional e-commerce context.

Reviewer 3: Discussion: Particular management ramifications for SMEs in developing nations.

The prior version addressed SMEs in passing, but did not really engage with their specific constraints. We have added a dedicated closing paragraph to Section 5.2.2 that takes this more seriously. The core argument is that SMEs cannot invest simultaneously in all fronts; privacy infrastructure, content production, and service responsiveness all compete for the same limited resources. The findings provide a clear basis for sequencing: privacy-related investments should come first, because without consumer confidence in data handling, returns on content and service investment are reduced. For SMEs operating in Chile, where Law 21.719 has recently established GDPR aligned data protection requirements but enforcement infrastructure is still being built, proactive privacy adoption is not only a compliance matter but also a way to differentiate from competitors who are slower to adapt. The paragraph includes practical low-cost examples relevant to small Instagram sellers, such as plain-language privacy statements in profile highlights and explicit communication about data use in transactional messages.

Reviewer 3: Comments on the Quality of the English Language

High academic English proficiency is shown in the manuscript. The text is clear conceptually, precisely, and formally. Small enhancements could be achieved by: • Cutting down on the amount of language used to support hypotheses. • To make lengthy paragraphs easier to read, they should be slightly condensed. • Making sure that terms are stylistically consistent (for example, "social media quality" versus "quality of social media").
There is no need for significant linguistic editing.

The repetition of the hypothesis is addressed under Recommendation 1 above. Regarding paragraph length, we identified two passages with redundant phrasing: the S-O-R discussion in Section 5.2.1 and the paragraph on social media quality dimensions in Section 5.2.2. Both have been trimmed by removing internally circular sentences, the kind that effectively restate what the paragraph has just argued in slightly different words. On terminology, we have gone through the manuscript and standardized to "social media quality" throughout, replacing the variant "quality of social media" that appeared in several locations. We also corrected a section numbering error (what was labelled "6.5" is now correctly "2.6.5") and resolved an inadvertent duplication in Section 3.3 where two sentences describing the two-stage approach appeared twice.