Review Reports - The Relevance of AI-Driven Marketing Through Design Thinking for a Strategic Approach to Resilient Problem-Solving and Communication for SMEs

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The submitted article deals with a very topical issue. The author correctly identifies a knowledge gap: although AI in marketing is becoming increasingly common, there is no framework for sustainably implementing these technologies with an emphasis on human-oriented solutions. Nevertheless, it would be appropriate to explain more clearly how his approach differs from previous studies. The research lacks a critical discussion of which claims in the literature are controversial and what their limitations are. The sample (N=3) is tiny, which significantly limits the generalizability of the results. The article declares a qualitative approach, but it contains several quantitative KPI tables. These are presented as pre/post comparisons, but without control groups or statistical tests of causality. Other factors may therefore influence the results. The interpretations of the results are sometimes overly optimistic—for example, the claim that DT+AI integration clearly leads to higher ROI is not supported by robust statistics. The discussion only minimally reflects negative results (SME3 with adoption problems is mentioned but not analysed in depth). The article has high practical value and is an interesting contribution to the discussion on innovation in SME marketing. However, from a methodological point of view, it is less well-founded and its quantitative results are illustrative rather than rigorous.

Recommendations:

Deepen the literature discussion – explain more clearly how the study differs from existing research.

Methodological transparency – explain the limits of pre-/post KPI comparisons and avoid causal claims without control data.

Balance the discussion by discussing the negative case (SME3) in more detail and identifying the risks of failure.

Future research should propose quantitative or longitudinal studies to more clearly confirm the robustness of the results.

Author Response

Comments 1: Deepen the literature discussion – explain more clearly how the study differs from existing research.

Response 1: Thank you for pointing this out. I do agree with you as there is no a sustainable implementation and execution framework as AI continues to develop. Due to this, this research does present an opportunity for DT based framework, which is explored here. Based on your suggestions, I made following modification.

First modification could be found in 1. Introduction paragraph 3. The whole paragraph is posted below, and modification is highlighted.

“This study addresses an essential gap by exploring the potential of integrating the Design Thinking philosophy into AI-driven marketing strategies. Design Thinking, emphasizing human-centric problem-solving, creativity, and iterative development, presents a unique perspective that complements AI's analytical abilities [4]. By combining these approaches, SMEs may improve their marketing effectiveness and customer engagement, drive innovation, and gain competitive differentiation. Hence, this research proposes an applicable framework built on the empathy–define–ideate–prototype–test approach across data sourcing, and experimentation, which is not discussed in the existing literature. “

Second modification could be found in 5. Conclusion, paragraph 3. The whole paragraph is posted below, and modification is highlighted.

“The results emphasized the constructive collaboration between design thinking components and AI-driven marketing capabilities like personalization, automation, and data-driven decision-making. SMEs using this integrated approach can achieve superior customer interaction, offering a competitive advantage in vibrant business conditions. The proposed model emphasized the need for user-focused strategies, adaptability, and iterative experimentation to maximize the impact of AI-driven marketing strategies. Moreover, the contribution is a practical, as it proposes end-to-end integration framework that refocuses AI marketing from “how well to predict” to “what is worth predicting,” facilitating more rapid learning cycles and more accountable results.”

Comments 2. Methodological transparency – explain the limits of pre-/post KPI comparisons and avoid causal claims without control data.

Repones 2: Thank you for pointing this out. I accept this recommendation as I acknowledge that I had to make the statement about this in limitations section, under methodology. The modification can be found under 3. Methodlogy, 3.8. Limitations and Ethical considerations, the last paragraph.I also posted the statement below.

“Lastly, the author acknowledges limitations regarding pre–post KPI comparisons. Notably, these types of comparisons are descriptive and cannot support causal claims because it may lack counterfactual data. Moreover, apparent “lifts” are easily mistaken for macro shocks. Furthermore, without a suitable control, differences may demonstrate background trends rather than the actual impact of the treatment.”

Comments 3. Balance the discussion by discussing the negative case (SME3) in more detail and identifying the risks of failure.

Response 3. Thank you for pointing this out. This was very helpful tip. I have accepted this suggestion and have modified the results section. The modification can be found under 4. Discussions, paragraph 5. I also posted the statement below.

“All three SMEs acknowledged the importance of prioritizing user needs in developing AI-driven marketing solutions, showing customer-centricity. They also emphasized the value of interdisciplinary alliances in implementing AI-driven marketing projects. Besides, creativity and experimentation were facilitated across all SMEs to innovate and offer continuous improvements for integration. Moreover, across SMEs, the availability of resources and expertise for AI implementation was a significant factor. Moreover, internal resistance and organizational readiness were highlighted. Lastly, from tables 5 and 6, it was revealed that SME3 failed to apply DT outcomes into operational decisions for AI-driven marketing. The direct failures were an insight-to-action gap and siloed incentives that blocked cross-functional implementation. Moreover, the subsequent risks were associated with mis-targeted promotions that threatened margins, data privacy exposure, vendor lock-in, and backlash against brick-and-mortar operations when recommendations ignored inventory and labor constraints.”

Comments 4. Future research should propose quantitative or longitudinal studies to more clearly confirm the robustness of the results.

Response 4. Thank you for pointing this out. I have accepted this suggestion and have modified the future research statement which is more clear now. The modification can be found under 5. Conclusions, paragraph 5. I also posted the statement below.

“Longitudinal research is recommended. The future research should explore employee perspectives on integrating design thinking and AI-drive marketing, by quantitatively measuring customer engagement, impact of adaption and readiness and employee satisfaction. The longitudinal research demonstrates long-term impact and offers actionable recommendations.”

Reviewer 2 Report

Comments and Suggestions for Authors

Attached , you can find my comments !

Comments for author File: Comments.pdf

Author Response

First of all, I am greatful for comments, suggestions and recommendations.

Comments 1: The limitation of this study is its relatively modest theoretical foundation, as out of a total of 47 references in the text, only 2 are books or volumes, while the rest are short articles

Response 1: Thank you for pointing this out. This is a great observation, and I did expect that I might get asked about this. I have performed a systematic literature review and have been very selected on articles as I wanted concrete ones addressing AI-driven marketing, DT in Marketing, and lastly show intersection. The focus was very narrow but effective as it allowed me to highlight the empirical research and explore the research gap more profoundly. The modification can be found in Apendix section. I added Appendix C to demonstrate systematic literature review.

Comments 2: The authors state in the Abstract and in the content of the study that they selected/analyzed 3 SMEs from 11 companies intensive in digital technologies; Table 1 on p. 7 should probably be included at the end of section 3.2, pp. 5-6, because in its current form it is not clear what the authors want to show. In any case, under Table 1, but also in the following tables, the authors should indicate the source. In addition, I recommend that the authors include in subsection 3.2 some general information about the 11 companies initially included in the study, such as: - average number of employees; - country; - region/locality; - some information from the history of each company; - competitive position on the market; - other similar information.

Response 2: Thank you for pointing this out. Even though I agree regarding the detailed information, I have signed consent form not to disclose details of the companies and sizes. All I could confirm that all three SMEs fall in the threshold that I had. Please don’t judge me on this as it took me some time to convince SMEs to allow to perform research.

Regarding the table 1. I believe it is in the right position on the table as it demonstrates the stepwise process how the selection occurred.

Comments 3. The numbering of sections 3. Research Methodology, p. 5, is incorrect, as the results are also numbered 3; the other two sections must also be renumbered.

Response 3: Thank you for pointing this out. This has been fixed and renumbered.

Comments 4. In section 3 Research Methodology, I suggest that the authors formulate 2-3 Research Questions and a flowchart of the entire study to provide a clear picture of the entire study.

Response 4: Thank you for pointing this out. I have accepted your suggestion and posted a research question under 3. Research methodology, section 3.1 “At last, the author posted the main research question, to what extent does integrating Design Thinking into AI-driven marketing improve SME marketing effectiveness, and through which mechanisms?”

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors
It was with great pleasure that I received the invitation to review your manuscript.
However, I have a few comments to make:

• What is the main question addressed by the research? The main issue has to do with the philosophy of Design Thinking applied to marketing strategies.
• Do you consider the topic original or relevant to the field? Does it
address a specific gap in the field? Please also explain why this is/ is not
the case. This topic is relevant in today's world.
• What does it add to the subject area compared with other published
material? This manuscript sheds new light on the application of Design Thinking philosophy to marketing. • What specific improvements should the authors consider regarding the
methodology? The methodology is very clear to any reader of the manuscript. They should only make the following correction: the authors say that it is a qualitative study, but when they use descriptive statistics, more specifically proportions, the study should also be considered quantitative.
• Are the conclusions consistent with the evidence and arguments presented
and do they address the main question posed? Please also explain why this
is/is not the case. The conclusions are consistent with the analyses performed. However, the authors do not present limitations or practical and theoretical implications. They should add them.
• Are the references appropriate?

The references are appropriate.
• Any additional comments on the tables and figures.

The results presented in Tables 4 and 5 would be more interesting if they included statistical tests to compare proportions.

My Best Regards

Author Response

I am greatful for the such a feedback and recommendations. As i believe there is a space for learning. Please see my comments. I do agree with all of them, and I did my best to comment and make some adjustments where was applied.

Comment 1:What is the main question addressed by the research? The main issue has to do with the philosophy of Design Thinking applied to marketing strategies.

Response 1: Thank you for the comment. Indeed, this work studies how DT philosophy could be integrated, therefore, offering the potential framework for use. At the moment, there is not such framework.

Comment 2: Do you consider the topic original or relevant to the field? Does it
address a specific gap in the field? Please also explain why this is/ is not
the case. This topic is relevant in today's world.
Response 2: Thank you for the comment. Yes, as long as AI is progressing, we shall search for the tools and framework that could be used to better assist marketing specialists and practicioners.

Comment 3: What does it add to the subject area compared with other published
material? This manuscript sheds new light on the application of Design Thinking philosophy to marketing.

Response 3: Thank you for the comment. Especially, introducing the framework that has a future potential for sustainable use for SMEs and practitioners.

Comment 4: What specific improvements should the authors consider regarding the
methodology? The methodology is very clear to any reader of the manuscript. They should only make the following correction: the authors say that it is a qualitative study, but when they use descriptive statistics, more specifically proportions, the study should also be considered quantitative.
Response 4: Thank you for the comment and observation. Indeed, this is qualitative research using case study approach. I only used the quantitative data to show Pre/Post measurements. There was qualitative way to show but it will not have a rigor. Therefore, I used the metrics approach. I hope this answers you question.

Comment 5: Are the conclusions consistent with the evidence and arguments presented
and do they address the main question posed? Please also explain why this
is/is not the case. The conclusions are consistent with the analyses performed. However, the authors do not present limitations or practical and theoretical implications. They should add them.

Response 5: Thank you for the comment. I have accepted the recommendation and added and enhanced the statement on limitation. “The qualitative case study method was conducted with rigor using systematic protocols, triangulation, and evidence coding, which provided credibility and analytic depth. However, subsequent quantitative testing would have strengthened external validity and causal assumption.

“

Comment 6. Are the references appropriate?

The references are appropriate.
• Any additional comments on the tables and figures.

The results presented in Tables 4 and 5 would be more interesting if they included statistical tests to compare proportions.

Response 6: Thank you for the comment. Absolutely agree, but for this work, I would resisted added more statistical information, as I need more of them for the follow up research. I currently used only 2-3 statements, but I prefer not to change this qualitative research to the mixed method.

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript tackles a topical and practically relevant intersection and offers a detailed, practitioner-facing narrative of how teams attempted to orchestrate processes, governance and experimentation. As an exploratory account it has merit. At present, however, the scholarly core is not yet aligned with the claims the paper wishes to make. The research question is stated in broad terms and key constructs “integration”, “marketing effectiveness”, and especially “sustainability”, are not sufficiently defined or operationalised. “Sustainable” tends to be used as a descriptive adjective for durable or efficient results rather than as a measurable ESG outcome; this gap affects the argument from the title onwards and leaves the reader unsure what, precisely, has been demonstrated. Methodologically, the multiple-case design and the triangulation of interviews, internal artefacts and performance indicators are coherent choices for an exploratory study. Where the design becomes brittle is in the evaluation of effects. The pre–post comparisons over short windows, without a counterfactual, cannot address seasonality, exogenous shocks or serial correlation; the use of Hedges’ g on monthly KPI series compounds the problem by implying effect magnitudes under assumptions that are unlikely to hold. The qualitative analysis is carefully described, but the paper reports a target inter-coder reliability rather than the values actually achieved, and the sampling through professional networks creates a favourable selection that should be acknowledged with more force. The permission given to companies to review factual passages is understandable from an access perspective, yet it also raises the possibility of desirability bias and should be discussed as a limitation. The results sections are readable and contain specific numbers that will interest practitioners, but they place very different funnels - e-commerce, product-led software and physical retail - on the same plane, which blurs important sectoral differences. Longer KPI histories, uncertainty intervals and an explicit treatment of seasonality would make the patterns more credible. If the ambition is to argue for improvement attributable to the integration of Design Thinking and AI, a strengthened design interrupted time series, difference-in-differences or synthetic controls will be required. If, instead, the intention is to remain strictly exploratory, then the language of the claims should be moderated to avoid suggestions of statistical significance or causal impact. The manuscript would benefit from a tighter engagement with the central literatures on evaluation of managerial interventions and AI-enabled marketing, pruning peripheral citations and clarifying precisely how the paper extends or challenges current knowledge. Minor issues of English usage and consistency remain; a careful copy-edit in UK spelling, along with standardising terminology and correcting small factual slips, would improve readability. In its current form I see clear practical value - particularly in the governance routines and process descriptions - but the scholarly contribution is not yet secured. A major revision should either (i) operationalise sustainability and reinforce the evaluation design so that the empirical claims match the ambition of the title, or (ii) reframe the study as an exploratory practice paper with moderated claims, making mechanisms and contextual contingencies the central contribution. Either path would sharpen the argument and bring the paper into alignment with the standards of inference expected by the journal.

Author Response

Comment 1: The manuscript tackles a topical and practically relevant intersection and offers a detailed, practitioner-facing narrative of how teams attempted to orchestrate processes, governance and experimentation. As an exploratory account it has merit. At present, however, the scholarly core is not yet aligned with the claims the paper wishes to make.

Response 1: Thank you for the comment. The major objective of this was to explore the extent DT philosophy can be integrated in Ai-driven marketing and offer a sustainable framework for the use. I have attempted to demonstrate it with almost 18 month work. I believe this work delivers practical value, as well as scholar value for the future discourse, especially when quantitative research is performed. I have addressed this limitation in the conclusion “The qualitative case study method was conducted with rigor using systematic protocols, triangulation, and evidence coding, which provided credibility and analytic depth. However, subsequent quantitative testing would have strengthened external validity and causal assumption.:

Comment 2: The research question is stated in broad terms and key constructs “integration”, “marketing effectiveness”, and especially “sustainability”, are not sufficiently defined or operationalised. “Sustainable” tends to be used as a descriptive adjective for durable or efficient results rather than as a measurable ESG outcome; this gap affects the argument from the title onwards and leaves the reader unsure what, precisely, has been demonstrated.

Response 2: Thank you for the comment. I understand this standpoint, but my work is not related directly to ESG. I wont be able to respond further as my research doesnot address ESG. The objective was to to explore the extent DT philosophy can be integrated in Ai-driven marketing and offer a sustainable framework for the use.

Comment 3: Methodologically, the multiple-case design and the triangulation of interviews, internal artefacts and performance indicators are coherent choices for an exploratory study. Where the design becomes brittle is in the evaluation of effects.

Response 3: Thank you for the comment. I do agree that qualitative case study methods to lack statistical generalizability, but the 18 month work with metrics, and data creates a strong value. I have addressed this in limitations second under subsection 3.8 as well as in conclusion.

Comment 4: The pre–post comparisons over short windows, without a counterfactual, cannot address seasonality, exogenous shocks or serial correlation; the use of Hedges’ g on monthly KPI series compounds the problem by implying effect magnitudes under assumptions that are unlikely to hold.

Response 4: Thank you for the comment. Yes, the assumptions and results would change in the future as AI-driven marketing continues to grow however, DT framework allows to control results, and create future predictions. Again, I am thankful for your comments on this.

Comment 5: The qualitative analysis is carefully described, but the paper reports a target inter-coder reliability rather than the values actually achieved, and the sampling through professional networks creates a favourable selection that should be acknowledged with more force.

Response 5: Thank you for the comment and advise. The only way to connect to business will be using the professional network. However, as my paper states, I took me times and process to obtain all permission before I could even do anything.

Comment 6: The permission given to companies to review factual passages is understandable from an access perspective, yet it also raises the possibility of desirability bias and should be discussed as a limitation.

Response 6: Thank you for this. For goodness and trustworthiness, every data obtained from report reviews, and interviews, must be communicated to providers so they can check validity and statements (I I correctly presented it) and use afterwards. I am just following the qualitative research rules.

Comment 7:The results sections are readable and contain specific numbers that will interest practitioners, but they place very different funnels - e-commerce, product-led software and physical retail - on the same plane, which blurs important sectoral differences.

Response 7: Thank you for this comment. I have nothing to add here.

Comment 8: Longer KPI histories, uncertainty intervals and an explicit treatment of seasonality would make the patterns more credible. If the ambition is to argue for improvement attributable to the integration of Design Thinking and AI, a strengthened design interrupted time series, difference-in-differences or synthetic controls will be required. If, instead, the intention is to remain strictly exploratory, then the language of the claims should be moderated to avoid suggestions of statistical significance or causal impact.

Response 8: Thank you for the suggestion. This is very valuable and I recognize it. I have made changes to the paper, where I also recommend future research using quantitative results to explore correlations and regressions. I do have statistical data, which is going to be used in another article. But I agree your statements and make me more confident for my next work.

Comment 9: The manuscript would benefit from a tighter engagement with the central literatures on evaluation of managerial interventions and AI-enabled marketing, pruning peripheral citations and clarifying precisely how the paper extends or challenges current knowledge.

Response 9: Thank you for this. I actually added appendix c to show how literature was collected and used.

Comment 10: Minor issues of English usage and consistency remain; a careful copy-edit in UK spelling, along with standardising terminology and correcting small factual slips, would improve readability.

Response 10: I have no comment on this.

Comment 11: In its current form I see clear practical value - particularly in the governance routines and process descriptions - but the scholarly contribution is not yet secured.

Response 11: thank you for this. I have modified paper with all suggestions given from all reviewers and my paper now would be good.

Comment 12: A major revision should either (i) operationalise sustainability and reinforce the evaluation design so that the empirical claims match the ambition of the title, or (ii) reframe the study as an exploratory practice paper with moderated claims, making mechanisms and contextual contingencies the central contribution. Either path would sharpen the argument and bring the paper into alignment with the standards of inference expected by the journal.

Response 12: This has been addressed.

Thank you again for all the suggestions. I have learned a lot and also receive motivation to continue working on my next paper using quant data which I have colleted.

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors
Thank you for making the changes I suggested.
My best regards

Author Response

Comment 1: Dear Authors, Thank you for making the changes I suggested.My best regards

Response 1: Thankful for the suggestions and advice. Feedback was helpful for improving not just the paper, but also learning.

Reviewer 4 Report

Comments and Suggestions for Authors

Thank you for your responses and for the work invested in the revision. While I appreciate the effort, the replies do not resolve the methodological and conceptual issues flagged in the first round, and in several instances they decline to engage with the core points. Below I explain, comment by comment, why the manuscript in its current form still falls short of the journal’s standards.

C1 – Alignment between claims and scholarly core.
Your reply reiterates the paper’s practical intent and notes that the study is exploratory, but it does not show how the claims have been realigned with the evidence. The manuscript still reads as if it aims to demonstrate effectiveness rather than to generate propositions. This needs either a reframing of the contribution or stronger evidence; at present, neither is provided.

C2 – Definition/operationalisation of “integration”, “marketing effectiveness” and “sustainability”.
You state that the work is not about ESG and therefore cannot address this point, yet the paper continues to use “sustainable” in a way that invites an ESG interpretation. Either operationalise the construct with measurable outcomes or remove the term and replace it with language that matches what is actually assessed (e.g., durability, governance robustness). As it stands, the claim remains ambiguous and unsupported.

C3 – Exploratory design vs. evaluation of effects.
Acknowledging the lack of statistical generalisability does not address the concern about how effects are evaluated. The manuscript still presents outcome contrasts as if they evidenced improvement attributable to DT×AI integration. If the study is exploratory, these should be explicitly presented as descriptive patterns, not effects.

C4 – Pre–post windows, absence of counterfactual, and Hedges’ g.
The response does not commit to removing the effect-size calculations or to introducing a design that can handle seasonality and shocks. Stating that “DT framework allows to control results and create future predictions” is not a methodological remedy. Either withdraw the causal language and all effect-size claims or redesign the evaluation (e.g., ITS/DiD/synthetic controls) with sufficient data. Without that, the current analyses are not acceptable.

C5 – Inter-coder reliability and selection bias.
You justify recruitment via professional networks but do not report the achieved inter-coder reliability or meaningfully acknowledge the favourable selection introduced by access routes. The paper needs the actual coefficients (e.g., Cohen’s κ) and a frank discussion of how sampling affects external validity.

C6 – Company fact-checking and desirability bias.
“Member checking” for factual accuracy is acceptable, but the reply asserts it as a rule rather than engaging with the bias it may introduce. The manuscript should document the scope of company review, safeguards used, and why residual desirability bias is unlikely (or must be accepted as a limitation).

C7 – Heterogeneous funnels pooled together.
You indicate you have “nothing to add”. This is not sufficient. Results need to be restructured by sector, with claims limited to within-sector patterns and any cross-sector statements clearly identified as propositions contingent on context.

C8 – Longer KPI histories, uncertainty intervals, and claim moderation.
Thank you for recognising the point; however, moving the quantitative agenda to “another article” leaves this manuscript still making claims it cannot support. If causal claims are removed and uncertainty communicated (intervals, seasonality discussion), this should be visible in the current text. Otherwise, the issue remains unresolved.

C9 – Engagement with the central literatures.
Adding an appendix on how literature was collected does not substitute for a tighter, critical synthesis that positions the paper relative to evaluation of managerial interventions and AI-enabled marketing. The manuscript still needs pruning of peripheral citations and a precise statement of how it extends current knowledge.

C10 – Language and consistency.
“ No comment” suggests the copy-edit has not been undertaken. The manuscript requires UK spelling, consistent terminology, and correction of factual slips.

C11 – Practical value vs. scholarly contribution.
Stating that the paper is now “good” does not address the misalignment between claims and evidence. Either reframe the contribution as practice-oriented (mechanisms, governance routines, contingencies) or upgrade the evaluation design; at present, neither path is implemented.

C12 – Choice between two revision paths.
The response says “addressed” but does not specify which path was taken or how. From the replies and the text as described, the sustainability construct is still un-operationalised and the evaluative design remains unchanged. Consequently, the manuscript continues to over-claim relative to its evidence base.

Comments on the Quality of English Language

The English must be improved before the manuscript can be considered further. Please arrange a professional copy-edit. Recurrent issues include inconsistent spelling and terminology, non-idiomatic phrasing, grammatical errors (subject–verb agreement, article use, prepositions), punctuation and spacing mistakes, duplicated words, and inconsistent hyphenation and capitalisation. Choose one variety of English (UK or US) and apply it consistently. Replace non-idiomatic expressions (for example, “attainment to the fair” should be “attendance at the fair”). Maintain tense discipline (past for Methods and data collection; present for general statements). Define acronyms at first use and use them consistently. Ensure figure and table captions are self-contained, labels match the text, and the basis for percentages is stated. These language corrections are required prior to acceptance.

Author Response

C1 – Alignment between claims and scholarly core. - Your reply reiterates the paper’s practical intent and notes that the study is exploratory, but it does not show how the claims have been realigned with the evidence. The manuscript still reads as if it aims to demonstrate effectiveness rather than to generate propositions. This needs either a reframing of the contribution or stronger evidence; at present, neither is provided.

Reponses 1: After I consulted with my fellow researcher colleague, the statement on in abstract on design of research had to be modified. This is also modified throughout the paper. Therefore, I changed it to “The author used a qualitatively driven mixed data case study design” as is represents the work better. This way, qualitative research highlights the meaning, depth, and context that I bring to discourse. It underscores the quantitative metrics that I provide, delivering structure and magnitude that complement the qualitative insights. This blend will be a mixed-data case study, which, in research, is sometimes called a qualitatively driven mixed methods design.

C2 – Definition/operationalisation of “integration”, “marketing effectiveness” and “sustainability”. - You state that the work is not about ESG and therefore cannot address this point, yet the paper continues to use “sustainable” in a way that invites an ESG interpretation. Either operationalise the construct with measurable outcomes or remove the term and replace it with language that matches what is actually assessed (e.g., durability, governance robustness). As it stands, the claim remains ambiguous and unsupported.

Reponses 2: Thank you note. I accept this, and I chose to use word “resilient” as it addresses and focuses on adaptability and strength over time, which also supports my statements on integration and effectiveness.

C3 – Exploratory design vs. evaluation of effects. - Acknowledging the lack of statistical generalisability does not address the concern about how effects are evaluated. The manuscript still presents outcome contrasts as if they evidenced improvement attributable to DT×AI integration. If the study is exploratory, these should be explicitly presented as descriptive patterns, not effects.

Reponses 3: Thank you for the note. I have made adjustments. I am modified information for qualitative mixed-method case study analysis.

C4 – Pre–post windows, absence of counterfactual, and Hedges’ g. - The response does not commit to removing the effect-size calculations or to introducing a design that can handle seasonality and shocks. Stating that “DT framework allows to control results and create future predictions” is not a methodological remedy. Either withdraw the causal language and all effect-size claims or redesign the evaluation (e.g., ITS/DiD/synthetic controls) with sufficient data. Without that, the current analyses are not acceptable.

Reponses 4: In Pre-Post summaries I am only using marketing metrics to show impacts and demonstrate results of observation. Therefore, I have removed Hedges’ g from the writing.

C5 – Inter-coder reliability and selection bias. - You justify recruitment via professional networks but do not report the achieved inter-coder reliability or meaningfully acknowledge the favourable selection introduced by access routes. The paper needs the actual coefficients (e.g., Cohen’s κ) and a frank discussion of how sampling affects external validity.

Reponses 5: I have looked to my transcribed data, and I was able to prepare paragraph statement addressing Cohen’s k. Thank you for this valuable advice.

Reponses 6: I have all the documented and based on that I am showing inclusion and exlusion criteria, later showin in stepwise selection process.

C7 – Heterogeneous funnels pooled together.- You indicate you have “nothing to add”. This is not sufficient. Results need to be restructured by sector, with claims limited to within-sector patterns and any cross-sector statements clearly identified as propositions contingent on context.

Reponses 7: I did not mean to ignore the suggestion. I did agree to the statement, but currently, there are 3 SME’s on which 17 month work was performed. If I would have more SME’s in process, it would have been great to create cross industry comparisons. I apologize the misunderstanding I created.

Reponses 8: Thank you very much, however, I cannot add more statistical data here, as I mentioned the quant data from 17 month is used for another article with different perspective. I already changed design for this paper to qualitative mixed data driven case study approach.

C10 – Language and consistency.
“ No comment” suggests the copy-edit has not been undertaken. The manuscript requires UK spelling, consistent terminology, and correction of factual slips.

Reponses 10: I had 2 native speakers re-read this work, and they have not identified any issues in my writing. I also re-read several times, and fixed minor issues.

Reponses 11: Again sorry for misconfusion. I mean that I have modified the work, and it was good for that moment. I have applied for corrections.

Reponses 12: I didn’t realize that. I have chosen 2^nd, as I changed the approach to qualitative mixed-data driven

Round 3

Reviewer 4 Report

Comments and Suggestions for Authors

Thank you for a substantive and careful revision. The manuscript is now coherently positioned as a qualitatively driven mixed-data case study. These changes materially strengthen transparency and align the contribution with the paper’s exploratory intent. Before acceptance, I ask for minor textual refinements so that the narrative is perfectly consistent with the non-causal stance adopted elsewhere. Please replace any residual causal wording in the Abstract and Conclusions (e.g., “significantly improves”, “produce a higher ROI”) with non-causal phrasing such as “is associated with” or “observed improvements in”, and retain the caveat that pre–post contrasts are sensitive to seasonality and exogenous shocks in the absence of a counterfactual. If “sustainable” is intended to mean durability or resilience rather than ESG outcomes, either use “resilient” consistently - including in the title and keywords - or delimit that meaning explicitly early in the Introduction to avoid misinterpretation. Please also remove any remaining template placeholders, correct a few small slips (e.g., “Small and Medium-sized Enterprises”, “empathize” for the Design Thinking stage, and the “St. Paul, MN” state code), and ensure that the CVPAT/PLSpredict narrative matches the tabled results in all waves. The English is generally publication-ready; these are minor copyedits and consistency adjustments. With these changes in place, I would support acceptance.

Comments on the Quality of English Language

While the prose is generally fluent, the revised manuscript still contains language and editorial problems that are material and must be corrected in the paper itself before acceptance. The most serious issue is a mismatch between wording and design: the Abstract and Conclusions still use causal verbs and claims (e.g., “significantly improves”, “produce a higher ROI”) even though the study is now framed as a descriptive mixed-data case study without a counterfactual. This is not a stylistic preference; it changes the meaning of the findings. Please replace all causal phrasing with non-causal formulations (e.g., “is associated with”, “observed improvements in”) and keep this consistent across the manuscript. A second substantive problem is terminological. The paper alternates between “sustainable/sustainability” and “resilient/resilience”. In scholarly usage, “sustainable” is commonly read as an ESG construct; because the study does not operationalise ESG outcomes, the current wording is misleading. Either use “resilient/resilience” consistently (including in the title and keywords) or explicitly delimit the intended meaning of “sustainable” at the start. There are also clear editorial lapses that must be fixed: unrevised template placeholders remain in the body of the text; the Design Thinking stage is labelled “emphasize” rather than “empathize”, which mischaracterises the method; a US location is given as “St. Paul, MS” instead of MN; and there are bibliographic inaccuracies (e.g., a reference title and DOI that do not match the cited source). These are not matters of style; they undermine the reliability of the manuscript. I acknowledge that native speakers have read the text; however, fluency is not the issue here. The problems above concern precision, terminological coherence and factual accuracy. Please address them comprehensively and visibly in the manuscript. Once these publication-blocking items are corrected, I will support acceptance after minor revision.

Author Response

Comment 1. Thank you for a substantive and careful revision. The manuscript is now coherently positioned as a qualitatively driven mixed-data case study. These changes materially strengthen transparency and align the contribution with the paper’s exploratory intent.

Response 1: Thank you for your feedback. I am thankful.

Comment 2: Before acceptance, I ask for minor textual refinements so that the narrative is perfectly consistent with the non-causal stance adopted elsewhere. Please replace any residual causal wording in the Abstract and Conclusions (e.g., “significantly improves”, “produce a higher ROI”) with non-causal phrasing such as “is associated with” or “observed improvements in”, and retain the caveat that pre–post contrasts are sensitive to seasonality and exogenous shocks in the absence of a counterfactual.

Response 2: Thank you for these suggestions. I have applied the changes in abstract and conclusion as per your guidance.

Comment 3: If “sustainable” is intended to mean durability or resilience rather than ESG outcomes, either use “resilient” consistently - including in the title and keywords - or delimit that meaning explicitly early in the Introduction to avoid misinterpretation.

Response 3: Thank you very much for the information. I applied these suggestions. I modified the title, and also offered explanation to delimit the meaning in introduction, paragraph 1.

Comment 4: Please also remove any remaining template placeholders, correct a few small slips (e.g., “Small and Medium-sized Enterprises”, “empathize” for the Design Thinking stage, and the “St. Paul, MN” state code), and ensure that the CVPAT/PLSpredict narrative matches the tabled results in all waves.

Response 4: Thank you for the suggestion. Have applied the suggestions and corrected them. I checked and results and tables are consistent.