Avalilação PREreview de Data-Driven Decision-Making in Marketing

Summary:

This paper presents a comprehensive and methodologically rigorous analysis of how data analytics has transformed marketing decision processes over the past decade. The authors employ a PRISMA 2020 framework and an LRSB (Literature Review Systematic Bibliometric) approach to identify key trends, conceptual frameworks, and research gaps.

Using Scopus-indexed publications as the data source and VOSviewer for bibliometric visualization, the paper maps how topics such as artificial intelligence, customer analytics, personalization, and marketing intelligence have evolved and interconnected over time. The findings highlight that organizations embracing data-driven strategies show improved market adaptability, campaign precision, and customer retention.

Importantly, this work moves the field forward by offering both a methodological roadmap for future systematic reviews in marketing and by clarifying how data-centric approaches can replace intuition-based decision-making with evidence-based strategies grounded in measurable outcomes.

Good / Positive Feedback:

Strong methodological foundation: The use of the PRISMA 2020 framework adds a high level of credibility and reproducibility to the study. It ensures that the review process is transparent, systematic, and well-documented. This adherence to structure helps position the research as a reliable reference for future meta-analyses in marketing.

Clarity in visualization and conceptual mapping: The inclusion of keyword co-occurrence and co-citation networks gives the reader a tangible sense of the field’s intellectual structure. The visualizations, although technical, are well chosen and reflect the authors’ effort to make bibliometric data accessible to both academic and industry audiences.

Bridging theory and practice: One of the paper’s greatest strengths is how it links conceptual findings with practical implications. The discussion connects bibliometric clusters (e.g., customer behavior analytics, AI-driven personalization) to marketing strategies that enhance efficiency, agility, and evidence-based leadership.

Major issues

1. Limited interpretive depth beyond bibliometric mapping: While the paper provides rich visual and quantitative analysis, it falls short in translating those patterns into actionable insights. For example, Figure 6 (keyword co-occurrence network) highlights “artificial intelligence” and “customer analytics” as dominant nodes. However, the paper does not delve deeply into why these topics have risen or how their integration changes marketing performance. A more interpretive narrative connecting clusters to real-world use cases—such as personalization strategies, customer lifetime value prediction, or omnichannel optimization—would strengthen the paper’s impact.

2. Insufficient methodological transparency in inclusion criteria: The authors mention that Scopus was used as the primary database under PRISMA guidelines, but do not detail the search strings, date ranges, or exclusion filters applied. This lack of granularity limits reproducibility. For example, it’s unclear whether conference papers or non-English studies were excluded, or how duplicates and bias were handled. Adding a small methodological appendix with explicit search terms and screening flowcharts would improve transparency.

3. Underdeveloped ethical and managerial perspective: This paper briefly references data ethics, but given the dominance of privacy, consent, and bias concerns in data-driven marketing, the ethical dimension deserves fuller exploration. For instance, how can marketers adopt DDDM while ensuring compliance with data protection laws such as GDPR (General Data Protection Regulation) or CCPA (California Consumer Privacy Act)? Similarly, what governance structures should organizations implement to prevent algorithmic bias? Addressing these questions would expand the practical relevance and credibility of the study.

Minor issues

1. Transitions between sections could be smoother: The paper occasionally shifts abruptly between methodology and results without summarizing the previous section’s key findings. For example, the transition from “PRISMA methodology” to “VOSviewer results” could use a short linking paragraph summarizing how the dataset evolved from raw Scopus entries to final bibliometric maps. This would make the narrative flow more natural for readers unfamiliar with bibliometric research.

2. Terminology consistency: The authors sometimes alternate between terms such as “LRSB,” “systematic review,” and “bibliometric analysis” without clear distinction. This can confuse readers about the methodological framework. A short glossary or definitional table early in the paper could clarify these differences—for instance, explaining how LRSB combines quantitative mapping with systematic synthesis.

3. Visual clarity in figures: Some visualizations (e.g., co-citation maps) contain overlapping labels that are difficult to read. Enhancing contrast, enlarging font size, or including callouts summarizing key clusters (e.g., “AI-driven analytics,” “customer engagement models”) would make these visuals more user-friendly. Moreover, adding interpretive captions below each figure would improve accessibility for readers unfamiliar with VOSviewer outputs.

Competing interests

The author declares that they have no competing interests.

Use of Artificial Intelligence (AI)

The author declares that they used generative AI to come up with new ideas for their review.