Big data in consumer behavior research: a systematic review of data sources, analytical methods, and research questions

Abstract

While big data offers unprecedented opportunities for decoding consumer behavior, researchers face significant challenges in navigating the fragmentation of heterogeneous data sources and the complexity of integrating advanced analytical techniques with behavioral theory. To address these gaps, this study conducts a systematic literature review following the PRISMA 2020 protocol, analyzing 127 peer-reviewed articles published between 2012 and 2023. Moving beyond prior reviews that focus on isolated domains or specific data types, this research contributes an integrated framework that maps specific data types (structured vs. unstructured) to analytical methods (e.g., machine learning, text mining) and their application in solving core behavioral research questions. By systematically aligning these elements, the framework demonstrates how big data can be operationalized to address theoretical gaps in consumer research. The analysis reveals that big data is extensively employed to analyze consumption patterns, sentiment, and decision-making pathways, yet challenges regarding data quality, algorithm interpretability, and privacy protection remain prevalent. Practically, this review provides a methodological roadmap for scholars and practitioners to leverage big data for precise behavioral prediction and personalized marketing, while underscoring the urgent need for interdisciplinary collaboration to address ethical compliance in data-driven strategies.