A Big Data–Driven Methodology for Conducting Marketing Research in the Food Products Market: A Conceptual Framework for Uzbekistan

Abstract

The digitalisation of food production, distribution, retailing, and consumption has created large volumes of heterogeneous and rapidly changing data. Point-of-sale transactions, loyalty programmes, e-commerce clickstreams, delivery applications, social media, online reviews, mobile location records, Internet of Things sensors, weather information, and supply-chain systems can provide a more continuous view of food markets than conventional surveys alone. However, the availability of data does not automatically produce valid marketing knowledge. This article develops a Big Data–driven methodology for conducting marketing research in the food products market, with a proposed application to Uzbekistan. The framework organises the research process around five properties of Big Data volume, velocity, variety, veracity, and value and integrates transactional, behavioural, attitudinal, contextual, and operational data. It specifies six methodological layers: research-problem formulation, data-source mapping, data governance and integration, analytical modelling, insight validation, and managerial activation. The proposed design combines descriptive, diagnostic, predictive, and prescriptive analytics to support market monitoring, demand forecasting, consumer segmentation, price and promotion evaluation, assortment decisions, and early detection of changes in consumer sentiment. Particular attention is given to data quality, identity resolution, selection bias, privacy, security, algorithmic fairness, and the difference between correlation and causality. Since an integrated empirical database is not yet available, the paper presents a conceptual and operational protocol rather than fabricated results. The framework contributes by adapting Big Data marketing research to the distinctive characteristics of food markets and emerging-market data environments.