A multi-level AI-driven product policy model for retail chains as a conceptual framework for marketing consulting

Abstract

The rapid development of omnichannel distribution systems and the increasing need to improve the speed and accuracy of marketing decision-making in product policy create a strong demand for the implementation of algorithmic data analysis methods and artificial intelligence technologies. This study aimed to develop the theoretical and methodological foundations of product policy management in retail enterprises through the design of a conceptual multi-level AI-driven product policy model. The research was based on a combination of general scientific and specialised methods, including analysis and synthesis, comparative analysis, systems analysis, structural-logical modelling, abstraction, and generalisation. The proposed model was conceptualised as an adaptive self-regulating system that integrates data, machine learning algorithms, and managerial decision-making processes and serves as a methodological platform for marketing consulting aimed at implementing intelligence-supported decisions in retail practice. The study provided a theoretical justification for the transition toward AI-driven product policy by identifying the system-forming role of data in marketing decision-making, analysing the capabilities of machine learning algorithms, and specifying directions for their application in demand forecasting, assortment optimisation, and automated inventory management. As a result, a conceptual multi-level model of AI-driven product policy was developed, integrating data infrastructure, machine learning analytics, managerial decision-making, and marketing consulting support into a unified adaptive management cycle. The model functions as an integrated closed-loop system with a feedback mechanism and includes the stages of data collection and integration, algorithmic forecasting and AI-based analytics, managerial decision-making for assortment optimisation and automated inventory management, and KPI-based performance monitoring. The practical significance of the research lies in the possibility of applying the proposed model as a methodological tool for marketing consulting aimed at increasing the adaptability of retail enterprises, reducing operational costs, accelerating inventory turnover, and strengthening competitive positions through data-driven management.