Problematic aspects of predictive analytics in digital marketing

DOI: 10.31673/2412-9070.2026.029610

Abstract

The article provides a comprehensive examination of the theoretical foundations, technological determinants, and applied challenges associated with the development and implementation of predictive analytics in digital marketing. The study synthesizes current scientific insights and industry evidence to reveal how machine learning algorithms, artificial intelligence models, and large-scale data processing infrastructures reshape contemporary marketing decision-making. Particular emphasis is placed on the role of Big Data ecosystems, cloud computing services, and customer data platforms (CDP) as core enablers of scalable predictive modelling and personalized communication strategies. The research demonstrates that predictive analytics significantly enhances the accuracy of forecasting consumer behavior, supports the optimization of pricing and promotional mechanisms, and strengthens the personalization of marketing interactions across digital channels.
At the same time, the article identifies a set of critical constraints that limit the efficiency and reliability of predictive systems in real-world business environments. These include data quality issues, organizational resistance to data-driven transformation, algorithmic opacity, ethical dilemmas, and the phenomenon of concept drift, which leads to the degradation of model performance under dynamic market conditions. The study also highlights the increasing relevance of explainable AI, ethical frameworks for automated decision-making, and hybrid predictive–prescriptive architectures capable of generating scenario-based managerial recommendations in real time. The analysis of current trends for 2023–2025 demonstrates a shift toward hyper-personalization, dynamic pricing, generative AI integration, and the creation of unified customer profiles, all of which reinforce the strategic importance of predictive analytics in digital marketing ecosystems. The findings underline the necessity of advancing methodological tools, institutional readiness, and ethical safeguards to ensure the effective and responsible adoption of predictive technologies in marketing practice.

Keywords: predictive analytics, machine learning, digital marketing, consumer behavior forecasting, Big Data, CDP, personalization, dynamic pricing, recommendation systems, prescriptive analytics.