AGENTIC ARTIFICIAL INTELLIGENCE AS A NEW PARADIGM OF MARKETING AUTOMATION: OPPORTUNITIES, CHALLENGES AND ECONOMIC FEASIBILITY

Authors

  • Oleksii Ya. Yarmoliuk National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
  • Victor L. Sibruk National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine
  • Andrii Yu. Tryvailo National Technical University of Ukraine "Igor Sikorsky Kyiv Polytechnic Institute", Ukraine

DOI:

https://doi.org/10.30857/2786-5398.2025.6.10

Keywords:

agent-based artificial intelligence, marketing automation, multi-agent systems, business process, Model Context Protocol, performance marketing, digital economy

Abstract

The article examines agentic artificial intelligence (agentic AI) as a new paradigm of marketing automation that replaces traditional approaches based on rigidly programmed workflows. The study provides a conceptual distinction between generative and agentic AI: while the former focuses on automating content creation, the latter is capable of autonomously setting sub-goals, making decisions, and coordinating the execution of marketing tasks in real time without continuous human intervention. The architectural components of agentic systems in marketing are systematized, including reasoning (planning models), action tools (tools/APIs), memory (contextual and long-term), and feedback loops. Key application areas are identified, such as autonomous campaign management, dynamic audience segmentation, omnichannel communication orchestration, predictive pricing, and automated A/B testing. A comparative analysis of leading agentic marketing platforms is conducted, including Salesforce Agentforce 360, HubSpot Breeze, IBM watsonx Orchestrate, and Zeta Global AI Agent Studio. The study also identifies key challenges hindering large-scale adoption, including reliability issues in agent chains (e.g., agent hallucinations in execution contexts), the lack of standardized ROI evaluation metrics, regulatory uncertainty regarding autonomous decision-making in advertising, and security risks associated with integrating agents into corporate data environments. Special attention is given to the Model Context Protocol (MCP) as a standard for inter-agent communication and its role in enabling multi-agent marketing ecosystems. The paper argues that the economic feasibility of agentic AI in marketing is determined not only by the reduction of operational costs but also by the ability of such systems to generate competitive advantages through rapid response to market signals and a level of personalization that exceeds human capabilities.

Downloads

Download data is not yet available.

Published

2025-12-23

How to Cite

Ярмолюк, О. Я., Сібрук, В. Л., & Тривайло, А. Ю. (2025). AGENTIC ARTIFICIAL INTELLIGENCE AS A NEW PARADIGM OF MARKETING AUTOMATION: OPPORTUNITIES, CHALLENGES AND ECONOMIC FEASIBILITY. Journal of Strategic Economic Research, (6), 104–113. https://doi.org/10.30857/2786-5398.2025.6.10

Issue

Section

MODERN TENDENCIES AND MANAGEMENT PROBLEMS