Implementation of Agentic AI, Multi-Agent Sytems and Retrieval-Augmented Generation In Digital Marketing: A Systematic Literature Review

Abstract

The rapid advancement of artificial intelligence has significantly transformed digital marketing practices, particularly through the emergence of Large Language Models (LLMs), Retrieval-Augmented Generation (RAG), and Agentic AI architectures. This study aims to analyze the evolution, implementation, opportunities, and limitations of Agentic AI using a Multi-Agent System approach in supporting digital marketing strategies. The study employs a Systematic Literature Review (SLR) method by synthesizing relevant academic publications published between 2020 and 2025 from indexed databases. The review focuses on the integration of autonomous AI agents, collaborative multi-agent workflows, and retrieval enhanced language models within digital marketing environments. The findings indicate that Agentic AI enables more adaptive and data driven marketing processes through automated decision making, audience segmentation, campaign optimization, and personalized content generation. Furthermore, the integration of RAG enhances contextual accuracy and reduces the limitations of static LLM knowledge. Despite its strategic potential, several challenges remain, including system reliability, hallucination risks, computational complexity, ethical concerns, and organizational readiness. This study contributes to the growing academic discussion on Agentic AI implementation in digital marketing and provides insights for future development of intelligent marketing systems in business environments.