CRMAgent: A Multi-Agent LLM System for E-Commerce CRM Message Template Generation

2025-07-11

AI Summary

In e-commerce private channels (e.g., IM, email), merchants commonly lack both professional expertise and scalable tools to author high-conversion CRM message templates. To address this, we propose a multi-agent large language model system specifically designed for e-commerce private domains, featuring a novel three-tiered generation mechanism: (i) population-based learning to iteratively optimize historically underperforming templates; (ii) semantic retrieval to adapt high-performing analogous templates; and (iii) rule-driven zero-shot fallback to ensure robustness. The system integrates multi-agent coordination, pattern-transfer learning, and a hybrid rule engine to jointly support template generation and actionable, step-by-step writing guidance. Experimental evaluation demonstrates that the generated templates significantly outperform human-crafted ones in audience alignment and marketing effectiveness, achieving an average 23.6% improvement in message conversion rate.

Technology Category

• Multiagent Systems: Agent Communication
• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

In e-commerce private-domain channels such as instant messaging and e-mail, merchants engage customers directly as part of their Customer Relationship Management (CRM) programmes to drive retention and conversion. While a few top performers excel at crafting outbound messages, most merchants struggle to write persuasive copy because they lack both expertise and scalable tools. We introduce CRMAgent, a multi-agent system built on large language models (LLMs) that generates high-quality message templates and actionable writing guidance through three complementary modes. First, group-based learning enables the agent to learn from a merchant's own top-performing messages within the same audience segment and rewrite low-performing ones. Second, retrieval-and-adaptation fetches templates that share the same audience segment and exhibit high similarity in voucher type and product category, learns their successful patterns, and adapts them to the current campaign. Third, a rule-based fallback provides a lightweight zero-shot rewrite when no suitable references are available. Extensive experiments show that CRMAgent consistently outperforms merchants' original templates, delivering significant gains in both audience-match and marketing-effectiveness metrics.

Problem

• Generates persuasive CRM message templates for e-commerce
• Improves message quality via multi-agent LLM system
• Enhances retention and conversion with actionable guidance

Innovation

• Group-based learning from top-performing messages
• Retrieval-and-adaptation of successful template patterns
• Rule-based fallback for zero-shot rewriting