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

TL;DR

CRMAgent is a multi-agent LLM system that enhances e-commerce CRM message template generation by learning from top-performing messages, retrieving similar successful templates, and providing fallback rewriting, significantly improving message effectiveness.

Contribution

This paper introduces CRMAgent, a novel multi-agent system leveraging LLMs for scalable, high-quality CRM message template generation with learning, retrieval, and rule-based fallback modes.

Findings

CRMAgent outperforms original merchant templates in effectiveness.

The system achieves significant improvements in audience match.

It demonstrates robustness across various campaign scenarios.

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.