Agentic Approach to Using Neo4J Knowledge Graphs in Salesforce

Session Track: Knowledge Graphs

Session Time: 05:15 am - 05:30 am November 6

Session description

In this session, the presenter will showcase an agentic approach to integrating Neo4j knowledge graphs within Salesforce through the Model Context Protocol (MCP). Salesforce, a leading CRM platform, offers powerful case management capabilities. However, resolving complex cases often demands deeper insights and contextual relationships among customer data that are traditionally hard to model within standard relational systems. Neo4j, a graph database optimized for capturing relationships, enables the modeling of rich knowledge structures. Using MCP—an emerging protocol that facilitates natural language-driven interactions between systems—this solution allows Salesforce agents to query and interact with a Neo4j-powered knowledge graph directly from within the Salesforce environment. Attendees will learn how to architect and implement this integration, including building MCP plugins, mapping Salesforce data into a graph structure, and enabling natural language queries through tools like Claude. By the end of the session, you will be equipped to leverage graph-based insights to accelerate case resolution and enhance customer support outcomes in your Salesforce implementation.