
A real estate broker in the U.S. is now managing more than 11,000 leads with a team of eight humans and one AI agent named Russ. The agent has its own email address, scores leads against multiple signals, and sends daily CRM briefs to human agents with follow-up recommendations.
Marcus Rush, founder of Rush Home, built the system using Claude and Zapier MCP. According to a LinkedIn post by Zapier co-founder Wade Foster, the setup lets Rush build workflows that his CRM could not natively support.
From Trigger Limits to Custom Workflows
Rush Home previously relied on whatever triggers its CRM happened to offer. If a workflow required a trigger the CRM did not provide, the team either worked around it or dropped the idea. Zapier MCP changed that.
Zapier MCP is a Model Context Protocol implementation that connects AI tools like Claude, ChatGPT, and Cursor to more than 9,000 apps, including Gmail, Slack, and Salesforce. The company launched one-click installation earlier this year and now reports 195,000 MCP servers created, 4.6 million tool calls completed, and 250,000 apps connected.
Rush built Russ to evaluate leads using signals such as buyer versus seller intent, response timing, and call duration. The agent then emails each human agent a daily summary with recommended actions. "I can build exactly what I want," Rush said in the post. "We're not limited to a trigger anymore."
A Workflow Built for Scale
Rush Home's setup demonstrates how small teams can use AI agents to handle large volumes of operational data without hiring additional staff. The AI agent processes lead scoring continuously, surfaces insights, and distributes actionable intelligence to the team.
Zapier MCP eliminates the need to wait for native CRM integrations or build custom API connections. According to the company's MCP landing page, the tool is designed to bridge the gap between AI reasoning and execution. "Your AI is great at thinking. It drafts, summarizes, plans, and researches. But ask it to check your email, update a spreadsheet, or log a task and then it hits a wall," the page states.

For enterprise customers, Zapier MCP includes account-level restrictions, managed connections, and workspace controls. IT teams can govern how AI agents access business tools across the organization, regardless of which AI platform employees use.
MCP Adoption Accelerates
Zapier has released multiple MCP-related updates in recent months. The company shared how its customer advocacy team built an AI-native system using MCP and the Zapier SDK. It also launched an SDK that lets developers execute integrations from AI coding tools like Cursor and Replit.
Rush Home's use case illustrates the potential for small businesses to build custom automation without technical teams or large budgets. The ability to design workflows around specific business logic, rather than available triggers, opens up new operational possibilities for companies managing high volumes of leads, customer interactions, or repetitive tasks.
Zapier MCP is available now with one-click installation at zapier.com/mcp. The tool works with AI platforms that support the Model Context Protocol standard.
Sources
1 checkedHow we cover tool news: Create With's tool desk drafts these reports with AI from the sources listed above and checks them against those sources before publishing.







