Use Quotient in Other AI Apps
Connect Quotient to AI apps like Claude, Cursor, and Codex so they can do work in your Quotient account on your behalf.
MCP (Model Context Protocol) is an open standard that lets AI apps securely connect to other software.
This guide covers one direction: connecting other AI apps and agents to Quotient, so they can do work in your Quotient account without leaving the app you're already using. Quotient can also connect to outside tools like GitHub or Notion; if that's what you're looking for, see Connect Tools to Quotient.
With Quotient's MCP server, AI apps like Claude Code, Claude Cowork, Codex, and Cursor can take action in your Quotient account through a secure connection.
Common Use Cases
- Create product marketing materials for a feature you just shipped with Claude Code.
- Recreate an email template made in Claude Design.
- Pull marketing analytics and campaign data for a strategy memo written in Claude Cowork.
- Draft social posts or create campaigns without switching back to Quotient.
How Quotient's MCP Server Works
Quotient's MCP server exposes two main types of tools:
- Action tools: These mirror what you can do in Quotient yourself — creating campaigns, launching emails, publishing social posts, reading memories, and so on. Use them for direct, well-defined actions.
- Agent-to-agent tools: Instead of acting directly, these let another app talk to Quotient the same way you would. For example, Claude Code can send Quotient a message and get a response back. This is the best way to handle complex tasks that need a lot of Quotient-specific know-how.
Some tasks are best handled by talking to Quotient directly rather than calling individual tools:
- Creating and editing email broadcasts or email templates
- Defining complex flows, e.g. welcome series, customer win-back, and CRM automations
- Working with dynamic audience segments
- Analyzing data and creating analytics reports
These tasks rely on deep knowledge of Quotient's features, best practices, and internal formats, so it's usually best to ask Quotient to do them for you rather than stitching the individual steps together yourself.
Tips for Getting the Most Out of It
- Reach for agent-to-agent on anything complex. If a task touches emails, flows, segments, or analysis, just ask Quotient in plain language instead of chaining individual tools together.
- Be specific. The more context you give — audience, goal, tone, dates — the better Quotient's output, the same as if you were briefing a teammate.
- Lean on memory. Quotient draws on your saved brand and business memory, so the more you've taught it, the better your results across every connected app.
- Start small. Connect one app first, try a simple task like drafting a social post, then expand once you're comfortable.