MCP: Run your research from inside Claude, Cursor, or ChatGPT
Design studies, build screeners, invite your customers, and synthesize what they told you, all without leaving the chat. Every finding backed by the real session highlights behind it.
What is MCP?
Model Context Protocol (MCP) is an open standard that lets AI tools securely connect to — and work inside — external software.
For Great Question, that means Claude, Cursor, and ChatGPT can now do your research for you. Not just read what's in your repository — actually design studies, invite your customers, and bring the findings back.
For Great Question, that means Claude, Cursor, and ChatGPT can now do your research for you. Not just read what's in your repository — actually design studies, invite your customers, and bring the findings back.
MCP
What you can do
70 new things your AI can do across the research workflow: from designing studies to verifying findings against the real highlights behind them.
Works with tools used by
🔍
Search sessions
New
Find interviews & tests
Search your research repository by topic, participant, or keyword. Surface relevant insights without leaving your workflow.
"Find interviews about checkout friction"
📋
List studies
New
Browse active research
See what research is in progress across your organization. Filter by team, owner, or status.
"Show active studies from Mobile team"
🎤
Create interviews
New
Moderated studies
Spin up interview studies with guides, participant limits, and incentives — all from your AI assistant.
"Create an interview about pricing"
📝
Create surveys
New
12+ question types
Build surveys with multiple choice, scales, open-ended questions and more. Perfect for quick validation.
"Create a 5-question NPS survey"
Built for product builders
Four ways teams are using Great Question inside their AI tools right now.
📄 Write PRDs with the customer voice
Pull customer quotes, pain points, and feedback directly into your product specs. No more switching tabs or digging through old research decks.
"Find interviews about payment friction, then help me write the problem statement for my PRD"
📄 Research-informed roadmap prioritization
Query what research exists before prioritizing features. Identify gaps in your understanding and create quick validation studies to fill them.
"What research do we have about settings? Create a survey to fill the gaps."
📄 Launch a study without leaving your chat
Describe what you want to learn. Your AI designs the study, writes the screener, invites your customers, and books the calendar, all from a single prompt.
"Recruit 30 weekly-active PMs for an interview about our new settings flow. Screen out anyone at companies under 100 people."
📄 Synthesize research results instantly
Ask a question, run the research that answers it, and synthesize the results, without leaving Claude.
"Create a weekly check-in survey for beta testers"
Get started in 3 steps
1
Add the MCP server
Run this command in Claude Code to add Great Question as an MCP server.
claude mcp add --transport http great-question https://greatquestion.co/api/mcp/v12
Authorize access
Your browser opens to Great Question. Review the permissions and click Authorize. Your existing role permissions apply.
3
Start using
Ask Claude to review past research and create a new study. Search sessions, list studies, or create new research, all from your terminal.
See it in action
Watch how to build a PRD with voice of customer data using Claude and Great Question.
The most comprehensive MCP integration for research
Great Question is the most comprehensive MCP out there, with over 70+ ways you can use the MCP to operate Great Question.
You will be among the first to use MCP with your own research!
You will be among the first to use MCP with your own research!
Capability
Great Question
Other Platforms
Native MCP Server
✓
Limited*
Search Research Sessions
✓
✓
List Studies
✓
✗
Create Interview Studies
✓
✗
Create Surveys
✓
✗
OAuth 2.0 Security
✓
Varies
Enterprise Role Permissions
✓
Limited
*Some platforms offer MCP through third-party connectors like Zapier with limited functionality.
Enterprise-grade security
Your research data stays protected with the same security you expect from Great Question.
🔍
OAuth 2.0 Authentication
Industry-standard authorization flow. Your credentials never touch AI tools directly.
👥
Role-Based Permissions
Your existing Great Question permissions apply automatically. See only what you're authorized to see.
🛡️
SOC 2 Type II
Complete data separation between accounts. Enterprise-grade security certifications.
Request Access
Great Question's MCP integration is available upon request.
Ready to run research from a chat?
Design studies, invite your customers, and synthesize what they told you, all from inside Claude, Cursor, or ChatGPT.