What is MCP?
Model Context Protocol (MCP) securely connects AI tools like Claude, ChatGPT, Gemini, and Cursor to your research in Optimal, enabling your team to further explore and analyze your Optimal study data. Instead of searching through studies, copying results, or exporting data, you can ask your AI assistant questions and get answers informed by your research.
Once connected, MCP turns your repository into something you can ask directly, so all that past and current research is always instantly accessible. You can also integrate Optimal data and findings into other AI-powered workflows with tools already connected with your AI tools, such as Slack.
For example, you can ask to:
Find the most valuable UX opportunities for improvement.
Summarize findings across multiple studies.
Compare insights over time.
Pull participant quotes on a specific topic.
Generate reports, presentations, or stakeholder summaries based on your research.
Connect your Optimal data with existing tools and workflows already connected to your AI tool e.g. Slack
Read-only access
The Optimal MCP server provides read-only access to your research.
This means your AI assistant can retrieve and analyse information from your Optimal Workspace, but it cannot modify your data or create new data in Optimal, such as create a new study. Your studies, participants, results, and workspace remain unchanged.
What Optimal MCP can do
Once connected, your AI assistant can help you:
Analyze studies: Understand patterns, findings, and trends across research automatically.
Cross-study synthesis: Identify recurring themes across multiple studies.
Pull key insights: Extract findings from individual studies without manual review.
Search & explore research: Filter studies by creator, title, participant group, or timeframe.
Analyze transcript insights & sessions: Surface usability issues, pain points, and behavioral patterns.
Turn insights into deliverables: Automatically format findings into summaries and stakeholder-ready outputs. Get more ideas here.
Connect with other tools & workflows connected to your AI tool: Create alerts and automate next steps across the tools you already use to turn insights into action with less effort e.g. create a Slack notification when a participant completes a study, share milestones, create a JIRA ticket and follow-up tasks
What Optimal MCP can't do
The Optimal MCP server cannot currently:
Create, edit, delete, or launch studies
Build and add screeners
Recruit participants
Change workspace settings
Update participant information
Schedule participants
Create a workflow automation
Your AI assistant can only access the data you have permission to view in Optimal.
Getting started
Getting started only takes a few minutes.
1. Choose an MCP-compatible AI tool
The Optimal MCP server works with AI tools that support Model Context Protocol, including:
ChatGPT
Claude
Gemini
Microsoft CoPilot
Other compatible MCP clients
Log into your AI tool account.
2. Connect your Optimal Workspace
For guidance on setup, please reference the specific guides below:
or use the generic guide, which covers the details you'll need for any MCP client.
Tip: if needed, you can prompt your AI tool to give guidance on troubleshooting.
Once connected, the AI tool can securely access the Optimal resources and tools available to your account, based on your permissions.
Authentication
Optimal uses OAuth 2.0 for authentication. During setup, you'll be redirected to sign in and grant access to your account. No API keys are required, and access is scoped to your user permissions within your organization.
Once connected, your AI tool will be able to securely access your research.
3. Start asking questions
You can use natural language prompts in your AI tool such as:
"Summarise the key usability issues from our checkout studies."
"What have participants said about navigation across the last six months?"
"Find evidence supporting a redesign of our onboarding flow."
"Compare findings from our mobile and desktop usability tests."
"Create an executive summary of our recent research."
The more specific your question, the more focused your AI assistant's response will be.
Tips for better results
Ask about specific studies, products, or timeframes when possible.
Reference participant groups or research methods to narrow your search.
Ask follow-up questions to explore findings in more depth.
Request outputs in the format you need, such as a report, presentation, table, or stakeholder summary.
MCP works best when you treat your AI assistant like a research partner, asking questions, exploring findings, and iterating on the results.
MCP access
If you cannot currently connect to your AI tool, your organization may need to enable MCP for your Workspace.
To turn on MCP for your users, admins can go to Workspace Settings > Features to change their preferences.
If you are an organization admin of multiple Workspaces, please navigate to your Admin Team Workspace. Then go to Manage Organization > Features to change your preferences.