Getting started with the Conjointly API and MCP server
Conjointly offers an API connection and an MCP server to make it possible for you to work with your Conjointly data programmatically. Use the API when you are writing your own code or integrating Conjointly with other systems. Use the MCP server when you want an AI assistant, such as Claude or ChatGPT, to work with your data for you in plain language.
Both give you access to the same data, with the same permissions as your Conjointly account.
Conjointly MCP server
MCP (Model Context Protocol) is an open standard for connecting AI assistants to other services. With the Conjointly MCP server connected, you can ask an AI assistant about your experiments without writing any code, for example to:
- List and search your experiments.
- Read reports, analysis results and respondent data.
- Run analyses and request exports.
You sign in with your Conjointly account when you connect, so there is no access token to manage.
Set it up in a few steps with Claude, ChatGPT, Microsoft 365 Copilot, Mistral Vibe or another AI assistant:
Conjointly API
An API, short for ‘Application Programming Interface’, is a set of rules that lets programs communicate with each other, standardising requests to the server and processing of the response from the server. The Conjointly API lets your code interact with the Conjointly platform for a wide range of tasks, including:
- Launching experiments.
- Inviting respondents.
- Viewing and exporting results.
The Conjointly API has predictable resource-oriented URLs, accepts JSON-encoded request bodies, returns JSON-encoded responses, and uses standard HTTP response codes, authentication, and verbs. These mean you can interact securely with our API from any client-side web application.
To start, create an access token, then see Making an API call.
Let’s build together: Share your suggestions
The Conjointly API and MCP server are still growing and evolving. We want to hear your feedback on how we can improve them to better suit your needs.