Bring Churnkey to your AI agent
The Churnkey MCP server brings churn analytics, at-risk accounts, and saved revenue into the AI tools your team already uses. Ask a question. Get the answer. No dashboard required.
MCP is the open standard for connecting AI agents to your data.
Across 447 failed payments, Payment Recovery clawed back $173,419 of $214,756 at risk—an 81% recovery rate. By decline reason:
| Decline reason | Attempts | At risk | Recovered | Rate |
|---|---|---|---|---|
| insufficient_funds | 171 | $79,983 | $68,175 | 85% |
| do_not_honor | 54 | $33,393 | $31,269 | 94% |
| generic_decline | 34 | $21,028 | $18,619 | 89% |
| transaction_not_allowed | 31 | $16,575 | $14,134 | 85% |
insufficient_funds is the volume game—38% of attempts, recovered at 85% with retry-til-payday timing. do_not_honor quietly recovers best at 94%.
Works with
Claude
Cursor
ChatGPT
Perplexity
Codex
VS Code and more 
and more Your data lives in dashboards. Your team would rather not.
Every retention question turns into a context switch. Open the dashboard. Pick the right view. Filter by plan, by cohort, by month. Export. Reformat. Share. Repeat.
By the time the answer's in hand, the moment for the next question is gone. So you stop asking.
The Churnkey MCP MCP (Model Context Protocol) is an open standard that lets your AI agent connect to tools and data sources like Churnkey. server gives AI agents permission-aware access to your retention data (churn analytics, at-risk accounts, saved revenue, session histories, and more).
How it works
Retention answers,
where your team already works.
The MCP server is a tiny package that connects your Churnkey workspace to any AI agent that speaks the Model Context Protocol. Install it once, and every question your team has about churn, saves, or revenue gets a real answer—backed by your real data.
No new dashboard to learn. No new tab to forget about. Just the tools your team already trusts, now with retention context.
Start here with a simple npm install.
A single command drops the MCP server directly into your AI agent of choice. No build step, no infrastructure, no waiting on engineering.
Connect once.
Paste your Churnkey API key into your AI agent's config. The same snippet works across Claude Desktop, Claude Code, Cursor, ChatGPT, Codex, and the terminal.
Ask in plain language.
"What's my save rate this quarter?" "Show me at-risk accounts." "Which offer converted best last month?" Whatever you'd ask a teammate, ask your agent instead.
Act on the answer.
Every response comes with the numbers behind it. Pipe it into a report, share it in Slack, or keep moving—the context is right there in the conversation you're already having.
Built for everyone
Churn isn't one team's problem.
Neither are the answers.
RevOps, founders, CS, engineers—everyone has questions about churn. Pick a role to see the questions they ask, and the answers the MCP server hands back.
Cohort analysis
"Why did Enterprise churn spike in February?"
Get the cohort breakdown and the drivers behind it—sliced by plan, geography, or signup month. No SQL, no waiting on the data team.
Plan comparison
"Compare save rate between Pro and Team this quarter."
See save rate, recovered MRR, and offer performance side by side, with the QoQ trend baked in.
Churn drivers
"Which cancel reason grew the most last month?"
Rank the fastest-rising cancel reasons from your Cancel Flows by change, so you know what to fix next.
Forecasting
"Forecast next month's churn from current signals."
Project gross churn from intent and usage signals, with a confidence range—no spreadsheet gymnastics.
Set up in 60 seconds
Three steps from install to first answer.
Install.
Drop the package onto your machine with a single command.
Add to your AI agent.
Paste your Churnkey App ID and Data API key into the MCP config for Claude Desktop, Cursor, ChatGPT, or Codex CLI.
{ "mcpServers": { "churnkey": { "command": "npx", "args": ["-y", "@churnkey/mcp"], "env": { "CHURNKEY_APP_ID": "your_app_id", "CHURNKEY_API_KEY": "your_api_key" } } } }
Ask your first question.
Open your AI tool and start asking. The MCP server handles the rest.
Built with care
Your data stays yours.
The MCP server is auditable, predictable, and tightly scoped.
API-key auth.
Your Churnkey API key stays in your local config. Requests go directly from your machine to Churnkey—nothing routes through anyone else's servers.
Read-only.
The server exposes queries, not mutations. Your agent can read your retention data—it can't change it. Configuration stays where it belongs.
Open source.
Audit the code yourself. The MCP server is open source at github.com/churnkey/mcp.
Compound effect
Retention answers, on the same surface as the work.
When churn data is a question away, the question gets asked more often. Teams that bring retention into their AI workflow notice patterns earlier, fix them faster, and compound small wins.
The MCP server is a small package. The behavior change it unlocks is not.
Still unsure?
Frequently asked questions
Enterprise-grade security, built in.
Churnkey meets top industry standards for data security and compliance.
Bring Churnkey to your AI agent.
The MCP server is free, open source, and ready the moment you are. Install it now and ask your first question in 60 seconds.
