- CRITICAL
- WATCH
- STABLE
- HUMAN
Short answer
How do I find customers at risk of churning?
Your agent pulls health scores and specific performance metrics to spot declining trends across your accounts. You'll get a clear list of which customers need a check-in right now. This breakdown shows you how to set that up.
Where your insights land.
Risk categorization.
The AI sorts your data into actionable buckets based on health indicators.
CRITICAL
Immediate intervention
The AI flags accounts where health scores have dropped below your threshold. These go straight to your priority list.
WATCH
Declining engagement
The agent spots accounts where usage metrics are trending downward. You'll see these before they become critical.
STABLE
Healthy accounts
The AI identifies customers with consistent, high-performing metrics. These stay on your standard monitoring list.
HUMAN
Manual review needed
The AI highlights accounts with conflicting data points. You'll need to step in and investigate these personally.
The workflow
What your AI does when the data arrives.
The agent moves from broad scanning to deep investigation automatically.
Scan your customer base
The AI pulls a full list of your current customers to establish a baseline.
list_customersAnalyze health indicators
The agent looks at specific KPIs to see which accounts are actually struggling.
list_customer_kpisDeep dive into accounts
Once a problem is spotted, the AI pulls the full profile for that specific account.
get_customer_detailsSummarize risk levels
The AI compares the gathered metrics to identify which accounts are most likely to churn.
list_customer_kpis
Try it
Copy these to start.
Use these prompts to kick off your analysis.
Starting points
These are opening lines. Swap out the specific account names or metric types to match your setup.
Custify Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_7d7Mwghw15m8yaUYZomXwVHrIl5ehOmM1upP12IY/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Custify capabilities are ready to use.
{
"mcpServers": {
"custify-mcp": {
"url": "https://edge.vinkius.com/vk_preview_7d7Mwghw15m8yaUYZomXwVHrIl5ehOmM1upP12IY/mcp"
}
}
}List all customers in Custify and tell me which ones have a health score below 50.
Check the KPIs for Acme Corp and tell me if their usage is dropping.
Find all customers who haven't had a significant increase in engagement this month.
Get the details for Globex Corporation and summarize their current health status.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Custify once, then ask.
Just provide your credentials via the secure link. Once connected, your agent can immediately start querying your customer data.
Connect Custify to your AIFAQ
How this task behaves.
- 01
Can the AI change a customer's health score?
No. The AI can only read and report on the data currently in Custify.
- 02
Will the AI automatically email my customers?
No. The AI identifies the risk, but you must handle the actual communication.
- 03
Can I ask for specific KPI trends?
Yes. You can ask the AI to look for specific patterns in the metrics it retrieves.
- 04
Does the AI delete any customer records?
No. This MCP is read-only for the tools provided.
- 05
Can the AI create new customer profiles?
No. The available tools are limited to listing and retrieving existing data.
More questions about Custify? See everything the Custify Connector can do
Connect Custify to Claude, Cursor, ChatGPT & more
