• NEW
  • RETAINED
  • RESURRECTED
  • CHURNED

How to classify user cohort movements with your AI

Connect your data once. Your AI then sorts every user into specific lifecycle buckets.

Ask AI about this page

Short answer

How can I classify user cohort movements with my AI?

Your AI compares current user lists against historical data to identify who is new, who stayed, who came back, and who left. You get a clear breakdown of these movements.

The results.

Where your users land.

The AI sorts every user into one of these four categories.

NEW

First-time users

The AI finds users appearing in the current list who have no prior history. You get a count of your newest acquisitions.

RETAINED

Steady users

The AI identifies users present in both the current and previous lists. This shows you who is sticking around.

RESURRECTED

Returning users

The AI flags users who were inactive in the last period but showed up again now. This tracks your win-back success.

CHURNED

Lost users

The AI finds users from the previous list who are missing from the current one. You'll need to decide how to handle these lost accounts.

The workflow

What your AI does when the data arrives.

The AI processes your raw lists through a specific logic to ensure every user is accounted for.

  1. Ingest user lists

    Your AI reads the raw user identifiers from your provided source files.

    classify_user_flow

  2. Compare historical states

    The AI checks the current list against previous snapshots to find matches and gaps.

    classify_user_flow

  3. Assign cohort buckets

    Every user is tagged as new, retained, resurrected, or churned based on their movement.

    classify_user_flow

  4. Flag anomalies

    The AI identifies users that don't fit standard patterns for your manual review.

    classify_user_flow

Try it

Copy these to start.

Use these prompts to trigger the classification immediately.

Starting points

Replace the bracketed text with your actual file names or database table names.

Growth Accounting Framework Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_BwNh9Of8HL1D6Dm9nKDKxuaH0EFsiLPmdaOR37af/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Growth Accounting Framework capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "growth-accounting-framework-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_BwNh9Of8HL1D6Dm9nKDKxuaH0EFsiLPmdaOR37af/mcp"
    }
  }
}
Copy into chat04

  • Classify the users in [users_july.csv] into new, retained, resurrected, and churned buckets.

  • Compare the user list from [current_month.csv] against [last_month.csv] to show cohort movements.

  • Look at the user data in [production_db] and tell me how many users were resurrected this period.

  • Identify all churned users by comparing [active_users.csv] with [previous_users.csv].

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Growth Accounting Framework once, then ask.

Just link your data source. Your credentials stay encrypted, and you can start asking questions in your preferred AI client right away.

Connect Growth Accounting Framework to your AI

FAQ

How this task behaves.

  • 01

    Can the AI automatically update my database with these categories?

    No. The AI only reads your data and provides the classification results in your chat interface.

  • 02

    Does the AI delete any of my user records?

    No. The AI has no permission to delete or modify your original data.

  • 03

    Can I use this with any user list?

    Yes, as long as the list contains unique identifiers like email addresses or user IDs.

  • 04

    What happens if a user is in both lists but has been inactive for a year?

    The AI follows the logic of the current list comparison. If they are in both, they are classified as retained.

  • 05

    Can the AI create new user accounts?

    No. This MCP is strictly for analyzing and classifying existing user movements.

  • More questions about Growth Accounting Framework? The Connector page answers them. See everything the Growth Accounting Framework Connector can do

Connect Growth Accounting Framework to Claude, Cursor, ChatGPT & more