Use dothemathbefore from Claude or Cursor

dothemathbefore exposes its calculation engine as an MCP (Model Context Protocol) server, so Claude or Cursor can run LTV, churn-curve, and payback calculations directly — using the exact same logic as this app. Add it as a remote server, no local install required.

1. Add it to Claude

In Claude Desktop, open Settings → Connectors → Add custom connector, or edit your claude_desktop_config.json directly with this:

{
  "mcpServers": {
    "dothemathbefore": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://customer-value-lab.lovable.app/mcp",
        "--transport",
        "http-only"
      ]
    }
  }
}

2. Restart Claude

Fully quit and reopen Claude Desktop (not just close the window) so it picks up the new connector.

3. Try it

Ask Claude something like: "Estimate LTV for a subscription product, continuous need, direct purchase, €40 ARPU, 6% monthly churn, 75% margin, €150 CAC." Or paste a share link copied from the app — "Load my assumptions from this link: …" — to bring your onboarding answers straight into the chat. When you're happy with the numbers, ask "Import these results into the dashboard" to get a link back into the app.

Available tools

load_assumptions_from_link

Paste a share link from the app (or just its 'd' parameter) to import all the assumptions from your onboarding — business description, sector, need nature, access model, churn curve, ARPU, churn, margin, CAC — straight into the conversation.

calculate_ltv

Full LTV calculation (gross, net, discounted), LTV:CAC ratio, payback period, and the per-cycle curve, from raw churn/ARPU/margin/CAC assumptions.

recommend_churn_curve

Given the nature of the customer need and how customers first get in, recommends a churn curve shape and explains why.

estimate_ltv

End-to-end: describe the business in plain terms and get a full LTV estimate in one call — the same flow the app's onboarding walks you through.

import_results_to_dashboard

Builds a link that opens results worked out in this conversation straight in the dothemathbefore dashboard, so you can keep tweaking them with sliders instead of chat.