Voice Tools

Conversion Tracking

Track conversion events when callers complete goal actions — purchases, bookings, sign-ups, and more.

Conversion Tracking

The Conversion Tracking tool records conversion events when a caller completes a goal action during the call. Use it to measure call effectiveness, track revenue, and feed data into your analytics pipeline.

How It Works

Caller: "Yes, I'd like to sign up for the premium plan."
Agent: "Welcome to Premium! I've set up your account."
-> Agent calls track_conversion with event_type="premium_signup"
-> Conversion recorded with room, agent, and user metadata

Configuration

{
  "analyticsEnabled": true,
  "analyticsTrackConversions": true,
  "analyticsConversionEvent": "purchase"
}
SettingTypeDefaultDescription
analyticsEnabledbooleanfalseEnable analytics features
analyticsTrackConversionsbooleanfalseEnable conversion tracking
analyticsConversionEventstringconversionDefault conversion event name

LLM Tool

Tool name: track_conversion

Parameters:
  event_type: string (conversion event name, defaults to configured event)
  details: string (additional context about the conversion)

Returns: "Conversion tracked successfully."

Conversion Data

Each conversion is recorded with:

{
  "room_name": "room_abc123",
  "event_type": "purchase",
  "details": "Caller purchased premium plan at $49/month",
  "user_id": "user_123",
  "agent_id": "agent_456",
  "timestamp": "2026-03-07T14:30:00Z"
}

Example — Create Agent with Conversion Tracking

curl -X POST https://api.thinnest.ai/v1/agents \
  -H "Authorization: Bearer $THINNESTAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Sales Agent with Tracking",
    "model": "gpt-4o-mini",
    "instructions": "You are a sales agent. When the caller completes a purchase or signs up for a plan, use the track_conversion tool to record the event with details like plan name and amount.",
    "voiceEnabled": true,
    "transcriber": { "provider": "deepgram", "model": "nova-2-conversationalai" },
    "voice": { "provider": "deepgram", "voiceId": "aura-2-thalia-en" },
    "analyticsEnabled": true,
    "analyticsTrackConversions": true,
    "analyticsConversionEvent": "purchase"
  }'

On this page