Analytics

Track call volume, platform-minute usage, carrier spend, sentiment distribution, and agent performance metrics.

Spend Overview

GET/v1/analytics/spend

Returns aggregated metrics across calls for the authenticated tenant. Filter by date range, agent, or call status.

Query Parameters

Parameter	Type	Required	Default	Description
from	string	No	None	Inclusive start date in YYYY-MM-DD format. When omitted, no start bound is applied and all calls are aggregated (the time_series window still defaults to 30 days ago).
to	string	No	None	Inclusive end date in YYYY-MM-DD format. When omitted, no end bound is applied and all calls are aggregated (the time_series window still defaults to today).
agent_id	uuid	No	All agents	Restrict metrics to a specific agent
status	string	No	All statuses	Filter by call status: completed, failed, in_progress

Example Request

curl "https://api.rymi.live/v1/analytics/spend?from=2026-03-01&to=2026-03-31&agent_id=550e8400-e29b-41d4-a716-446655440000" \
  -H "Authorization: Bearer YOUR_API_KEY"

Response 200

{
  "total_calls": 140,
  "completed_calls": 120,
  "in_progress_calls": 2,
  "average_duration_seconds": 187,
  "success_rate": 0.72,
  "total_minutes": 1450,
  "customer_revenue": 58.0,
  "provider_cost": 46.5,
  "gross_margin": 11.5,
  "gross_margin_pct": 19.83,
  "time_series": [
    {
      "date": "2026-03-01",
      "calls": 8,
      "minutes": 34,
      "customer_revenue": 2.4,
      "provider_cost": 1.9,
      "gross_margin": 0.5
    }
  ],
  "status_distribution": {
    "completed": 120,
    "failed": 18,
    "in_progress": 2
  },
  "sentiment_distribution": {
    "positive": 40,
    "neutral": 61,
    "negative": 8,
    "unknown": 31
  },
  "end_reason_distribution": {
    "caller_hangup": 72,
    "agent_ended": 40,
    "error": 8
  },
  "agent_performance": [
    {
      "agent_id": "550e8400-e29b-41d4-a716-446655440000",
      "calls": 90,
      "minutes": 740.5,
      "customer_revenue": 31.2,
      "provider_cost": 25.1,
      "gross_margin": 6.1,
      "evaluated_calls": 64,
      "passed_calls": 46,
      "success_rate": 0.719,
      "sentiment": { "positive": 30, "neutral": 48, "negative": 6, "unknown": 6 },
      "sentiment_score": 0.286
    }
  ],
  "drop_off_points": [
    { "bucket": "<30s", "count": 3 },
    { "bucket": "30-60s", "count": 7 },
    { "bucket": "60-120s", "count": 25 },
    { "bucket": ">120s", "count": 45 }
  ],
  "filters": {
    "from": "2026-03-01",
    "to": "2026-03-31",
    "agent_id": "550e8400-e29b-41d4-a716-446655440000",
    "status": null
  }
}

Response Fields

Field	Type	Description
total_calls	integer	Total call count matching the filters
completed_calls	integer	Calls that completed successfully
in_progress_calls	integer	Calls currently active
average_duration_seconds	integer	Average call duration in seconds
success_rate	number	Ratio of evaluated calls that passed their evaluation (0.0–1.0)
total_minutes	number	Total billed minutes across completed calls
customer_revenue	number	Total customer-billed revenue in USD across completed calls
provider_cost	number	Total underlying provider cost in USD across completed calls
gross_margin	number	customer_revenue − provider_cost in USD
gross_margin_pct	number | null	Gross margin as a percentage of revenue; null when revenue is 0
time_series	array	Daily breakdown: { date, calls, minutes, customer_revenue, provider_cost, gross_margin }
status_distribution	object	Call count grouped by final status
sentiment_distribution	object	Call count grouped by post-call sentiment (positive, neutral, negative, unknown)
end_reason_distribution	object	Call count grouped by why the call ended (e.g. caller_hangup, agent_ended)
agent_performance	array	Per-agent breakdown: { agent_id, calls, minutes, customer_revenue, provider_cost, gross_margin, evaluated_calls, passed_calls, success_rate, sentiment, sentiment_score }
drop_off_points	array	Duration histogram of non-passing calls, bucketed <30s, 30-60s, 60-120s, >120s
filters	object	Echo of applied filters

Errors

Status	Meaning
401	Missing or invalid API key
500	Server error while querying analytics

INFO

Analytics data is aggregated from call records plus post-call intelligence. Real-time minute counts may be supplemented from live usage tracking when available.

Cost Scope

provider_cost (and the derived gross_margin/gross_margin_pct) are aggregated from each completed call's recorded provider_cost. For self-serve BYOC, this reflects carrier spend only when the connected provider reports costs and the gateway persists them. LLM/STT/TTS provider token-level cost metering is on the roadmap.