Aggregates metric values and conversation counts over a time window and returns one series per measure, with a total for the window and per-interval buckets. The same query a dashboard widget runs, without saving anything.

• measures: what to aggregate. A METRIC measure names its metric by slug; CONVERSATION counts conversations; UPTIME reads health checks. The aggregation defaults by the metric's output type.
• timeRange: LAST n DAYS/WEEKS/MONTHS, a CALENDAR_ALIGNED period (TODAY, THIS_MONTH, ...), FIXED from/to, or SINCE. Defaults to the last 30 days; at most 731 days. Calendar periods and buckets align to timezone (IANA, default UTC).
• filters apply to every measure; a measure's own filters add to them.
• breakdown on a measure returns one series per value (largest 50 kept).
• comparisonGroups (2 to 5) run every measure once per named filter set; comparisonMode: PREVIOUS_PERIOD adds the preceding window's values.

Values are in the metric's display units. A PASS_RATE is a 0 to 1 share; sampleSize 0 means no data. A query the analytics store cannot answer returns 400 naming the unsupported part.

{
  "measures": [
    {
      "metric": "sentiment_score",
      "aggregation": "AVG"
    },
    {
      "type": "CONVERSATION",
      "aggregation": "COUNT"
    }
  ],
  "timeRange": {
    "type": "LAST",
    "value": 7,
    "unit": "DAYS"
  },
  "interval": "DAY",
  "timezone": "Europe/London",
  "filters": [
    {
      "type": "MODALITY",
      "value": "call"
    }
  ]
}