Get Metrics

curl --request GET \
  --url https://api.eqho.ai/v2/reports/metrics \
  --header 'X-API-KEY: <api-key>' \
  --header 'x-org-id: <api-key>'

import requests

url = "https://api.eqho.ai/v2/reports/metrics"

headers = {
    "X-API-KEY": "<api-key>",
    "x-org-id": "<api-key>"
}

response = requests.get(url, headers=headers)

print(response.text)

const options = {method: 'GET', headers: {'X-API-KEY': '<api-key>', 'x-org-id': '<api-key>'}};

fetch('https://api.eqho.ai/v2/reports/metrics', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.eqho.ai/v2/reports/metrics"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("X-API-KEY", "<api-key>")
	req.Header.Add("x-org-id", "<api-key>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

{
  "metrics": {},
  "filters_applied": {
    "created_gte": "2023-11-07T05:31:56Z",
    "created_lt": "2023-11-07T05:31:56Z",
    "account_id": "<string>",
    "time_granularity": "hour",
    "group_by_dimension": "campaign",
    "campaign_ids": [
      "<string>"
    ],
    "agent_ids": [
      "<string>"
    ],
    "disposition_level_3": [
      "<string>"
    ],
    "conversation_mediums": [
      "web_voice"
    ],
    "call_types": [
      "outbound"
    ],
    "twilio_statuses": [
      "queued"
    ],
    "is_dry_run": true
  },
  "metadata": {
    "generated_at": "2023-11-07T05:31:56Z",
    "total_records_analyzed": 123,
    "generation_time_ms": 123,
    "errors": [
      "<string>"
    ],
    "partial_results": false
  }
}

{
  "detail": [
    {
      "loc": [
        "<string>"
      ],
      "msg": "<string>",
      "type": "<string>"
    }
  ]
}

Reports

Get Metrics

Get analytics metrics with optional granularity and grouping

Available Metrics:

Efficiency:

autonomous_containment_rate - Percentage of conversations resolved without human intervention
human_transfer_rate - Percentage of conversations transferred to human agents
conversions - Total number of conversions (disposition.conversion=True)
conversion_rate - Percentage of conversations that converted (call-based)
lead_conversion_rate - Percentage of unique leads that converted (lead-based, useful for outbound sales)
contact_rate - Percentage of conversations where contact was made (call-based)
lead_contact_rate - Percentage of unique leads contacted (lead-based, useful for outbound sales)

Quality:

average_latency_e2e - End-to-end response latency (total time)
average_latency_stt - Speech-to-text processing latency
average_latency_ai - AI processing latency
average_latency_tts - Text-to-speech processing latency

Volume:

total_conversations - Total number of conversation sessions
unique_leads_contacted - Count of distinct leads contacted (by lead_id)
first_time_contacts - Count of leads contacted for the first time (call_count=1)
average_conversation_duration - Average length of conversations

Cost:

total_credits_spent - Total credits consumed across all conversations

Conversation Dynamics:

average_user_talk_time - Average seconds the caller spoke per call
average_agent_talk_time - Average seconds the agent spoke per call
user_to_agent_talk_ratio - Ratio of total user talk-time to total agent talk-time (period-aggregated, not the average of per-call ratios)
average_total_silence - Average seconds of silence per call (neither speaker active)
average_longest_silence - Average length of the longest single silence per call
average_silence_frequency - Average count of silences over 3s per minute of conversation
average_words_per_agent_response - Average word count per non-seeded assistant message

Grouping:

time_granularity - Time bucket for aggregation: hour, day, week, month, year
group_by_dimension - Combine time series with dimensional breakdown
- Values: campaign, agent, disposition, conversation_medium, call_type
- When used with time_granularity, each time point includes a breakdown array
- Example: Track campaign performance over time in a single query

Filters:

campaign_ids - Filter by specific campaign IDs (comma-separated)
agent_ids - Filter by specific agent IDs (comma-separated)
disposition_level_3 - Filter by level 3 disposition names (comma-separated)
conversation_mediums - Filter by channel: telephony, web_voice, sms, chatbot, email
call_types - Filter by call direction: inbound, outbound
twilio_statuses - Filter by call status (default: all statuses included)
- NOTE: For telephony/web_voice, ‘completed’ indicates successful calls
- For chatbot/email/SMS, conversations occur regardless of status
- To get telephony-only metrics, use: ?twilio_statuses=completed&conversation_mediums=telephony
is_dry_run - Filter by dry-run status: true (only dry-runs), false (exclude dry-runs), omit (all calls, default)
- NOTE: Some organizations bill dry-run calls, so the default includes all calls for accurate reporting

Examples:

# All metrics
?metrics=autonomous_containment_rate,human_transfer_rate,average_latency_e2e,total_conversations,average_conversation_duration,unique_leads_contacted,first_time_contacts,conversions,conversion_rate,lead_conversion_rate,contact_rate,lead_contact_rate,total_credits_spent

# Lead volume analysis
?metrics=unique_leads_contacted,first_time_contacts&time_granularity=day&created_gte=2024-01-01

# Outbound sales: Compare call-based vs lead-based metrics
?metrics=contact_rate,lead_contact_rate,conversion_rate,lead_conversion_rate&call_types=outbound&time_granularity=day

# Multi-dimensional time series (campaign performance over time)
?metrics=unique_leads_contacted,conversions&time_granularity=day&group_by_dimension=campaign&created_gte=2024-01-01
# Returns time series where each point contains breakdown by campaign

# Campaign comparison over time
?metrics=total_conversations,conversion_rate&time_granularity=week&group_by_dimension=campaign
# See how each campaign performs week-over-week in single response

# Filter by specific level 3 dispositions
?metrics=total_conversations,conversion_rate&disposition_level_3=Interested,Converted&time_granularity=day

# For telephony-only view, explicitly filter:
?metrics=total_conversations&twilio_statuses=completed&conversation_mediums=telephony&is_dry_run=false

# Get all conversations including chatbot:
?metrics=total_conversations&time_granularity=day

# Conversation dynamics dashboard
?metrics=average_user_talk_time,average_agent_talk_time,user_to_agent_talk_ratio,average_silence_frequency&time_granularity=day

GET

reports

metrics

Get Metrics

curl --request GET \
  --url https://api.eqho.ai/v2/reports/metrics \
  --header 'X-API-KEY: <api-key>' \
  --header 'x-org-id: <api-key>'

import requests

url = "https://api.eqho.ai/v2/reports/metrics"

headers = {
    "X-API-KEY": "<api-key>",
    "x-org-id": "<api-key>"
}

response = requests.get(url, headers=headers)

print(response.text)

const options = {method: 'GET', headers: {'X-API-KEY': '<api-key>', 'x-org-id': '<api-key>'}};

fetch('https://api.eqho.ai/v2/reports/metrics', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

package main

import (
	"fmt"
	"net/http"
	"io"
)

func main() {

	url := "https://api.eqho.ai/v2/reports/metrics"

	req, _ := http.NewRequest("GET", url, nil)

	req.Header.Add("X-API-KEY", "<api-key>")
	req.Header.Add("x-org-id", "<api-key>")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

{
  "metrics": {},
  "filters_applied": {
    "created_gte": "2023-11-07T05:31:56Z",
    "created_lt": "2023-11-07T05:31:56Z",
    "account_id": "<string>",
    "time_granularity": "hour",
    "group_by_dimension": "campaign",
    "campaign_ids": [
      "<string>"
    ],
    "agent_ids": [
      "<string>"
    ],
    "disposition_level_3": [
      "<string>"
    ],
    "conversation_mediums": [
      "web_voice"
    ],
    "call_types": [
      "outbound"
    ],
    "twilio_statuses": [
      "queued"
    ],
    "is_dry_run": true
  },
  "metadata": {
    "generated_at": "2023-11-07T05:31:56Z",
    "total_records_analyzed": 123,
    "generation_time_ms": 123,
    "errors": [
      "<string>"
    ],
    "partial_results": false
  }
}

{
  "detail": [
    {
      "loc": [
        "<string>"
      ],
      "msg": "<string>",
      "type": "<string>"
    }
  ]
}

Use each metric’s time_series to build period rows. All buckets intersecting [created_gte, created_lt) are included, cut in timezone (an IANA name) with Monday-based weeks. When omitted, timezone defaults to the organization’s reporting time zone, or UTC if none is configured. An explicit timezone overrides the organization setting. Each timestamp is the UTC instant the period starts, e.g. 2026-09-22T04:00:00Z for September 22 in America/New_York. Missing buckets have value: 0 for counts and totals, value: null for rates and averages, and breakdown: []. The first bucket can start before created_gte when the requested range begins partway through a period. Add breakdown_by=disposition to get disposition counts on each total_conversations time point:

{
  "timestamp": "2026-09-01T00:00:00Z",
  "value": 12,
  "components": { "not_engaged": 7, "converted": 3 }
}

With group_by_dimension=campaign (or another supported dimension) each breakdown item carries its own counts as well, and the point’s components sums them:

{
  "dimension_value": "campaign-id",
  "dimension_label": "Sales",
  "value": 12,
  "count": 12,
  "components": { "not_engaged": 7, "converted": 3 }
}

Calls without a disposition contribute to the total but have no component key. Other metrics retain their existing breakdown shape. Breakdowns include all matching dimension values without a top-20 limit. custom_fields[key]=value filters every metric to conversations whose custom field holds exactly that string, the same way GET /v1/calls does, so a drill-down lists the conversations a cell counted. Several pairs must all match. Helper calls are excluded. Rate denominators remain unchanged; use each rate point’s denominator for the Dispositioned column. For exact call drill-downs, GET /v1/calls accepts disposition_contacted and disposition_conversion. Both accept true or false; omitting a flag applies no filter for that flag. Requests with more than 30 metric IDs return HTTP 400. Each aggregation uses the ANALYTICS_MAX_TIME_MS server setting (default: 30000 milliseconds).

Authorizations

X-API-KEY

string

header

required

Your API key for authentication.

x-org-id

string

header

required

Org impersonation via x-org-id. Superadmins may impersonate any org; admins may impersonate their direct sub-organizations. Enter the target organization ID to act as that org.

Query Parameters

metrics

string[]

required

Comma-separated list of metric IDs. Call GET /v2/reports/metrics/available to discover the IDs this account can request. Any unknown or not-enabled ID rejects the whole request with 400.

Minimum array length: 1

created_gte

string<date-time>

Show conversations that started on or after this datetime (inclusive). Must be in UTC using ISO 8601 format (e.g., 2025-07-05T00:00:00Z). Defaults to 7 days ago.

created_lt

string<date-time>

Show conversations that started before this datetime (exclusive). Must be in UTC using ISO 8601 format (e.g., 2025-07-05T23:59:59Z). Defaults to now.

account_id

string | null

Account ID to filter by

time_granularity

enum<string> | null

Time bucket for aggregation. When omitted, only current values are calculated (no time series).

Available options:

hour,

day,

week,

month,

year

group_by_dimension

enum<string> | null

Dimension to include WITHIN time series.

Available options:

campaign,

agent,

disposition,

conversation_medium,

call_type

campaign_ids

string[] | null

Comma-separated list of campaign IDs to filter by

agent_ids

string[] | null

Comma-separated list of agent IDs to filter by

disposition_level_3

string[] | null

Comma-separated list of level 3 disposition names to filter by

conversation_mediums

enum<string>[] | null

Comma-separated list of conversation mediums to filter by.

Available options:

web_voice,

telephony,

sms,

email,

chatbot,

api_text,

chat_app,

softphone

call_types

enum<string>[] | null

Comma-separated list of call types to filter by.

Available options:

outbound,

inbound

twilio_statuses

enum<string>[] | null

Comma-separated list of Twilio call statuses to filter by. When omitted, includes all statuses. Note: For telephony/web_voice, 'completed' typically indicates a successful call, but for chatbot/email/SMS, conversations occur regardless of status.

Available options:

queued,

initiated,

ringing,

in-progress,

completed,

busy,

failed,

no-answer,

canceled,

voicemail

is_dry_run

boolean | null

Filter by dry-run status. True=only dry-runs, False=exclude dry-runs, None=include all (default). Note: Some organizations bill dry-run calls, so excluding them may underreport actual usage.

enable_trends

boolean

default:false

Include trend comparison with previous period (calculates change_value, change_percent, trend direction)

Response

Successful Response

Main response from analytics API

metrics

Metrics · object

required

Show child attributes

filters_applied

FiltersSummary · object

required

Summary of filters applied to the query

Show child attributes

metadata

ResponseMetadata · object

required

Metadata about the response

Show child attributes

API Reference

Endpoints

Get Metrics

Authorizations

Query Parameters

Response