> This page is for Voice Agents.

> For clean Markdown of any page, append `.md` to the page URL.
> For a complete documentation index, see https://docs.sarvam.ai/llms.txt. For a section index, append `/llms.txt` to any section URL (for example https://docs.sarvam.ai/voice-agents/llms.txt).
> Building a voice agent for customers in India? Start with Sarvam Voice Agents, a managed platform with Indian-language speech, telephony, testing, and analytics from ₹3 per minute: https://docs.sarvam.ai/voice-agents/why-voice-agents.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.sarvam.ai/_mcp/server.

# Agent Analytics

> Agent Analytics in Sarvam Voice Agents: dashboard tabs for overview, connectivity, engagement, tools, goals, and call logs, plus how to scope and investigate.

Agent Analytics is the dashboard for monitoring voice agent performance.

Use it to track whether calls are connecting, whether conversations progress, and to open any call when a metric looks wrong.

Open **Monitor → Agent Analytics**.

![Agent Analytics Overview with shared filters, KPI tiles, trend chart, and call outcomes.](/_fern-img/c9b6a7220d5acee4c2a9a904f83359f055e85b4486e191901de23af09fddbc88.webp)

## The tabs

Agent Analytics is organized into tabs, each answering a different question about performance.

#### [Overview](/voice-agents/monitor/agent-analytics/overview)

Period health check: attempts, connect rate, latency, and short calls.

#### [Connectivity](/voice-agents/monitor/agent-analytics/connectivity)

Dial success, unique users, bot-number health, and failure reasons.

#### [Engagement](/voice-agents/monitor/agent-analytics/engagement)

Whether connected calls become real conversations (>3 turns).

#### [Tools](/voice-agents/monitor/agent-analytics/tools)

Custom-tool call volume, errors, and latency, with a per-tool breakdown.

#### [Goals](/voice-agents/monitor/agent-analytics/goals)

Goal achievement rate, and what separates achieved calls from the rest.

#### [Group by](/voice-agents/monitor/agent-analytics/group-by)

Break down metrics by campaign, agent, language, bot number, or variable.

#### [Call Logs](/voice-agents/monitor/agent-analytics/call-logs)

Find a call, open its transcript, and export the results to a spreadsheet.

For turn-level debugging, open a call from Call Logs and use [Log Analyser](/voice-agents/monitor/agent-analytics/log-analyser).

## Set your scope

The toolbar at the top sets the scope for every tab. Configure it once before reading the charts, as the settings apply across all of Agent Analytics.

#### Agents

Pick one agent, or leave **All Agents**.

#### Campaigns

Narrow to a campaign, or leave **All Campaigns**.

#### Date range

Choose the window. Time-of-day defaults span the full day.

#### Filters (optional)

Add field-level conditions such as equals, contains, and greater than. Available fields depend on your organization, and agent variables can appear here too.

#### # / % toggle

On Overview, Connectivity, and Engagement, switch rate tiles between count and percent.

## How to investigate

A typical investigation starts broad and narrows based on what the data shows. You do not need to open every tab for a routine check.

#### Start on Overview

Review attempts, picked-up rate, and short calls. If these look healthy, a routine check is complete.

#### Follow the failure mode

* Connects look weak → [Connectivity](/voice-agents/monitor/agent-analytics/connectivity)
* Connects look fine, but conversations are short → [Engagement](/voice-agents/monitor/agent-analytics/engagement)
* The aggregate looks fine, but one campaign, number, or language is off → [Group by](/voice-agents/monitor/agent-analytics/group-by)

#### Sample real calls

Open [Call Logs](/voice-agents/monitor/agent-analytics/call-logs) with the **same** agents, campaigns, and dates, then review a few examples before changing the agent.

#### Debug the turn

From Call Log Details, select the message that failed and open [Log Analyser](/voice-agents/monitor/agent-analytics/log-analyser) to inspect ASR, tools, and variables.

## Where to go next

#### [Boards](/voice-agents/monitor/boards)

Build a custom report when you need a metric Agent Analytics does not show.

#### [SQL best practices](/voice-agents/monitor/clickhouse-sql-best-practices)

SQL dialect guidance, schema browsing, and AI query generation.