> 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 full documentation content in one file, see https://docs.sarvam.ai/llms-full.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.sarvam.ai/_mcp/server.

# Overview

Monitor is where you review how your deployed agents perform in production. It brings together connection health, conversation quality, individual call records, and custom reporting in one place.

Open **Monitor**. It contains two areas:

* **Agent Analytics** — ready-made dashboards for day-to-day health checks and call investigation, covering connect rates, engagement, call logs, and turn-level debugging.
* **Boards** — custom reports you build with SQL when the fixed Agent Analytics tabs do not cover what you need.

## Choosing where to start

Use Agent Analytics for most routine monitoring, and move to Boards when you need a report it does not provide.

#### Check production health

Open [Agent Analytics](/conversations/monitor/agent-analytics). Start on Overview, then move to Connectivity or Engagement depending on whether the issue is with dialing or conversation quality.

#### Investigate a single call

Open [Call Logs](/conversations/monitor/agent-analytics/call-logs), select the call, then use [Log Analyser](/conversations/monitor/agent-analytics/log-analyser) on the turn that failed.

#### Build a custom report

Use [Boards](/conversations/monitor/boards) to write SQL, or describe the question and let AI draft the query.

For SQL dialect guidance, schema browsing, and AI query generation, see [SQL best practices](/conversations/monitor/clickhouse-sql-best-practices).

## Where to go next

#### [Agent Analytics](/conversations/monitor/agent-analytics)

Connect rates, engagement, call logs, and turn-level debugging, out of the box.

#### [Boards](/conversations/monitor/boards)

Build your own dashboards with SQL, charts, and reusable filters.