> ## Documentation Index
> Fetch the complete documentation index at: https://docs.ringg.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> To search or read these docs, connect the Ringg docs MCP server at https://docs.ringg.ai/mcp (no key needed; tools search_ringg_ai and query_docs_filesystem_ringg_ai) instead of browsing pages one by one. Setup for each client: https://docs.ringg.ai/resources/mcp-server/docs-mcp-server/mcp-for-ai-assistants.md
> The dashboard labels agents "Assistants"; these docs say "agent". Both mean the same object, identified by agent_id.
> API base URL: https://prod-api.ringg.ai/ca/api/v0. Send the workspace API key in the X-API-KEY header, from server-side code only.
> Placing a call dials a real phone number and spends credits. Confirm the number with the user before calling.
> To act on a Ringg workspace (place calls, edit agents, read transcripts), use the separate Ringg workspace MCP server at https://mcp.ringg.ai/mcp: https://docs.ringg.ai/resources/mcp-server/workspace-mcp-server/overview.md

# QA reviews

> A QA re-scores a call by hand, parameter by parameter, and sees their Human score next to the AI's. Reviews can trigger an email to the agent.

A QA review is a QA's own scoring of a call, done whenever they choose. It sits next to the AI's score and doesn't replace it; to change the official score of a call, the agent [disputes](/analytics-platform/disputes/overview) it and a QA resolves the dispute.

In [Ringg Analytics](https://analytics.ringg.ai): **Sidebar → Interactions →** open a call. Owners, admins and QAs can review.

## Review a call

**1. Open the call**

Open it from **Interactions**, or **Open full page** for more room.

**2. Rate the parameters you check**

Expand a category and pick a rating for each parameter you check, with a comment saying why. Parameters you don't rate keep the AI's rating in your **Human** score.

**3. Mark the call fatal if it is**

Tick **Mark interaction as fatal** to set the Human score to 0, whatever the ratings, and say why.

**4. Save**

Click **Save review**, or **Discard** to drop your changes.

> [!TIP]
>
> The scorecard shows **AI** and **Human** scores side by side.

![Call Analysis for a Missed EMI follow-up call by Arjun Mehta with the recording, the transcript, and AI 85 next to Human 95, a View detailed comparison link and the Mark interaction as fatal option](https://mintcdn.com/ringgai-0da73d20/CZ_hnbckkb2zP9lz/media/analytics-qa-reviews/review.webp?fit=max\&auto=format\&n=CZ_hnbckkb2zP9lz\&q=85\&s=8956f2699865de1ba00ac12b8c6d56cc)

**View detailed comparison** lists the parameters where you and the AI disagree, or says "Fully aligned". Every save is kept in the call's review history, marked **Added**, **Edited** or **Cleared**, with who made it.

## Rules

* A QA set to **Rate and comment on every parameter** can't save until each parameter has a rating and a comment: "Rate and comment on every parameter to save: N left." Clearing a whole review is always allowed.
* A QA in **Blind review** sees AI ratings, reasons and summaries as `***` while they score, so their ratings stay independent. Admins set both on **Team** (see [QA settings](/analytics-platform/roles-and-access#qa-settings)).
* Use **Human Evaluation** in the **Interactions** filter to list reviewed or unreviewed calls.

## Share reviews with agents

To email the agent each time a QA saves a review, add a notification trigger that starts when **A QA reviews a call**. The ready-made **QA Review Shared** template lists the QA's score against the AI's and every rating and comment. See [Notifications](/analytics-platform/set-up/notifications).

## Next steps

> **[Notifications](/analytics-platform/set-up/notifications)**
>
> Email agents when a QA reviews their call.

> **[Review a dispute](/analytics-platform/disputes/review-a-dispute)**
>
> Re-score a call an agent disputed.

> **[Report cards](/analytics-platform/coach-agents/report-cards)**
>
> Turn reviews into coaching.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.