> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.ada.cx/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.ada.cx/_mcp/server.

# Update Dixa CSAT

After a handoff ends and the end user returns to the AI Agent, you can pass a CSAT rating back to the Dixa conversation. This lets the CSAT collected by Ada be recorded against the original Dixa conversation.

## Prerequisites

Before configuring CSAT updates, make sure the following are in place:

* The Dixa Utilities app and the Dixa handoff are configured.
* The AI Agent is configured to wait for the handoff to finish, so the CSAT update runs after the handoff ends and the end user is back with the AI Agent.
* The Dixa rating type is decided. Dixa supports `Csat` (values 1–5) and `ThumbsUpOrDown` (values 0–1). Use `Csat` for a 1–5 scale, or `ThumbsUpOrDown` for a binary thumbs up / thumbs down rating. The rating type must match in both API calls described below. See the Dixa API reference for [offering a rating](https://docs.dixa.io/openapi/dixa-api/v1/tag/Ratings/#tag/Ratings/operation/postConversationsConversationidRatingsOffer) and [submitting a rating](https://docs.dixa.io/openapi/dixa-api/v1/tag/Ratings/#tag/Ratings/operation/postConversationsConversationidRatingsRatingidSubmit).

> **Note**
>
> Enabling the wait-for-handoff behavior requires a setting that is enabled by your Ada team. Contact your Ada representative to have it configured.

## Configuration

When a handoff starts, three metavariables are stored on the conversation and are used to submit the rating back to Dixa:

* `agent_system_dixa_user_id`
* `agent_system_dixa_conversation_id`
* `agent_system_dixa_agent_id`

![The three Dixa metavariables stored on a conversation in Ada's Conversation dashboard](/_fern-img/b0a2c5b9fad90e6c2e8cbb009eaadfd5897721b12c6d3135422dcbddbd1b3bf5.webp)

**To configure the CSAT update:**

1. On the **Human Agent** tab in the Ada dashboard, enable **Human Agent Survey for chat** so that end users are prompted with the survey when they close the human agent chat.

![The Human Agent CSAT survey setting enabled in the Ada dashboard](/_fern-img/f672e458e82be6f003c27e9f4ac8af7da151825023a44bb0ef9c2dfcb2cb8047.webp)

2. In the **Handoff to agent** action, select both the **Pause conversation here until handoff ends** and **Track as Handoff** check boxes.

![The Pause conversation here until handoff ends and Track as Handoff check boxes selected](/_fern-img/bf6f227fc0c535cdbc85eec3490b6ac8d8e88d650e5e7538c09c514df025b63a.webp)

3. Add two HTTP Request blocks to submit the CSAT rating to Dixa.
   * The first block offers a rating on the conversation. Send a `POST` request to the Dixa ratings `offer` endpoint for `agent_system_dixa_conversation_id`, with the `dixa token` in the **Authorization** header. In the body, set `userid` to `agent_system_dixa_user_id`, `agentId` to `agent_system_dixa_agent_id`, and `ratingType` to your chosen rating type. Save the returned `data.id` as a variable to use in the next request.

![The first HTTP Request block posting to the Dixa ratings offer endpoint](/_fern-img/59f3cdc17dab9e4b6a463ea5df7f99af4003106830d188e276c12549dfc836ab.webp)

* The second block submits the rating value. Send a `POST` request to the Dixa ratings `submit` endpoint, using the conversation ID and the rating ID returned by the first block. In the body, set `ratingResult.value` to the captured CSAT score and `ratingResult._type` to the same rating type used above.

![The second HTTP Request block posting to the Dixa ratings submit endpoint](/_fern-img/4af0332e009ba94a8894a6ef0efa05f9a2afb634fcef8019261ef9d5c98717a3.webp)

---

Have any questions? Contact your Ada team, or email us at [](mailto:help@ada.cx?subject=Help%20Docs%20inquiry).