# User feedback

### What are Feedback collections?

Feedback collections let you capture quick structured user feedback on an AI response (just like the 👍/👎 controls you see with popular LLM chat experiences). Because a feedback collection is a global workspace resource, you can reuse the same collection across user nodes and basic nodes.

If using [Touchpoint](/platform/nlx-platform-guide/ai-applications/deployment/managing-channels/creating-an-api-channel.md#touchpoint-api-channel), feedback collections are rendered automatically in the chat UI with no additional frontend work required. If you prefer a fully branded or custom experience, you can render the feedback UI in your own frontend and submit feedback using your own interface.

{% hint style="info" %}
Feedback collections are currently supported in chat-based experiences. Voice feedback collections are unsupported at this time.
{% endhint %}

Common use cases:

* Measure response quality (helpful vs. not helpful) on key moments like knowledge base answers
* Identify knowledge gaps (patterns in bad ratings and comments)
* Compare experiences across flows or versions of an application
* Add pulse checks after high-risk turns (refunds, account access, compliance messaging)

To access, click *Resources* in your workspace menu and choose *Feedback collections*

{% @arcade/embed flowId="jE758tAXzNKERQHU03cK" url="<https://app.arcade.software/share/jE758tAXzNKERQHU03cK>" %}

### Set up a collection

{% @arcade/embed flowId="hZzvbUNnsEajkKvTi4ua" url="<https://app.arcade.software/share/hZzvbUNnsEajkKvTi4ua>" %}

Create and configure a feedback collection once and then reuse it anywhere.

When setting up a New feedback collection, set the following on its *Configuration* tab:

* *Name*: Internal name for the collection (e.g., “Post-answer rating”)

Optional:

* *Comments* (toggle): When enabled, users can leave an optional written comment
* *Question* (toggle): The prompt shown to the user (e.g., “How was your experience?”)
  * NOTE: Keep the question short and neutral, as users respond more when it feels effortless
* *Labels*: Customize the text shown to users for each rating option (e.g., Good/Bad), including the *Comments* label, if enabled (e.g., Tell me more)

#### Option 1: Create from Resources page

1. Select *Resources* in your workspace menu and choose the *Feedback collections* card&#x20;
2. Select *New feedback collection*
3. Configure the collection fields and click *Save*

#### Option 2: Create while in a flow

You can also create a feedback collection directly from a node in your flow:

1. Choose an eligible node (User choice, User input, or Basic)
2. Click *+Add functionality* and select *Collect feedback*
3. In the collection dropdown, select +*Create new collection*
4. Configure the collection fields and click *Save*

### Use a collection

{% @arcade/embed flowId="YcoTBwMsHNYrb6GdPen2" url="<https://app.arcade.software/share/YcoTBwMsHNYrb6GdPen2>" %}

Attach feedback collection prompts directly to a node so the user can rate the preceding experience.

1. Open a flow and select the node where you want to request feedback (only user or basic nodes)
2. In the node’s side panel, select *+ Add functionality*
3. Choose *Collect feedback*
4. Select the feedback collection (e.g., "Post-answer rating") from the dropdown
5. Click *Save* on the Canvas

### Monitor feedback

<figure><img src="/files/ayG0nuBvt5O6JVTIGYSV" alt=""><figcaption><p>Feedback tab on application</p></figcaption></figure>

Once your flow is deployed with an app and users begin responding, you can review results per application.

1. Go to *Applications* in your workspace menu and select your app
2. Click the *Feedback* tab
3. Choose the feedback collection (if more than one appear in flows attached to your app) from the dropdown to view:
   * Rating distribution (Good vs. Bad)
   * Entries list, including:
     * Timestamp
     * Source (where it occurred)
     * User utterance
     * The application message that was rated
     * The rating and any comment

{% hint style="info" %}
Use search and filters to narrow down results to specific periods or patterns.
{% endhint %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.nlx.ai/platform/nlx-platform-guide/monitoring/user-feedback.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.