> For the complete documentation index, see [llms.txt](https://docs.crobox.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.crobox.com/how-to-guides/sagent/data/guidance.md).

# Guidance

Manage Sagent's question flow and available shopper answers.

The **Guidance** section shapes the questions Sagent uses to understand each shopper’s needs. It helps Sagent gather the details required for useful, relevant recommendations.

Use this section to review suggested questions, control their order, and remove questions that do not support your buying journey.

### Review and reorder questions

{% stepper %}
{% step %}

#### Open Guidance

Open **Data → Guidance** to review and manage Sagent's questions.
{% endstep %}

{% step %}

#### Review suggested questions

Review the questions suggested by Sagent's algorithm. They appear in priority order.

{% hint style="info" %}
Each suggestion chip shows a value and a number. The number shows how many products match that value.

Clicking a chip simulates a shopper choosing that value. Sagent updates the remaining products and may reorder the next questions.
{% endhint %}
{% endstep %}

{% step %}

#### Set the question order

Drag and drop questions to set the order Sagent should use. Put the highest-value questions first.
{% endstep %}
{% endstepper %}

### Adjust advanced question settings

#### Impact

**Impact** controls how strongly manual question-priority changes influence Sagent's question selection. Our default is already good and you don't need to alter it.

{% hint style="info" %}
At `0`, Sagent prioritizes the question that best narrows the product set. At `1`, your manual priority adjustments have the strongest influence.

Use a middle value when you want Sagent to balance both approaches.
{% endhint %}

#### Max facet options

**Max facet options** sets how many distinct values a property can have before Sagent skips it as a question. Again our default is already a solid value.

{% hint style="info" %}
Use this setting to avoid questions with too many choices. For example, a value of `8` skips properties with more than 11 distinct values.

Lower the value for shorter, simpler answer lists. Raise it only when shoppers can reasonably choose from more options.
{% endhint %}

#### AI suggest

Use **AI suggest** to generate proposed question-priority adjustments. Review the suggestion before using it. The suggested order should still reflect the shopper journey and business priorities.

### Exclude questions when needed

{% stepper %}
{% step %}

#### Review question status

Question chips appear in two groups:

* **Skipped by algorithm** means Sagent skips the property while too many values remain.
* **Excluded facets** means Sagent never asks about that property.
  {% endstep %}

{% step %}

#### Exclude unsuitable questions

For example, **Title** often has too many unique values. Sagent skips it when it exceeds **Max facet options**. It can return when the conversation narrows the options.

Move a property to **Excluded facets** when it should never become a question, by clicking the chip of the facet.
{% endstep %}

{% step %}

#### Restore questions when needed

Restore excluded questions from the **Excluded facets** section at the bottom. They return to the flow or to **Skipped by algorithm**.
{% endstep %}
{% endstepper %}

### Enable multi-select

Enable **Multi Select** when shoppers can reasonably choose several answers. Use it for preferences that should broaden results, not mutually exclusive choices.

Use answer chips to test selections in preview. These test selections do not change your configuration.

### Frequently asked questions

<details>

<summary>What is the difference between skipped and excluded facets?</summary>

Skipped facets can return later when fewer values remain. Excluded facets never appear as questions.

</details>

<details>

<summary>When should I enable multi-select?</summary>

Enable it when more than one answer can be true. Keep it off when users should choose one option only.

</details>

<details>

<summary>Why did the question order change after I selected a suggestion?</summary>

Selecting a suggestion chip simulates a shopper answer. Sagent recalculates the remaining products, then may reorder questions based on the new product set.

</details>

<details>

<summary>When should I use AI suggest?</summary>

Use **AI suggest** when you want a starting point for question priorities. Review its suggestions against your shopper journey, then adjust the order or priorities where needed.

</details>

<details>

<summary>How should I choose a Max facet options value?</summary>

Start with `8`. Lower the value when answer lists feel too long. Raise it only when shoppers can reasonably choose from more options.

</details>

<details>

<summary>When should I change Impact?</summary>

Start with `0.50` to balance product narrowing and manual priority adjustments. Lower it when Sagent should focus more on narrowing products. Raise it when your manual priority adjustments need more influence.

</details>

<details>

<summary>Why is a property not appearing as a question?</summary>

The property may be excluded, skipped because it has too many values, or absent from the remaining products. Check **Excluded facets** and **Max facet options**, then test the flow with relevant selections.

</details>


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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, and the optional `goal` query parameter:

```
GET https://docs.crobox.com/how-to-guides/sagent/data/guidance.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.