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

# Create Widgets

Follow these steps to create a new AI-powered recommendation widget in Experro:

**1. Open the Recommendations Tab**

* In the left navigation panel, click **Discovery → Recommendations**.
* The menu has three further options: **Home**, **Analytics**, and **Widgets**.
* Select **Widgets** to see your existing widgets and then click on the **Add Widget** button to create a new widget.

  ![](https://mintcdn.com/experro/4fPtPqhyselHfqmM/images/discovery/recommendation_homescreen.png?fit=max\&auto=format\&n=4fPtPqhyselHfqmM\&q=85\&s=bdaa3e047006e88006dcc29c2a2c97ee)

**2. Launch the Add Widget Flow**

* Click **Add Widget**.
* A full-screen panel appears listing all available recommendation algorithms (RFY, Frequently Bought Together, Similar Products, etc.).

  ![](https://mintcdn.com/experro/8leTCfwCNMk4X8PP/images/discovery/create_widget.png?fit=max\&auto=format\&n=8leTCfwCNMk4X8PP\&q=85\&s=98c9ce0fe2e584add2c2afcfba31b66a)

**3. Choose Your Algorithm**

* Find and click the algorithm that fits your goal (e.g. “Recommended For You” for personalized picks).
* Each algorithm tile includes a brief use case description.

**4. Name & Describe**

* In the modal that appears, fill out:
  * **Widget Name**: A descriptive title (e.g. “Homepage RFY Carousel”).
  * **Description** (optional): Notes for your team about placement, audience, or context.
* Click **Save** to create your widget.

  ![](https://mintcdn.com/experro/8leTCfwCNMk4X8PP/images/discovery/create_widget_details.png?fit=max\&auto=format\&n=8leTCfwCNMk4X8PP\&q=85\&s=ab830bf6e27ea95dd635f163066a3139)

**5. Configure Your Default (Global) Rule**

* You’re taken to the widget’s Rules screen.
* A **Global** rule is selected by default—any merchandising or fallback settings configured here apply wherever this widget is rendered (homepage, PDP, cart).
* Global rule is mandatory; you cannot remove it.

**6. Add Page-Level Overrides (Optional)**

* To customize behavior on specific pages, click **Add Rule** and provide the following details :

  * **Rule Name**: A descriptive title.
  * **Description**: Notes for your team.
  * **Rule Type**: Select Homepage, Product, or Cart based on the context you want to target.

  ![](https://mintcdn.com/experro/8leTCfwCNMk4X8PP/images/discovery/custom_rule_widget.png?fit=max\&auto=format\&n=8leTCfwCNMk4X8PP\&q=85\&s=e333611a0d1b95ce9723772a61ff380d)

* Each page-level rule overrides the Global rule for that context, whether Homepage, Product, or Cart page.\
  \[!NOTE] You may only have one rule active for each context in a given time window.

**7. Edit a Rule**

* Click the **Edit Rule** option against the global rule or the **Edit** option under **Actions** next to custom created widget rules.
* Toggle **Merchandising** on/off and configure boost/pin/slot/sort/include/exclude rules.
* Toggle **Fallback Algorithm** on/off and select your backup recommendation engine.
* Click **Save** within the rule editor to apply.

  ![](https://mintcdn.com/experro/4fPtPqhyselHfqmM/images/discovery/edit_custom_rule_widget.gif?s=fd607eb24289aec7d651b2c3d81f1bae)

**8. Activate & Publish**

* Once all rules and settings are configured, toggle your widget **Active** if it is not already active.
* Navigate to **Content Library** to map the widget you created to the storefront.
* Read about [Publishing Widgets on Storefront](https://help.experro.com/experro_discovery/recommendations/mapping_widgets_to_storefront.md) to publish the widget on your storefront.

> **Key Concepts**
>
> * **Global Rule:** Applies everywhere the widget appears.
> * **Page-Level Rule:** Overrides Global on homepage, product, or cart.
> * **Only one rule per page type** may be active at any given time.
> * **Merchandising + Fallback** toggles live inside each rule’s editor.

Now that your widget and rules are in place, explore [Recommendation Algorithms](https://help.experro.com/experro_discovery/recommendations/recommendation_algorithms) to learn more about each algorithm.


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