> For the complete documentation index, see [llms.txt](https://helpcenter.channable.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://helpcenter.channable.com/manage-improve-product-data/ai-readiness-score/set-up-ai-readiness-attributes.md).
# Set up AI Readiness attributes
Generate, format, and map AI Readiness attributes for your Google Shopping channel.
Use AI-generated conversational attributes to enrich product data for Google Shopping.
### :memo-circle-check: Before you start
**You'll need**
* [A Core Pro plan](/account-billing/manage-your-subscription/channables-pricing-plans.md#core-plans-features)
* An active Google Shopping channel
* Conversion data imported, or a matching Google [Insights](/manage-improve-product-data/insights/channable-insights.md) dashboard (required to calculate `popularity_rank` values)
* A [Gemini or OpenAI subscription](/manage-improve-product-data/generate-with-ai/ai-generation-usage-and-costs.md) (to generate values for **Question and answer**, **Product highlights** and **Product details** in Channable)
**Good to know**
* It's best to start with a small group of products before expanding your setup
***
### Choose an attribute setup
Choose the setup that matches your attribute before following its instructions:
* Use AI generators and rule templates for **question and answer**, **product highlights**, and **product details**.
* Use the **popularity rank** rule template with Insights conversion data.
* Map existing product data for **related products**, **item group titles**, and **variant options**.
* Map document link URLs directly from a project field.
#### Attributes with AI generators and/or AI readiness rule templates
:messages-question: Question and answer (question_and_answer.question , question_and_answer.answer )
{% hint style="info" icon="lightbulb" %}
#### What this attribute is for
Generate question-and-answer pairs that help Google Shopping match products to shoppers' requirements.
For example, shoppers can ask whether shoes are waterproof or whether a portable charger includes a cable.
{% endhint %}
#### How to map in Channable
You can import question-and-answer values into project fields and map them in your Google Shopping feed.
With Core Pro and a [Gemini or OpenAI subscription](/manage-improve-product-data/generate-with-ai/ai-generation-usage-and-costs.md), you can also generate values in bulk with the **Question and answer** template.
{% embed url="" %}
{% stepper %}
{% step %}
#### Generate question-and-answer values
1. In the project containing your Google Shopping channel, go to **Generate with AI**.
2. Click **Generate with AI**. In the pop-up, select **Text**.
3. For **Text type** select **Question & answer**, and then click **Continue**.
4. Configure the generator settings:
1. At **Name**, enter a descriptive name, such as "Google Q\&A values".
2. At **AI connection**, [select the Gemini or OpenAI connection you want to use](/manage-improve-product-data/generate-with-ai/ai-generation-usage-and-costs.md).
3. At **Project fields to use to generate the text**, select the fields that provide context. By default, `description` and `title` are selected.
4. At **Field to save the optimized text**, create a project field for the generated values and name it `q_and_a`.
5. At **Output language**, select the language for generated values.
5. Click **Save & continue**.
6. At **Select items**, you can filter your items so you only generate values for specific items. If you leave this blank, we'll generate values for all your items.
1. If you add a condition, click **Save** to update the item count.
7. Click **Save & continue**.
8. At **Prompting**, review and edit the default prompt for Google's question-and-answer attribute.
9. Click **Generate 5 previews** to preview values for the selected items.
10. Review the previews, then click **Generate all**.
11. At **Generated values**, review the generated values before saving them to `q_and_a`:
* Click **Approve all** to approve every value.
* Click **Approve** or **Regenerate** to review values individually.
{% endstep %}
{% step %}
#### Extract values using the rule template
After generating values, use the **Q\&A** rule template to split them into separate question and answer project fields. Find the template in **AI Readiness rule templates**.
1. In the **Rules** step of your Google Shopping channel, click **Create rule**.
2. Select **AI Readiness rule templates > Q\&A**.
3. The template includes placeholders for the `questions` and `answers` project fields. Create both fields before completing the template.
4. Click `questions` > **+ Add new project field**.
5. Name your field `questions` and click **Create**.
6. Repeat steps 4 and 5 to create the `answers` field.
7. At `q_and_a`, select the project field where you saved the generated values.
8. Click **Save rule**. The rule splits the generated values into the `questions` and `answers` fields for the selected items.
{% endstep %}
{% step %}
#### Map your fields to Google's Question & Answer attributes
1. After creating your rule, go to the **Mapping** step.
2. Search for **question**.
3. Map the question and answer attributes:
1. For `question_and_answer.question`, map the `questions` field.
2. For `question_and_answer.answer`, map the `answers` field.
4. Click **Save**.
{% endstep %}
{% endstepper %}
Your Google Shopping feed can now send the mapped question-and-answer values for the selected items.
#### Extra: Configure question and answer by category
The default question-and-answer template works across product types, but you can create tailor your generator to specific categories.
{% hint style="info" %}
For clothing, use questions about fabric, care, and occasions. For shoes, use questions about fit, width, and activity.
{% endhint %}
{% stepper %}
{% step %}
#### Filter products by category
1. In your Q\&A generator in the **Select items** step, add an **IF** condition for your category field.
2. Select one category for the enrichment.
{% endstep %}
{% step %}
#### Tailor the prompt
1. Edit the prompt with questions that fit the selected category.
2. Generate and review values for a small set of items.
{% endstep %}
{% endstepper %}
:star: Popularity rank (popularity_rank)
{% hint style="info" %}
#### What this attribute is for
Lets Google know what items are your bestsellers so they know what to recommend if a shopper asks after it.
This is a number between 1 and 100, with 100 being your highest best seller.
{% endhint %}
To fill this attribute, you'll need to import sales data (like views or purchases over the last 30 days). If you use [Insights](/manage-improve-product-data/insights/channable-insights.md), you can [use your Insights dashboard data to create dynamic rules](/manage-improve-product-data/insights/use-insights-in-rules.md) in your Google Shopping feed to assign grades to items.
{% stepper %}
{% step %}
#### Apply the popularity rank rule template
1. In your Google Shopping feed channel, go to the **Rules** step.
2. Click **Create rule**.
3. Select **AI Readiness rule templates > Popularity rank**.
4. Select the conversion data that the rule uses to calculate each rank:
1. For `max_conversions`, click the project field and select **+ Add**. Add a project field and call it `max_conversions` .
2. For `conversions` , map the field containing your conversion data. If you have a Google Ads insights dashboard, click :chart-simple:**\[your Google ads insights dashboard] > conversions**.
6. Click **Save rule**.
{% endstep %}
{% step %}
#### Map popularity rank
1. In your Google Shopping channel, go to the **Mapping** step.
2. Search for **popularity rank**.
3. Map the project field containing the calculated rank to `popularity_rank`.
4. Click **Save**.
{% endstep %}
{% endstepper %}
#### Extra: Calculate popularity rank by category
By default, `popularity_rank` compares conversions across your full inventory. To calculate rank by category instead, select the field containing your product categories.
{% hint style="info" %}
If your category field contains a full path (for example, `Clothing > Shoes > Sneakers`) rather than one value per item, group by a single level within the category hierarchy for the clearest results.
For example, just "Shoes" or just "Sneakers," rather than the full path. You can do this using the [split text rule](https://helpcenter.channable.com/manage-improve-product-data/rules-bulk-edit-and-enrich/how-to-use-actions-then-in-rules/rule-action-split-text).
{% endhint %}
:highlighter: Product highlight (product_highlight)
{% stepper %}
{% step %}
#### Generate product highlight values
1. In the project containing your Google Shopping channel, go to **Generate with AI**.
2. Click **Generate with AI**. In the pop-up, select **Text**.
3. For **Text type** select **Product highlight**, and then click **Continue**.
4. Configure the generator settings:
1. At **Name**, enter a descriptive name, such as `Google product highlights`.
2. At **AI connection**, select the Gemini or OpenAI connection you want to use.
3. At **Project fields to use to generate the text**, select the fields that provide context about your items.
4. At **Field to save the optimized text**, create a project field for the generated values and name it `product_highlight`.
5. At **Output language**, select the language for generated values.
5. Click **Save & continue**.
6. At **Select items**, you can filter your items so you only generate values for specific items. If you leave this blank, we'll generate values for all your items.
1. If you add a condition, click **Save** to update the item count.
7. Click **Save & continue**.
8. At **Prompting**, review and edit the default prompt to generate product highlights that make sense for your items.
9. Click **Generate 5 previews** to preview values for the selected items.
10. Review the previews, then click **Generate all**.
11. At **Generated values**, approve or regenerate the generated values.
{% endstep %}
{% step %}
#### Apply the product highlight rule template
1. In your Google Shopping channel, go to **Rules**.
2. Click **Create rule**.
3. Select **AI Readiness rule templates > Product highlight**.
4. At `product_highlight`, select the project field where you saved the generated values.
5. Create project fields for any remaining placeholders in the template.
6. Click **Save rule**.
{% endstep %}
{% step %}
#### Map product highlights
1. In your Google Shopping channel, go to **Mapping**.
2. Search for **product highlight**.
3. For **product\_highlight**, map the project fields created by the rule template to the matching product highlight attributes.
4. Click **Save**.
{% endstep %}
{% endstepper %}
:magnifying-glass: Product detail (product_detail)
{% stepper %}
{% step %}
#### Generate product detail values
1. In the project containing your Google Shopping channel, go to **Generate with AI**.
2. Click **Generate with AI**. In the pop-up, select **Text**.
3. For **Text type** select **Product and detail**, and then click **Continue**.
4. Configure the generator settings:
1. At **Name**, enter a descriptive name, such as `Google product details`.
2. At **AI connection**, select the Gemini or OpenAI connection you want to use.
3. At **Project fields to use to generate the text**, select the fields that provide context.
4. At **Field to save the optimized text**, create a project field for the generated values and name it `product_detail`.
5. At **Output language**, select the language for generated values.
5. Click **Save & continue**.
6. At **Select items**, add a condition to generate values for matching items only.
7. Click **Save & continue**.
8. At **Prompting**, review and edit the default prompt for generating product details for your items.
9. Click **Generate 5 previews** to preview values for the selected items.
10. Review the previews, then click **Generate all**.
11. At **Generated values**, approve or regenerate the generated values.
{% endstep %}
{% step %}
#### Apply the product detail rule template
1. In your Google Shopping channel, go to **Rules**.
2. Click **Create rule**.
3. Select **AI Readiness rule templates > Product detail**.
4. At `product_detail`, select the project field where you saved the generated values.
5. Create project fields for any remaining placeholders in the template.
6. Click **Save rule**.
{% endstep %}
{% step %}
#### Map product details
1. In your Google Shopping channel, go to **Mapping**.
2. Search for **product detail**.
3. Map the project fields created by the rule template to the matching product detail attributes.
4. Click **Save**.
{% endstep %}
{% endstepper %}
{% hint style="info" icon="lightbulb" %}
#### **Tip**
Start with your highest-volume products or categories. Review the generated values before expanding the filter.
{% endhint %}
#### Attributes without templates
Related product
{% hint style="info" %}
#### What this attribute is for
Tells shoppers about products related to your item.
Google uses related products to inform shoppers about things like accessories, spare parts, and substitutes related to your item. For example, if someone buys an Espresso machine, the AI can naturally recommend filters that go with it.
{% endhint %}
The type of relationship (like `accessory` or `often_bought_with`) connected to the related product's ID number.
If you use an eCommerce platform, check if certain items are already linked to your main item. If so, and this data is imported, you can link the field directly in the **Mapping** step.
1. **Check the source first:** Check your eCommerce platform. If your store already knows which items are cross-sells, you can link this directly in the **Mapping** step.
2. **If missing, use a Rule:** Go to the **Rules** step and create a rule to assign related products.
* Supported values: `part_of_set` , `required_part` ,`often_bought_with` ,`substitute` , `different_brand` , `accessory`
**Example rule:**
1. In the **Mapping** step, search for "related".
2. For `related_product.relationship_type`, map the field containing the relationship type.
3. For `related_product.identifier_type` , map the field containing the id type for the relatied product.
4. For `related_product.identifier` , map the field containing the product ids of your related products.
5. Click **Save**.
Item group title
{% hint style="info" %}
#### What this attribute is for
If you map parent products to its variants using `item_group_id`, you can create a group title that connects your variant items to its main parent items.
For example, if you sell a T-shirt in Red, Blue, and Green, those different colors are the variants. Giving those items with variants a group title helps Google's AI understand the parent product in relation to its child items.
{% endhint %}
A clean, readable text name without the variant details (no sizes or colors).
You can build this name in the **Rules** step by combining information you already have.
**Example rule:**
1. Once you've created a rule, go to the **Mapping** step.
2. For `item_group_title` , map the field containing the product ids of your related products.
3. Click **Save**.
Variant option
{% hint style="info" %}
#### What this attribute is for
When you assign an item group title, the variant option tells Google what makes this specific item unique (like being "Size: L" and "Color: Blue"). Without this context, the AI might accidentally recommend an S when the shopper asked for a size L.
{% endhint %}
Google requires a specific format showing the name of the trait and the value: (e.g. `color:Blue` )
1. In the **Mapping** step, map the field containing the variant type for the attribute `variant_option.name` (e.g., field containing 'size', 'color').
2. For `variant_option.value`, map the project field containing the actual value tied to the variant name (e.g. if 'size' is mapped for the variant name, the mapped field should contain sizes like S, M, L etc.).
### Quick reference
| Attribute | Set up in Channable | Example value |
| --------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------ | ------------------------------------------------------------------- |
| [Question and answer `[question_and_answer]`](https://support.google.com/merchants/answer/17085211) | Generate question-and-answer pairs, then apply the matching rule template. | `"Does it support Bluetooth?":"It has full Bluetooth 6.0 support."` |
| [Product detail `[product_detail]`](https://support.google.com/merchants/answer/17085146) | Generate structured attribute-value pairs, then apply the matching rule template. | `General:Protection:"Scratch-resistant display"` |
| [Product highlight `[product_highlight]`](https://support.google.com/merchants/answer/17085146) | Generate product highlights, then apply the matching rule template. | `Supports 6 GHz Wi-Fi networks` |
| [Popularity rank `[popularity_rank]`](https://support.google.com/merchants/answer/17085297) | Apply the popularity rank rule template using conversion data. | `95.5` |
| [Related product `[related_product]`](https://support.google.com/merchants/answer/17085213) | Create the relationship type, identifier type, and identifier values in your product data. | `accessory:gtin:811571013579` |
| [Document link `[document_link]`](https://support.google.com/merchants/answer/17084656) | Map one or more PDF URLs from a project field. Separate URLs with commas. | `https://example.com/manual.pdf` |
| [Item group title `[item_group_title]`](https://support.google.com/merchants/answer/17085146) | Map this attribute with `[item_group_id]` for products with variants. | `Organic Cotton Men's T-Shirt` |
| [Variant option `[variant_option]`](https://support.google.com/merchants/answer/17085214) | Map each variant property with its name and value. | `Shoe width:narrow,size:8` |
---
# 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://helpcenter.channable.com/manage-improve-product-data/ai-readiness-score/set-up-ai-readiness-attributes.md?ask=&goal=
```
`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.
1. Once you've created a rule, go to the **Mapping** step.
2. For `item_group_title` , map the field containing the product ids of your related products.
3. Click **Save**.