Shopify metafields for AI shopping: Catalog Mapping

Plenty of Shopify stores keep their best product data out of the main description. Ingredients sit in a metafield that feeds a tab. The real product name is in a metafield because the title carries a SKU. Sizes and colors are encoded in tags. That works fine on your own storefront, where your theme knows where to look. Shopify Catalog, which supplies product data to AI shopping channels, doesn’t know unless you tell it.

Catalog Mapping is how you tell it.

What Catalog Mapping does

Shopify’s description: “You can map your Shopify Catalog product data to your preferred sources and influence how your products are represented on AI-driven sales channels” (Mapping your product data sources).

It covers three things:

• Product fields. Which source the catalog uses for a product’s title, description and category. Sources can be “product attributes, product metafields, or metaobject references.”
• Variant grouping. How separate products get grouped into one listing with options.
• Option names. What your variant options are called when shown to shoppers.

It doesn’t change your online store. Shopify pitches it as a way to change how products appear on AI platforms “without affecting the unique product structure or data that you have in your store.”

Who needs it

Shopify says mapping is “most helpful if your store uses custom data and grouping logic for products, such as metafields, metaobjects, tag prefixes, or separators/delimiters in product titles.” In practice, that’s stores where:

• The title is written for internal use, like “SKU-2231 / Blue / M,” and the customer-facing name lives in a metafield.
• The description is short and the detail lives in metafields that your theme shows as tabs.
• Each color or size is a separate product, and your theme stitches them together with tags or a metafield.
• Your product category is set in a custom field instead of Shopify’s standard product category.

If your title, description and variants are already complete in the standard fields, Shopify’s own note applies: “you might not need to make any changes.”

Setting it up

All of these steps are from Shopify’s mapping guide.

Map product fields

1. Go to Sales channels > Agentic.
2. In Sources, click Review how your product data is sent to the Shopify Catalog for Agentic Channels.
3. Under Product fields, choose a source for Product title, Product description and Product category.
4. Use Change in the Product summary section to preview a few different products.
5. Click Save.

Group variants

By default the catalog groups products using your Shopify Combined Listings setup. To use something else, turn on Custom variant grouping and choose a grouping method:

• Product title, with the delimiter your titles use, like a slash or a dash.
• Product metafield, choosing the metafield that holds the group.
• Product tag, entering the prefix your grouping tags start with.

Shopify notes that if you group by an unstructured metafield, the preview can’t show you the other products in the group.

Name your options

With custom grouping on, Product options lets you choose which options to show and set their display names. This is worth doing if your option names are codes. Shopify’s listing-quality check “checks whether your variant or option names contain acronyms or numbers that are hard for agents to understand or recommend” (Managing agentic storefronts).

What to expect afterward

Two caveats, both from Shopify. Mapping settings “are inputs that can influence Shopify Catalog. Changes are processed with a delay.” And mapping “doesn’t guarantee product placement, ranking, or exact wording in AI search results” (Shopify Catalog).

The second one matters. Mapping decides what the catalog reads. The catalog’s description, top features and selling points are still fields Shopify marks as “generated or enriched by Shopify” (Global Catalog docs). Better input makes a better summary more likely. It doesn’t let you write the summary. See What Shopify’s AI adds to your listings for what those fields are.

Category metafields and the standard taxonomy

Separate from mapping, Shopify has its own structured fields: the standard product category and the category metafields that go with it. Shopify says category metafields “help you add the right information to your products to make them more discoverable by visitors on your site, on marketplaces, and on search engines,” and that some are “added by default” based on Shopify Magic predictions (Category metafields).

Shopify’s agentic storefronts announcement points the same way: “You define your schema and group products by standard attributes and metafields so that agents accurately present your products in searches” (Shopify news, Dec 10, 2025).

Two things to check here:

• Set a product category on every product. It’s the anchor for everything else.
• Review the category metafields Shopify filled in for you. Predicted values are a starting point. If Shopify guessed a material or an age group, confirm it’s right, because those are exactly the attributes a shopper filters on.

A short checklist

1. Open the mapping preview for five of your products and read what the catalog receives for each.
2. If a title, description or category looks wrong or thin, map it to the field that has the real data.
3. If your variants show up as separate products, set up custom grouping.
4. Rename any option values that are codes.
5. Set product categories and review predicted category metafields.
6. Wait for the change to process, then check what the catalog says.

Step 6 is where Hearsay comes in. It compares what Shopify’s catalog tells AI agents about each product with your description, tags, product details fields and product page, and flags claims that contradict you or that you never made (how we check). After you edit a product it rechecks, holding alert emails for up to 72 hours so you aren’t alerted while the catalog catches up. Try the free scan on 5 products first.