SEO Guide · 2026

Shopify collections JSON-LD and AI agent indexing (2026)

Shopify collection pages are the most under-structured page type in e-commerce. They receive a BreadcrumbList by default — and nothing else. No ItemList, no price range, no CollectionPage typing. This makes them largely invisible to AI shopping agents for the high-intent category queries where they should dominate.

Why Shopify collection pages have almost no structured data

Shopify's theme architecture was not designed with collection-page structured data in mind. The product template has always been the structured data focus — products have SKUs, prices, and inventory, which map directly to well-understood schema.org types. Collection pages are treated architecturally as filter and navigation surfaces rather than as indexable, structured content entities.

The result: when Shopify renders a collection page, the structured data output is minimal. Inspect the source of any default Shopify collection page and you'll typically find a single application/ld+json block:

// Typical default Shopify collection page JSON-LD output:
{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [
    { "@type": "ListItem", "position": 1, "name": "Home", "item": "https://yourstore.com/" },
    { "@type": "ListItem", "position": 2, "name": "Women's Trail Shoes",
      "item": "https://yourstore.com/collections/womens-trail-shoes" }
  ]
}
// That is the entire structured data output. No ItemList, no CollectionPage, no products.

This is the complete structured data signal available to an AI agent crawling a Shopify collection page. The agent sees a breadcrumb trail — it can infer the page is a subcategory — but it cannot:

• Enumerate the products in the collection in a structured way
• Understand the price range of products in the collection
• Know whether the collection is manually curated or algorithmically sorted
• Connect the collection to a recognized product category in a knowledge graph

Default Shopify collection structured data vs. recommended

How AI agents index collection pages vs. product pages

Query types where collection pages appear in AI answers

CollectionPage vs. SearchResultsPage schema

There is an important schema.org distinction between CollectionPage and SearchResultsPage that matters for filtered collection pages in Shopify.

Use CollectionPage when:

• The page represents a manually curated or static product category
• Products are consistent across visits for a given sort order
• The collection has a meaningful title and editorial purpose (e.g. "Women's Trail Running Shoes", "New Arrivals")

Use SearchResultsPage (or a plain WebPage) when:

• The page is generated by an on-site search query
• The page is a facet-filtered result (e.g. /collections/all?filter.p.tag=waterproof)
• The content changes dynamically based on user-selected facets

Complete Liquid snippet for collection page JSON-LD

Add this snippet to your collection.liquid template or the equivalent section file in your theme. It outputs all three structured data types: CollectionPage, ItemList, and a full-path BreadcrumbList. The snippet handles both flat and nested collection hierarchies via a parent collection metafield.

{% comment %}
  Collection page JSON-LD: CollectionPage + ItemList + BreadcrumbList
  Add to: sections/collection-template.liquid or collection.liquid
  Requires metafield: seo.parent_collection_handle (optional, for nested breadcrumbs)
{% endcomment %}

{% assign collection_url = shop.url | append: collection.url %}
{% assign visible_products = collection.products | limit: 48 %}

{% comment %} Compute price range across visible products {% endcomment %}
{% assign prices_min = visible_products | map: 'price_min' | sort %}
{% assign prices_max = visible_products | map: 'price_max' | sort %}
{% assign low_price = prices_min | first | divided_by: 100.0 %}
{% assign high_price = prices_max | last | divided_by: 100.0 %}

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "CollectionPage",
  "name": {{ collection.title | json }},
  "description": {{ collection.description | strip_html | truncate: 300 | default: collection.title | json }},
  "url": {{ collection_url | json }},
  "about": {
    "@type": "Thing",
    "name": {{ collection.title | json }}
  },
  "numberOfItems": {{ collection.products_count }},
  "offers": {
    "@type": "AggregateOffer",
    "lowPrice": "{{ low_price }}",
    "highPrice": "{{ high_price }}",
    "priceCurrency": {{ shop.currency | json }},
    "offerCount": {{ collection.products_count }}
  },
  "mainEntity": {
    "@type": "ItemList",
    "name": {{ collection.title | json }},
    "numberOfItems": {{ visible_products | size }},
    "itemListElement": [
      {% for product in visible_products %}
      {
        "@type": "ListItem",
        "position": {{ forloop.index }},
        "item": {
          "@type": "Product",
          "name": {{ product.title | json }},
          "url": {{ shop.url | append: product.url | json }},
          "image": "https:{{ product.featured_image | img_url: '800x800' }}",
          "description": {{ product.description | strip_html | truncate: 200 | json }},
          "brand": { "@type": "Brand", "name": {{ product.vendor | json }} },
          {% if product.variants.first.barcode != blank %}
          "gtin": {{ product.variants.first.barcode | json }},
          {% endif %}
          "offers": {
            "@type": "AggregateOffer",
            "lowPrice": "{{ product.price_min | divided_by: 100.0 }}",
            "highPrice": "{{ product.price_max | divided_by: 100.0 }}",
            "priceCurrency": {{ shop.currency | json }},
            "offerCount": {{ product.variants.size }},
            "availability": {% if product.available %}"https://schema.org/InStock"{% else %}"https://schema.org/OutOfStock"{% endif %}
          }
        }
      }{% unless forloop.last %},{% endunless %}
      {% endfor %}
    ]
  }
}
</script>

{% comment %} BreadcrumbList — supports flat and nested collection hierarchies {% endcomment %}
{% assign parent_handle = collection.metafields.seo.parent_collection_handle %}
<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "BreadcrumbList",
  "itemListElement": [
    {
      "@type": "ListItem",
      "position": 1,
      "name": {{ shop.name | json }},
      "item": {{ shop.url | json }}
    }
    {% if parent_handle != blank %}
    {% assign parent_collection = collections[parent_handle] %}
    ,{
      "@type": "ListItem",
      "position": 2,
      "name": {{ parent_collection.title | json }},
      "item": {{ shop.url | append: parent_collection.url | json }}
    },
    {
      "@type": "ListItem",
      "position": 3,
      "name": {{ collection.title | json }},
      "item": {{ collection_url | json }}
    }
    {% else %}
    ,{
      "@type": "ListItem",
      "position": 2,
      "name": {{ collection.title | json }},
      "item": {{ collection_url | json }}
    }
    {% endif %}
  ]
}
</script>

Setting the parent collection metafield

For nested collection breadcrumbs to render the full path, store the parent collection handle on each child collection using a metafield. Set this via the Shopify admin metafield editor, or via the Admin REST API for bulk configuration:

// Shopify Admin API: set parent collection handle on a child collection
PUT /admin/api/2024-10/collections/{child_collection_id}/metafields.json
{
  "metafield": {
    "namespace": "seo",
    "key": "parent_collection_handle",
    "value": "womens-running",
    "type": "single_line_text_field"
  }
}

// Resulting breadcrumb JSON-LD for child collection:
// CatalogScan Store > Women's Running > Trail Shoes
// Positions:    1          2                3

How Google AI Mode uses collection page signals for category queries

Google AI Mode's shopping answer format for category queries — "best running shoes under $100", "most durable trail shoes for women" — draws on Shopping Graph data, product page structured data, and collection page signals simultaneously. Collection pages with ItemList contribute a meaningful editorial rank signal to this process.

When Google AI Mode processes a category query, it follows roughly this path:

1. Queries the Shopping Graph for products matching the intent and price filter
2. For each matched product, checks whether it belongs to a CollectionPage with an ItemList that includes the product at a specific position
3. Products at position 1–5 in a relevant CollectionPage ItemList receive an editorial rank boost — the store is signaling these are their top products in this category
4. The collection's about.name and description verify whether the collection matches the query's category intent
5. Price range from AggregateOffer enables price-filtered category answers ("under $100")

This means your product ordering within Shopify collections is a structured data signal that AI agents read directly. Products ranked first are more likely to appear in AI Mode category answers. Put your best-converting, best-reviewed, and most relevant products at the top of your collections — because that position is now machine-readable ranking data.

The 3 most impactful structured data signals for collection pages

Signal 1: ItemList with product positions

The single highest-impact addition for AI agent visibility. Without ItemList, collection pages cannot contribute to AI Mode category answers. With ItemList, the collection becomes a ranking signal source for the products it contains. Include products with their URLs, names, images, and nested AggregateOffer objects with price and availability for maximum signal density. The position field matters — AI agents interpret lower position numbers as higher editorial priority.

Signal 2: BreadcrumbList with full path

For stores with nested collections (Women's › Running › Trail › Waterproof), a full-path BreadcrumbList tells AI agents about your category hierarchy. This is critical for understanding which collection is the parent category and which is a subcategory — information that affects how AI agents route category queries through your catalog structure. Flat 2-level breadcrumbs (Home › Collection) provide minimal category context for deep taxonomies.

Signal 3: CollectionPage with about pointing to a Thing

The about property on CollectionPage connects your collection to a semantic concept. When you set "about": { "@type": "Thing", "name": "Women's Trail Running Shoes" }, you're declaring that this collection is authoritative for that concept. For maximum knowledge graph signal, add a sameAs reference to the equivalent Wikipedia category or Wikidata item where one exists — this grounds your collection in the AI agent's existing world knowledge about product categories.

Handling pagination in collection page structured data

Shopify collection pages paginate when the product count exceeds the theme's per-page setting (typically 24, 36, or 48 products). Each paginated page (e.g. /collections/womens-trail-shoes?page=2) has its own HTML and its own opportunity — or problem — with structured data.

The recommended approach for paginated collection pages:

• Add CollectionPage + ItemList JSON-LD only on page 1 (the canonical URL without a ?page= parameter). Subsequent pages add no ItemList — their products are implicitly part of the same collection, referenced by the page-1 schema.
• Ensure paginated pages 2+ have rel="canonical" pointing to the base collection URL, not to themselves — this prevents Google from treating paginated collection pages as separate entities with duplicate or partial ItemList data.
• For very large collections (>200 products), consider using Google's recommended "paginated" link signals (rel="prev" / rel="next") in addition to canonical — though these are no longer required by Google, they remain useful for AI agents that parse link headers.

The BreadcrumbList schema should be identical on all paginated pages (it reflects the collection's position in the hierarchy, not the page's position in pagination).

Testing your collection page structured data

After implementing the Liquid snippet, verify your output using Google's Rich Results Test and the Schema Markup Validator. For ItemList schema specifically, the most common validation errors to check for are: