Schema.org Product in JSON-LD: E-commerce Guide 2026

You’ve probably seen those Google results where a product page shows stars, price, and “In stock” directly under the title. That’s a rich result, and it’s what a properly implemented schema.org Product triggers. Sites that get them often see CTR increase by 15 to 40% at the same ranking position compared with results without rich results.

In 2026, implementation through JSON-LD is the official method recommended by Google. Here is the complete, testable, production-ready version.

Why JSON-LD instead of Microdata or RDFa

Three formats can add schema.org to a web page: Microdata (directly in the HTML), RDFa (HTML5 attributes), and JSON-LD (a separate structured data block).

Google has recommended JSON-LD since 2015, and by 2026 it has become the de facto standard. Reasons:

Data/presentation separation: JSON-LD lives in a <script type="application/ld+json"> inside the <head>, independently from the visible HTML
Easier to generate: one JSON block per page, easy to maintain
No conflict with your theme: no need to modify the HTML structure of templates
Better compatibility with modern frameworks (Next.js, Nuxt, etc.)

Microdata is still readable by Google but is no longer recommended for new implementations.

The 7 required (or recommended) fields for a Product

Here is a minimal Product JSON-LD that passes Google tests and enables rich results:

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Men's leather Derby shoe",
  "description": "Derby shoe in Italian leather, hand-finished, available in sizes 38 to 46.",
  "image": [
    "https://example.com/products/derby-01.jpg",
    "https://example.com/products/derby-02.jpg",
    "https://example.com/products/derby-03.jpg"
  ],
  "sku": "DERBY-BLK-44",
  "brand": {
    "@type": "Brand",
    "name": "Atelier Maison"
  },
  "offers": {
    "@type": "Offer",
    "url": "https://example.com/products/derby-cuir-homme",
    "priceCurrency": "EUR",
    "price": "149.00",
    "availability": "https://schema.org/InStock",
    "itemCondition": "https://schema.org/NewCondition"
  },
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.7",
    "reviewCount": "38"
  }
}

Breakdown:

@context and @type — declare that this is a schema.org Product
name — the product name (≤100 chars recommended)
description — short description (150-300 chars)
image — array of image URLs (minimum 1, ideally 3-6 in high resolution)
sku — unique product identifier (required for rich results)
brand — product brand
offers — pricing and availability structure (critical)
aggregateRating — average rating if you have reviews (triggers stars in the SERP)

Offer and AggregateRating: critical details

Offer

This is the most sensitive part. Four rules to follow:

priceCurrency in ISO 4217: "EUR", "USD", "GBP"… not "€", "$", or "£". This is a common error that causes rich results to fail.

price as a string, not a number: "149.00" not 149.00. Google accepts both in 2026, but the guidelines recommend a string.

availability as a full schema.org URL:

https://schema.org/InStock → in stock
https://schema.org/OutOfStock → out of stock
https://schema.org/PreOrder → preorder
https://schema.org/BackOrder → special order with lead time

priceValidUntil (optional but recommended): price expiration date in ISO 8601 format ("2026-12-31"). Google uses it to detect outdated prices.

AggregateRating

This triggers stars in the SERP. Requirements:

ratingValue between 1 and 5
reviewCount: minimum 1, but Google recommends 3+ for display
The matching reviews must be visible on the page — faking AggregateRating with nonexistent reviews can lead to a manual penalty

Tip: if you have few reviews, display them on your page through a widget (Yotpo, Judge.me, Trustpilot) and pull the count + average rating from their API for the JSON-LD.

Test with Rich Results Test

Before pushing to production, test every product page template with Google’s Rich Results Test.

It checks:

Valid JSON syntax
All required fields present
Values in expected formats (URLs, dates, currencies)
Eligibility for rich results

Common errors reported:

"availability" expected schema.org URL → you used "In stock" instead of the schema.org URL
"priceCurrency" invalid → currency is not ISO 4217
Missing field "sku" → no SKU in the Offer
Multiple products in single offer → you tried to put multiple products in one JSON-LD block

Once the test passes, Google Search Console reports performance and errors over 2-4 weeks through Enhancements → Products.

Implementation by platform

Shopify

Since 2023, Shopify automatically generates Product schema on product pages if you use an official theme. The problem: the default schema is often incomplete (no AggregateRating, no brand).

For a complete schema:

Edit your theme: Online Store → Themes → Edit code
Open snippets/product-structured-data.liquid or sections/main-product.liquid
Add the missing fields or replace it with a custom template

Alternative: install an app like SEO JSON-LD Schema or Smart SEO that manages schema across the full catalog on its own.

WooCommerce

WooCommerce also generates default schema starting from v5+, but it is incomplete. Recommended plugins to complete it:

RankMath (free): handles Product schema with Offer and AggregateRating
Yoast WooCommerce SEO (paid): enriches Yoast’s default schema
WP SEO Structured Data Schema: more granularity if you want to customize by template

Next.js / headless

If you use a custom frontend, inject the JSON-LD into the <head> through an SSR component:

<script
  type="application/ld+json"
  dangerouslySetInnerHTML={{
    __html: JSON.stringify(productSchema)
  }}
/>

Warning: generating JSON-LD client-side (through useEffect) does not work for SEO — Google crawls server-rendered output. Make sure the JSON-LD is present in the initial HTML returned by the server.

Typical mistakes to avoid

Contradictions between the visible page and the JSON-LD. If the page shows “In stock” but the JSON-LD says OutOfStock, Google ignores the JSON-LD. Sync both to the same data source.

Different price in JSON-LD and on the page. Same logic: Google detects the mismatch and ignores the rich result.

Multiple Product objects in the same <script>. Only one @type: Product per script. If you have variants, use @type: ProductGroup with hasVariant: [...] to link them.

Images not accessible. Image URLs must be public, not behind a login. Test each URL with curl to confirm it returns a 200.

Made-up AggregateRating. Showing 4.9 stars on a page with no visible reviews can lead to a manual penalty, and the risk is real.

What about variants?

If your product has multiple sizes or colors, there are 2 approaches:

Approach 1: one Product schema only (the parent)

Simpler, and suitable for 90% of cases. The parent schema describes the generic product, and the page handles variants through the selector. The displayed price is the default one, and availability is InStock if at least one variant is available.

Approach 2: ProductGroup with hasVariant

More precise but more complex. Recommended if your variants have very different prices or different availability states (out of stock in size 44 but not in 42). Google has supported ProductGroup since 2022.

Schema.org Product in JSON-LD: E-commerce Guide 2026