Which schema types does every e-commerce product page need?

Four: Product schema as the base entity, Offer schema nested inside it for price, currency, and availability, AggregateRating for review data, and BreadcrumbList so engines understand your category hierarchy.

Why is missing Offer schema the most common cause of price hallucinations?

The offers property is where AI models read price, currency, and availability. Without it, or with it filled in wrongly, the model has to infer the price from unstructured copy, which is easily misread and is the single most common source of price hallucinations.

What is the correct availability value in Product schema?

Use the full schema.org URL, for example https://schema.org/InStock, rather than a bare value like InStock. A wrong or shorthand availability value causes models to cite incorrect stock status.

Schema Markup for E-Commerce Products: The Definitive Guide (2026)

Schema markup is the structured data layer that tells AI answer engines, Google, and shopping assistants what your products actually are - not what your copy implies they are. If AI engines are going to make claims about your products, schema is how you control what they say.

Why schema matters more now than it did for classic SEO

Google's classic crawler could infer product details from page copy. AI answer engines need structured facts to cite. A page without a Product schema is a page the model has to guess about, and guessing is where hallucinations come from.

The four schema types every e-commerce product page needs

1. Product schema

The base entity. Include at minimum:

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "CloudWalk Lite Running Shoes",
  "description": "Lightweight trail running shoe for Indian terrain. Available in sizes 6–11.",
  "brand": { "@type": "Brand", "name": "YourBrand" },
  "sku": "CWL-001-BLK",
  "image": "https://yourdomain.com/images/cloudwalk-lite.jpg"
}

2. Offer schema (inside Product)

The offers property is where AI models get price, currency, and availability. Missing or wrong offer schema is the single most common cause of price hallucinations.

"offers": {
  "@type": "Offer",
  "price": "1899",
  "priceCurrency": "INR",
  "availability": "https://schema.org/InStock",
  "url": "https://yourdomain.com/products/cloudwalk-lite",
  "priceValidUntil": "2026-12-31"
}

3. AggregateRating schema (inside Product)

Rating data makes your product more citable in comparison queries. Models often surface rated products ahead of unrated ones when answering “best X for Y” queries.

"aggregateRating": {
  "@type": "AggregateRating",
  "ratingValue": "4.7",
  "reviewCount": "382"
}

4. BreadcrumbList schema

Breadcrumbs help AI models understand your product hierarchy and category. This is how a model knows your shoe is in “Running > Trail Running” and not “Casual Footwear.”

Organisation-level schema: the entity anchor

Every site needs an Organization schema in the root layout, not just on product pages. This anchors your brand identity:

{
  "@context": "https://schema.org",
  "@type": "Organization",
  "@id": "https://yourdomain.com/#organization",
  "name": "YourBrand",
  "url": "https://yourdomain.com",
  "logo": "https://yourdomain.com/logo.svg",
  "sameAs": [
    "https://twitter.com/yourbrand",
    "https://linkedin.com/company/yourbrand",
    "https://www.g2.com/products/yourbrand"
  ]
}

The sameAs array is how AI models resolve brand disambiguation. Without it, the model may confuse your brand with a similarly named company.

Common mistakes that cause hallucinations

Price in copy but not in schema: The model finds the price in copy, but copy is unstructured and easily misread.
Outdated priceValidUntil: An expired priceValidUntil date may cause the model to distrust your price and infer its own.
Wrong availability value: Use the full schema.org URL: https://schema.org/InStock, not just InStock.
Schema only on the homepage: Product schema belongs on product pages, not just the homepage.
JavaScript-rendered schema: If your schema is injected by client-side JavaScript, crawlers and AI models that do not execute JS will not see it.

Validation: the final check

Paste any product URL into Google's Rich Results Test. Every error in the output is a potential source of hallucinated data. Warnings are acceptable. Errors are not.

For teams shipping fast, running the Rich Results API against all product pages on every pull request keeps schema regressions from reaching production.

Audit your schema and fix it in one click

VritantAI Discover crawls your product pages, scores your structured data, and generates JSON-LD patch diffs ready for one-click deployment to Shopify or WooCommerce.

Run a GEO audit with Discover →