Glossary

Product schema: definition and examples

The Schema.org vocabulary that describes products in machine-readable form — most commonly emitted as Product JSON-LD in a PDP.

Last updated Q1 2026

Product schema is the Schema.org vocabulary used to describe products in machine-readable form, usually embedded as Product JSON-LD in a PDP.

In detail

Product schema is defined at schema.org/Product and covers the properties that search engines, AI engines, and shopping agents use to read a product without parsing free-form copy. The common implementation is JSON-LD — a script block in the PDP head containing a Product object with name, brand, sku, gtin, description, image, offers, and optionally aggregateRating. The same information can be expressed in microdata or RDFa; JSON-LD is the convention because it is cleanest to maintain.

Not every Product field is equally important. For AI visibility, offers (price, priceCurrency, availability), sku, gtin, and aggregateRating matter most. Name, description, and image are table stakes. Fields like material, color, and size become important in faceted category queries — "best ceramic mugs that are dishwasher safe" depends on the engine being able to read material and care data from the schema.

Why it matters

Product schema is how AI engines read a product without guessing. A PDP with thorough schema gives the engine answers to half the sub-queries in a query fan-out before any free-text parsing happens. A PDP with sparse schema forces the engine to infer, which is where hallucinations and misses enter.

For a Shopify brand, product schema is also the lowest-effort highest-yield PDP improvement available. A theme or schema app change that populates three new fields across the whole catalog can move visibility metrics within weeks.

Example

For example: a ceramic-mug brand's 12oz hand-thrown mug PDP emits Product JSON-LD with name, description, image, and an offers block. It is missing gtin, material, aggregateRating, and a populated category. Adding the four fields — via a metafield-backed schema app — lifts ChatGPT share of model for "best ceramic mug for gift" prompts from 6% to 18% across the next two weekly tracking runs, and moves the SKU into Google AI Overviews answers for three adjacent queries.

Related terms

PDP optimization — the broader practice schema fits inside.
SKU-level AEO — the discipline schema most directly supports.
AI discoverability — schema's upstream role in being found.
Hallucination detection — schema reduces hallucinations at the source.
llms.txt — the unstructured counterpart to schema for LLM crawlers.

Ask AI about product schema

Have your favorite AI engine summarize this for your specific use case.

Frequently asked questions

Which Product schema fields actually matter for AI search?

name, brand, sku, gtin (where applicable), description, image, offers (with price, priceCurrency, availability), and aggregateRating when real review data exists. Category and material matter for retrieval. Review is the single field that most often separates cited SKUs from uncited SKUs, because AI engines lean heavily on third-party review evidence when deciding which products to surface.

Do most Shopify themes emit Product schema correctly?

Partially. Most themes emit a minimal Product block but skip fields that matter for AI such as gtin, material, and structured aggregateRating. Some emit offers without priceCurrency. eCommerce Insights audits the JSON-LD the site actually serves and recommends specific additions or corrections, and supports writing through either theme customization or metafield-driven schema apps.

Is there a penalty for using structured data apps like Schema App or Yoast?

No. AI engines and search engines read whatever valid JSON-LD is present on the page. Whether it comes from a theme, a schema app, or hard-coded markup is invisible to the consumer. eCommerce Insights audits the emitted schema against what the Shopify admin says is true, so the tool source does not affect the analysis.

Related guides

Audit your Product JSON-LD and see what is missing. Try the eCommerce Insights product schema generator or read the Schema.org Product vocabulary.