Schema markup for stores: what to implement first, and what changed in 2026.

What schema should an ecommerce site implement first?

In priority order: (1) Product schema with price, availability, brand, SKU, and aggregate rating on every product page, kept current; (2) BreadcrumbList sitewide; (3) FAQ schema on category pages and guides — not product pages; (4) Article or HowTo on guides; plus Organization schema sitewide for entity identity. Everything else is situational.

Why Product schema outranks everything else

Product markup feeds rich results, Shopping Graph matching, and — increasingly — AI shopping agents assembling carts. It's the one schema type where staleness actively hurts: a price or availability mismatch between your markup, your page, and your Merchant Center feed can cost Shopping eligibility and agent trust simultaneously. Three-way consistency (feed = page = schema) is the rule; treat markup updates as part of the price-change workflow, not an afterthought.

Where does FAQ schema belong after the deprecation?

Google retired FAQ rich results in May 2026, so FAQ markup no longer earns the expandable SERP treatment — but the markup itself still does real work. It confirms to engines exactly which questions a page answers, which feeds People Also Ask alignment and AI-answer extraction. The placement rule: FAQ schema goes on category pages, buying guides, and sitewide FAQ pages — where buyers' pre-purchase questions live — and stays off product pages.

How do you know if a schema type is worth implementing?

Schema.org now publishes usage statistics per type (as of June 2026), showing how widely each is deployed — some types appear on tens of millions of domains, others on under a million. Before investing in an exotic type, check whether it's mainstream: widely-adopted types are the ones parsers, search features, and AI agents are built to consume. Marginal types are usually effort without an audience.

The mistakes that cost real money

• Schema contradicting the visible page. Ratings in markup that don't match displayed ratings, prices that lag a sale — these read as deception to validators and erode AI-agent retrieval confidence.
• Markup injected as literal text. A JSON-LD block that renders as visible code instead of executing means a template bug — validate the rendered page, not the template.
• Set-and-forget Product markup. Availability changes daily; markup generated at launch and never updated is wrong within a month.
• FAQ schema stuffed on product pages. Post-deprecation, it earns nothing there and dilutes the page's transactional signal. Move the questions (and markup) up to the category page.
• Skipping Organization schema. Entity identity — name, logo, sameAs links to official profiles — is how engines and LLMs reconcile your brand as one real thing. Without it, you're an ambiguous string.

Schema implementation is engineering, not plugin configuration — which is why it sits inside our technical SEO service, shipped in code and validated on the rendered page. For the on-page layer it supports (titles, copy, internal linking on PDPs and categories), see product & category optimization; for BigCommerce-specific markup work in Stencil themes, BigCommerce SEO is the deep end.