Schema (structured data)
Standardised markup that describes page content to search engines.
Schema, or structured data, is standardised markup — most commonly written in the schema.org vocabulary — that explicitly describes a page's content and entities to search engines and other machine consumers, rather than leaving them to infer that meaning from unstructured text and layout alone.
It is most commonly implemented in JSON-LD syntax, embedded in a page's `<head>` or `<body>` as a script block, though microdata and RDFa remain valid, less-used alternatives. Schema.org itself is a shared vocabulary maintained collaboratively by Google, Microsoft, Yahoo, and Yandex, defining thousands of types and properties from `Organization` to `Recipe` to `LocalBusiness`.
Not all schema is equal in practical value. Google explicitly documents a specific subset of types — Product, Review, FAQPage, HowTo (now restricted), Recipe, Event, JobPosting, LocalBusiness, and a handful of others — that can trigger visible rich results in classical search; markup outside that documented set can still support entity understanding and machine-readability without earning a visible SERP feature.
In the generative-search era, schema's role has expanded beyond rich results: it is a primary mechanism for strengthening entity resolution and factual grounding that AI Overviews and chat-based assistants draw on when deciding whether to trust and cite a specific claim, making comprehensive, accurate schema a strategic asset even on pages that will never earn a classical rich result.
How schema is processed by search and AI systems
When a crawler encounters valid JSON-LD, it parses the declared entities and properties as structured facts rather than inferring them from surrounding text — a `LocalBusiness` schema block with an explicit `priceRange`, `address`, and `openingHours` gives a search system unambiguous, directly-usable data that text alone would require far more confident natural-language understanding to extract reliably.
For rich results specifically, Google runs the parsed schema against its documented eligibility requirements for each type — required and recommended properties, content-matching rules (the markup must reflect content actually visible on the page), and quality guidelines — before deciding whether to render the associated visual feature.
Generative systems appear to use structured data as a grounding and confidence signal: a factual claim backed by explicit, valid schema is a lower-risk citation candidate than the same claim inferred purely from prose, since the schema removes ambiguity about what is actually being asserted.
Why schema matters more in 2026
As AEO and GEO strategies depend on confident entity attribution and factual grounding, schema has shifted from a rich-result-only tactic to core AI-search infrastructure. A business investing in schema purely for visual SERP features while ignoring its role in generative citation is under-using the same markup investment.
How to measure schema performance
Validate every schema-bearing page with Google's Rich Results Test and the Schema.org validator to catch syntax errors and missing required properties, then monitor the 'Enhancements' section of Search Console for the specific rich-result types the site targets, tracking impressions and click-through separately from non-enhanced results.
For generative-citation impact, there's no direct attribution tool; the practical approach is comparing citation rates in AI Overviews and chat-engine sampling before and after a significant schema rollout, alongside classical rich-result performance.
Common schema mistakes
The most common mistake is markup that contradicts visible page content — declaring a `Review` rating or `Product` price in schema that doesn't match what a user actually sees on the page, which violates Google's structured data guidelines and can result in manual action or suppressed rich results.
A second is exhaustive, unfocused tagging — applying dozens of schema types across a site with no clear priority, when a small number of correctly-implemented, feature-earning types (Product, FAQPage, LocalBusiness) deliver far more value than broad but shallow coverage.
A third is letting schema go stale — a `LocalBusiness` block with outdated hours or an `Organization` block missing a since-added sameAs profile — since schema is a factual declaration that needs the same maintenance discipline as any other structured business data.
- Schema data that contradicts visible on-page content.
- Broad, unfocused tagging instead of prioritising feature-earning types.
- Stale schema (outdated hours, prices, missing new sameAs links).
- Invalid JSON-LD syntax causing silent parsing failures.
A step-by-step schema implementation process
Audit existing schema across the site using a full-site crawl (Screaming Frog supports structured data extraction) to inventory what's currently implemented and validate it against Google's documented requirements.
Prioritise a short list of feature-earning types matched to the business: LocalBusiness and Organization sitewide, FAQPage on genuinely substantive Q&A content, Review/AggregateRating where reviews are genuinely collected and displayed, and Product where applicable.
Implement via JSON-LD, ensuring every declared property is accurate and matches visible page content exactly, and validate each template with the Rich Results Test before deploying sitewide.
Establish a maintenance process — schema should update automatically wherever possible (pulled from the same CMS fields driving the visible content) rather than being hand-maintained separately and inevitably drifting out of sync.
Schema for local Austin-area service businesses
LocalBusiness schema with accurate `address`, `geo`, `openingHours`, `priceRange`, and `areaServed` properties is the single highest-priority schema investment for a local service business, directly reinforcing the entity signals that both local-pack ranking and generative citation depend on, ahead of any more exotic schema type.
How schema relates to JSON-LD and entity
JSON-LD is the specific, recommended syntax used to implement schema.org markup; schema is the broader vocabulary and practice of structured data itself. Entity strength is the strategic outcome that comprehensive, accurate, consistent schema is a primary mechanism for building.
Schema priority by business type
| Local service business | LocalBusiness + FAQPage | Highest priority; directly reinforces entity and local signals. |
|---|---|---|
| Ecommerce | Product + Review/AggregateRating | Directly drives rich results and shopping-surface visibility. |
| Content/publisher | Article + Organization | Author/Person schema strongly recommended alongside. |
| Any site, sitewide baseline | Organization schema | Foundational entity signal regardless of vertical. |
Frequently asked questions
- What is schema markup used for?
- It's standardised structured data that explicitly describes a page's content and entities to search engines, most commonly implemented in JSON-LD, enabling both classical rich results and stronger entity/factual grounding for AI-generated answers.
- Does schema markup improve rankings directly?
- Not as a direct ranking factor for most types, but it can earn visible rich results that improve click-through, and it strengthens the entity and factual signals that both classical quality systems and generative citation appear to weight.
- What's the most important schema for a local business?
- LocalBusiness schema with accurate address, geo-coordinates, opening hours, price range, and service area, plus sitewide Organization schema — together these are the foundational entity signal most local sites should prioritise first.
- Can bad schema hurt my site?
- Yes. Schema that contradicts visible on-page content violates Google's structured data guidelines and can result in a manual action or suppressed rich results. Accuracy and content-matching are strict requirements, not suggestions.
- Is JSON-LD required for schema markup?
- No, microdata and RDFa remain valid alternatives, but JSON-LD is Google's recommended syntax because it's separated from visible markup, easier to generate programmatically, and easier to maintain and validate.
A local dental practice added LocalBusiness schema with accurate opening hours, price range, and service list, plus FAQPage schema on its ten most common patient questions. It began earning FAQ-rich results within three weeks and was cited in an AI Overview for a 'how much does a dental cleaning cost' query the following month.
Need this applied to your site?
We turn concepts like these into shipped work every week.