In 2015, structured data was a nice-to-have. You could add some JSON-LD to your pages and maybe get a star rating or a recipe card in search results. In 2026, structured data is how search engines understand what your page is about, who published it, when it was updated, and whether it answers the question the user asked. Google's AI Overviews, Bing's Chat answers, and every other generative search feature rely on structured data to identify authoritative sources.
If your pages do not have structured data, they are not broken. They just look like every other plain blue link in search results while your competitors show rich snippets with images, ratings, prices, and FAQ dropdowns.
What structured data actually does
Schema.org defines a vocabulary of types and properties that describe real-world entities: products, businesses, people, events, articles, recipes, software, courses. When you add JSON-LD markup to a page using this vocabulary, you are giving search engines a machine-readable description of the page's content.
Google uses this markup to generate rich results. A Product page with proper schema can show price, availability, and review stars directly in search results. An FAQ page with FAQPage schema gets expandable question-and-answer dropdowns. A LocalBusiness with complete schema appears in map packs with hours, phone number, and rating.
The key word is "proper." Google does not just check whether schema exists. It validates specific required fields per type. A Product without name and offers will not generate a rich result. A LocalBusiness without address will not appear in local results. A Recipe without image and recipeIngredient will not show the recipe card. Each type has its own set of required and recommended fields, and missing any required field means no rich result.
The most common problems
The Schema Validator checks your JSON-LD against Google's rich result required fields. After running thousands of pages through validation, the same issues come up repeatedly.
Missing @context. Every JSON-LD block needs "@context": "https://schema.org" at the top level. Without it, the markup is just a JSON blob with no semantic meaning. Some CMS plugins generate JSON-LD without the context declaration, which means the entire block is invisible to search engines.
Incorrect nesting. A Review inside a Product is valid. A standalone Review not nested inside a reviewable type will not generate rich results. An AggregateRating attached to a WebPage is ignored by Google. The parent-child relationship between schema types matters, and getting it wrong is the same as not having schema at all.
Mismatched data. Google's guidelines state that structured data must reflect the visible content on the page. If your JSON-LD says the product costs $29.99 but the page shows $34.99, that is a structured data spam violation. If your Article schema has a dateModified from 2023 but the page content references events from 2025, Google may flag the page as misleading.
Orphaned schema blocks. Pages sometimes accumulate multiple JSON-LD blocks from different plugins, themes, or manual additions. Two blocks declaring different Organization types for the same page confuse entity resolution. The JSON-LD Graph Linter catches these graph-level problems that a simple field-by-field validator misses.
Beyond validation: graph structure
Modern schema best practice uses @graph to connect multiple entities in a single JSON-LD block. A well-structured graph might connect an Article to its author (Person), the author to the publishing Organization, the Organization to its WebSite, and the WebSite to its SearchAction. Each entity references the others via @id, creating a knowledge graph that search engines can traverse.
This is where structured data moves from "add some fields to get a rich result" to "tell search engines exactly how your content ecosystem fits together." A complete graph helps Google's Knowledge Graph integration, AI Overview sourcing, and entity disambiguation. When two sites publish articles about the same topic, the one with a complete entity graph is more likely to be cited in AI-generated answers because the machine can verify the source's identity and authority.
The Rich Results Eligibility Audit checks per-type required and recommended fields across 16 schema types, giving you a checklist of what is missing on each page. Combined with the Schema Validator for syntax correctness and the Graph Linter for structural integrity, you get a complete picture of your structured data health.
Starting from zero
If your site has no structured data, start with the type that matches your primary page template. For a business site, that is LocalBusiness or Organization on the home page and Service or Product on service pages. For a blog, it is Article on every post with Person for the author and WebSite for the site entity.
The Schema Fix Bundle generates complete JSON-LD for 13 page types with all required fields pre-populated. Pick your page type, fill in the values, and paste the block into your template. It is not glamorous work, but it is the kind of foundational improvement that compounds over months. I wrote about this build-once, benefit-forever approach in The $97 Launch as one of the highest-ROI activities for a new web property.
Related tools
- Schema Validator validates JSON-LD syntax and Google required fields
- JSON-LD Graph Linter checks graph structure, dangling references, and orphan nodes
- Rich Results Eligibility Audit audits per-type field completeness across 16 types
- Schema Fix Bundle generates complete JSON-LD templates for 13 page types
- FAQ Schema Parity diffs FAQPage JSON-LD against visible Q&A content
Fact-check notes and sources
- Schema.org vocabulary maintained by Google, Microsoft, Yahoo, and Yandex since 2011. Source: schema.org/docs/about.html.
- Google rich result required fields per type: Google Search Central, structured data documentation for each type (Product, LocalBusiness, Article, etc.). Updated quarterly.
- JSON-LD recommended by Google as the preferred structured data format: Google Search Central, "Introduction to structured data markup in Google Search," 2024.
- AI Overviews using structured data for source identification: Google I/O 2024, "How AI Overviews work" presentation; Search Central documentation on "How Google Search uses structured data."
- Structured data spam policies: Google Search Central, "Structured data general guidelines," section on "Content guidelines" requiring structured data to match visible content.
This post is informational, not SEO-consulting advice. Google's structured data requirements evolve with each search update.
