PerfoMetrix

All articles · Technical SEO

Structured Data in 2026: What Still Earns Rich Results

By PerfoMetrix Team6 min read

Cover with the article title next to curly braces framing a star rating, in blue on a dark background

Structured data had a gold-rush phase. Add FAQ markup to every page and your result doubled in height; add HowTo steps and you got images in the results. Then Google narrowed or retired several of those features, and a lot of sites still carry markup that no longer does anything visible.

That doesn’t make structured data pointless. It still powers several rich result types, it helps search engines understand what a page is about, and it’s one of the cleaner ways to describe your products, articles and organization to any system that reads the page. The trick is knowing which types are worth maintaining today and what each one actually needs.

What structured data does and doesn’t do

Structured data is machine-readable information about a page, usually written with the schema.org vocabulary. It makes a page eligible for certain search features. It doesn’t guarantee them, and Google has said for years that it isn’t a general ranking factor.

Whether a rich result appears depends on the markup being valid and complete, on the page meeting the feature’s content guidelines, and on Google’s own decision for that query and device. Valid markup that shows nothing in results is common and not necessarily a sign of a problem.

JSON-LD is the format to use

Google supports JSON-LD, Microdata and RDFa, and recommends JSON-LD. It sits in a single script tag, doesn’t tangle with your HTML templates, and is easy to generate from the same data that renders the page:

<script type="application/ld+json">
  {
    "@context": "https://schema.org",
    "@type": "Article",
    "headline": "Structured Data in 2026: What Still Earns Rich Results",
    "datePublished": "2026-10-29",
    "dateModified": "2026-10-29",
    "author": { "@type": "Person", "name": "Jane Doe", "url": "https://example.com/team/jane" },
    "image": ["https://example.com/covers/structured-data.jpg"]
  }
</script>

Generate it server-side from your data model rather than writing it by hand per page. Hand-written markup drifts: the price changes on the page and stays wrong in the JSON.

What was retired or narrowed

If you’re auditing an older site, these are the usual suspects:

  • FAQ rich results were limited in 2023 to well-known, authoritative government and health sites. For everyone else, FAQPage markup no longer produces the expandable questions.
  • HowTo rich results were removed in 2023.
  • The sitelinks search box was retired in late 2024, so WebSite markup with a SearchAction no longer adds a search field to your result.
  • Google has continued to phase out less-used features since then. Its Search Gallery is the authoritative list of what’s currently supported; check it before investing in a type.

Retired markup doesn’t hurt you, and FAQPage still describes a real FAQ page accurately, so there’s no urgency to delete it. Just don’t expect it to change how your result looks, and don’t spend template work on it for new pages.

Types that still earn visible results

These are the types most sites can benefit from. Each has its own required and recommended properties in Google’s documentation; the notes below cover the parts people most often miss.

Product

The most valuable type for e-commerce. Product markup can feed both product snippets (price, availability and ratings under a regular result) and merchant listings in shopping surfaces.

  • A product snippet needs a name plus at least one of offers, review or aggregateRating.
  • offers should carry price, priceCurrency and availability, and they must match what the page shows.
  • Variants of one product can be described with ProductGroup and hasVariant, instead of repeating near-identical Product blocks.

Review snippets

Star ratings remain one of the most visible enhancements, with conditions. Reviews must be about a specific item, and ratings an organization collects about itself on its own site (for its LocalBusiness or Organization) aren’t eligible for stars. The rating also has to be visible on the page, not only in the markup.

BreadcrumbList describes the page’s position in your site hierarchy and helps Google understand your structure. It’s cheap to add from your navigation data and rarely goes wrong, as long as each item’s URL is a real, indexable page.

Article

Article markup has no required properties in Google’s documentation, but headline, image, datePublished, dateModified and author (with a name and ideally a profile URL) help Google pick the right image, dates and byline for news and article surfaces.

Organization

Organization markup on your home page lets you state your official name, logo, url and sameAs links to your profiles elsewhere. It’s a clean way to tell search engines which logo and profiles belong to you.

Domain-specific types

Recipe, Event, VideoObject, JobPosting, LocalBusiness, SoftwareApplication and others still produce rich results or feed dedicated experiences. If your pages genuinely are one of these, the markup is usually worth it, because the features are designed around exactly that data.

Rules that get sites into trouble

Google’s structured data guidelines come down to a few principles, and breaking them can lead to a manual action that removes rich results for the whole site:

  • Mark up what’s visible. Every value in the markup should be on the page for users: the price, the rating, the author, the questions and answers. Hidden or invented values are the classic violation.
  • Mark up the main content. A product page describes that product, not every product in a “related items” carousel.
  • Don’t misrepresent the page type. Adding Product to a blog post or Recipe to a page that only links to recipes earns nothing and risks a penalty.
  • Keep it current. Out-of-stock items marked InStock, past events still listed as upcoming, expired job postings without validThrough: all of these mislead users and are treated as quality problems.

You’ll see claims that a particular schema type is required to appear in AI-generated answers. Google’s guidance for its AI features says the same fundamentals apply as for regular search, and that no special markup is needed. Structured data still helps any system parse your page reliably, and it matters most when it matches clearly written visible content. Our guide on AI-readable content covers the page structure side of that.

Validating and monitoring markup

Three tools cover most of the work:

  1. Google’s Rich Results Test shows which Google features a page is eligible for and flags missing required properties. It only reports types Google supports.
  2. The Schema Markup Validator at validator.schema.org checks against the whole schema.org vocabulary, which is useful for types Google doesn’t use for rich results.
  3. Search Console’s enhancement reports show errors and warnings across all pages of a type, which is where template bugs become visible.

Treat errors (missing required properties) as bugs: the page isn’t eligible. Treat warnings (missing recommended properties) as opportunities, and fill them when the data exists.

In code, the most reliable approach is to test structured data the way you test any output: render a fixture page, parse every application/ld+json block, confirm the JSON parses, and check the properties your templates are supposed to emit. A broken comma in a template can silently remove markup from thousands of pages.

Where to start

If you’re picking up structured data on an existing site:

  • List the page types you have (product, article, category, home, location) and map each to the schema.org type that honestly describes it.
  • Check the rich results each type can earn in the Search Gallery, and drop plans that rely on retired features.
  • Generate JSON-LD from your data model, with required properties first and recommended ones where the data exists.
  • Validate one page per template, then watch the Search Console reports for the whole set.

Our structured data guide covers the basics of adding JSON-LD. The principle hasn’t changed in 2026: describe what’s really on the page, completely and accurately, and let the features follow where they still exist.

Tags:structured dataJSON-LDrich resultsschema.orgSearch Console