Structured Data for SEO The Complete Beginner’s Guide

Structured data is a standardized format usually JSONLD code that tells search engines explicitly what a piece of content is, rather than leaving them to infer it from the visible page alone. So if a page discusses an SEO consultant, structured data can state directly this is a Person, this is their job title, this is the organization they work for, this is their area of expertise. Search engines can often infer this from context anyway, but structured data removes the guesswork.

If semantic SEO is about writing content so its meaning is clear, structured data is about stating that meaning in a format machines don’t have to interpret at all. The two work together, but they’re not the same thing, and confusing them leads to two common mistakes expecting markup to substitute for actual content quality, and skipping markup because “the content already explains it clearly.”

This guide covers what structured data actually does, which schema types are worth using in 2026 (several widely repeated recommendations are now outdated), how to implement it correctly, and how to test that it’s working.

Structured Data for SEO vs. Schema Markup A Quick Distinction

These terms get used interchangeably, but technically

• Structured data is the organized information itself the facts about a page’s content, arranged in a predictable format.
• Schema markup is the vocabulary used to express that information, defined by the collaborative Schema.org project (backed by Google, Bing, Yahoo, and Yandex).

In practice, most people say “schema” to mean both. That’s fine for everyday use the distinction mainly matters if you’re reading Schema.org’s own documentation, which uses the terms precisely.

Is Structured Data for SEO a Ranking Factor?

No and Google has been consistent and explicit about this for years. Structured data doesn’t directly move rankings. What it does is help search engines understand a page’s content more precisely and, for some schema types, makes a page eligible for enhanced search features (rich results) that can improve clickthrough rate. Better understanding and better CTR can indirectly support performance, but adding schema to a thin or lowquality page won’t make that page rank better it just makes accurate metadata about a weak page.

This matters because a lot of published advice implies schema is a growth lever on its own. It isn’t. It’s a clarity layer on top of content that already needs to be good.

The Schema Types That Actually Matter in 2026

This is the section where a lot of existing guides including earlier drafts of this one are out of date. Google has meaningfully narrowed which schema types generate visible search features over the past three years, and treating this list as static is a common source of wasted implementation effort.

Article schema still standard and useful for blog posts, guides, and educational content. Helps establish headline, author, publish date, and publisher clearly. No major restrictions.

Organization schema useful for establishing a company’s identity, logo, and official social profiles (via the sameAs property), which supports entity recognition in the Knowledge Graph. Does not make a page eligible for starrating rich results see the review schema note below.

Person schema useful for author pages and expert bios, and increasingly relevant as Google weighs authorship and expertise signals. Should be applied consistently across every page where that person is credited.

Breadcrumb schema loweffort, still functional, helps both search engines and users understand site hierarchy. Worth implementing sitewide as a baseline.

FAQ schema largely retired as a visible feature.

Google restricted FAQ rich results to a narrow set of government and health sites in August 2023, then retired the FAQ rich result entirely across Google Search on May 7, 2026 including for those previously exempt sites. FAQ Page markup still validates and doesn’t cause errors, and Google has said unused structured data doesn’t harm a page, but it no longer produces the expandable dropdown in search results for any site. Don’t plan new content around it as an SEO tactic. Genuine, well organized Q&A content on a page can still help readers and can still help AI systems extract a direct answer to cite the value moved from “visible SERP feature” to “content clarity,” not away entirely.

Review and AggregateRating schema restricted for businesses reviewing themselves.

This is the correction most guides get wrong. Since 2019, Google has excluded self serving reviews from star rating rich results meaning if Organization or Local Business schema is used to display reviews about that same entity on its own site, no stars will show, regardless of whether the reviews are genuine. This applies even if the reviews are collected through a third party widget, because the business still controls what’s displayed.

Moreover, the restriction does not apply to Product, Recipe, Software Application, or similar entity types a store can legitimately mark up and display star ratings for a specific product it sells. For a consultancy like Almost SEO, this means Organization level review schema won’t produce visible stars; testimonials are still valuable content, but if star ratings are the goal, the working paths are third party review platforms (Trustpilot, Google Business Profile) or marking up a specific productized offering (a course, a template) rather than the business as a whole.

As of July 2026, Google also explicitly bans including fake or undisclosed incentivized reviews in either visible content or markup a real compliance risk, not just a best practice suggestion, since violations can trigger a manual action.

Product schema

Central for ecommerce, still fully supported for price, availability, and (per the exemption above) legitimate productlevel reviews.

Local Business schema

Still useful for local search visibility (address, hours, service area), separate from the reviewstar restriction discussed above.

Event schema

still functional for webinars, workshops, and conferences, though eligibility and display format have narrowed over time; check current documentation before relying on it for a specific display feature.

How To schema

Deprecated on desktop search results as of 2023, following the same pattern as FAQ. Not worth implementing for a visible rich result, though the underlying content structure (clear numbered steps) remains good practice regardless of markup.

The pattern across all of these Google has been steadily narrowing which schema types produce a visible search feature, while maintaining that the underlying markup still helps machine understanding of a page. Treat schema decisions based on what’s currently eligible for a rich result, not on lists that were accurate two or three years ago.

JSONLD The Format to Use

Three formats can express structured data JSONLD, Microdata, and RDFa but JSONLD is Google’s recommended format and the one worth defaulting to. It’s a selfcontained script block, separate from the page’s visible HTML, which makes it easier to implement, test, and update without touching the content itself.

A basic Article schema example looks like this

<script type=”application/ld+json”>

{

  “@context” “https//schema.org”,

  “@type” “Article”,

  “headline” “Structured Data for SEO The Complete Beginner’s Guide”,

  “author” {

    “@type” “Person”,

    “name” “Author Name”

  },

  “publisher” {

    “@type” “Organization”,

    “name” “AlmostSEO”,

    “logo” {

      “@type” “ImageObject”,

      “url” “https//almostseo.com/logo.png”

    }

  },

  “datePublished” “20260806”,

  “dateModified” “20260806”

}

</script>

This sits in the page’s <head> and doesn’t need to duplicate visible content it describes what’s already on the page in a machinereadable format.

Structured Data For SEO and Entity Recognition

Structured data supports entity SEO directly through properties like sameAs, which links an entity (a person, a business) to its existing presence elsewhere a Wikipedia page, a Wikidata entry, an official LinkedIn or social profile. This gives search engines a way to confirm “this AlmostSEO mentioned in an article is the same AlmostSEO with this website and these social profiles,” rather than treating each mention as a fresh, unconnected reference.

For a growing content site, the compounding value here is real every new article that consistently identifies the same author and organization entities reinforces the same knowledge graph connections, rather than starting from zero each time.

Implementation Process

1. Identify the content type first. A blog article gets Article schema; a service page gets Organization or Service schema; an FAQformat page (for its content value, not the richresult mechanism) can still use FAQPage schema even without expecting a visible SERP feature.

2. Define the main entity clearly. Every page should have one unambiguous subject a specific person, product, service, or piece of content. Markup that tries to describe multiple unrelated entities on one page tends to confuse rather than clarify.

3. Keep information consistent across every platform. Name, address, and other identifying details should match exactly across the website, Google Business Profile, social profiles, and any directories. Inconsistency between these sources is one of the more common reasons entity recognition stays weak even with markup in place.

4. Test before publishing. Run the page through Google’s Rich Results Test to confirm the markup validates and check which features (if any) it’s eligible for. This step is frequently skipped, and a single JSON syntax error can silently invalidate an entire markup block.

5. Maintain it. Structured data that describes something that later changes a price, an author who’s left, an event date needs updating alongside the content itself. Stale markup that contradicts the visible page is a policy risk, not just an accuracy gap.

Structured Data and AI Search

Google’s own guidance on AI Overviews and AI Mode states there’s no special schema requirement for a page to be used or cited in AI generated answers structured data should simply match the visible content on the page, not substitute for it. The more reliable lever for AI citation is well organized, directly answered content; markup at most reinforces what’s already clear from the text.

That said, clean structured data still has a supporting role consistent entity markup helps establish which brand, author, or organization is actually saying something, which matters for the trust signals AI systems weigh when deciding what to cite. It’s a supporting signal, not a shortcut around content quality.

Common Mistakes

Treating FAQ schema as an active ranking or visibility lever. As covered above, this changed materially in 2026 implementing it for that reason is wasted effort now.

Adding review schema to Organization or LocalBusiness pages, expecting stars to appear. They won’t, under Google’s selfserving review restriction regardless of how genuine the reviews are.

Skipping the Rich Results Test. Unvalidated markup can fail silently; a page can carry broken schema for months with no visible symptom until someone checks it directly.

Letting markup drift out of sync with visible content. Schema describing something no longer true on the page is both inaccurate and a policy risk.

Implementing every available schema type regardless of relevance. More markup isn’t inherently better irrelevant or inaccurate schema adds maintenance overhead without adding understanding.

Quick Audit Checklist

Before publishing or reviewing an existing page

• Does the schema type match what the page is actually about?
• Does every claim in the markup match what’s visibly on the page?
• Has it been run through the Rich Results Test with no errors?
• Is the entity information (name, author, organization) consistent with every other page and platform?
• If review/rating schema is present, is it on an eligible entity type (Product, not Organization/LocalBusiness)?
• Is any FAQ or HowTo schema present there for genuine content value, not for a richresult expectation that no longer applies?

Frequently Asked Questions