Schema Markup in 2026: What It Does, What It Doesn't, and Why It Still Matters

Schema markup has been around since 2011. For most of that time, it was considered a relatively benign feature for informing rich snippets. 

Then AI search arrived, and for a while, it looked like schema had hit the big time. GEO articles - including this one (since updated) - were suddenly talking about schema like it was the silver bullet for being cited in Google AI Overviews and ChatGPT. Turns out, that wasn’t quite right. 

In May 2026, Google published its first official guide to optimising for generative AI search, and promptly turned schema’s limelight off. It turns out schema isn’t any more important than it was before. 

That raises the question: Is schema markup still worth bothering with in 2026? Happily, the short answer remains an emphatic yes. Just (probably) not because of AI.

So, What’s the Official Position on Schema Markup? 

Google’s new guide isn’t long, and it’s worth a careful read. But, keeping things short, structured data isn’t a requirement of generative AI search. There is no special schema markup you need to add. 

As others have said, this is quite the correction to a lot of what has been said about search over the past 12-24 months.

All of that said, schema hasn’t been rendered suddenly worthless. And Google isn’t the only AI search platform out there. It may be the dominant force in its market, but ChatGPT, Claude, Perplexity, (etc…) do not necessarily follow the same rules. 

Besides schema’s value has always extended far beyond just AI search.

💡 Further reading: How to Optimise Your Website for AI Search

Quick Refresher: What is Schema Markup?

But let’s not get ahead of ourselves. What is schema markup?

Schema markup is structured data added to a webpage’s HTML (usually written in JSON-LD) that helps search engines understand exactly what they’re looking at, which in turn supports AI search and LLM results.

There are hundreds of schema types and over 800 vocabulary terms in total, but some of the most common help search engines understand things like… 

• Product information: price, availability, reviews
• Business information: address, phone numbers, opening hours
• Event information: date, time, location
• FAQs
• Articles and blog content

Tools like Google’s Rich Results Test or the Schema.org validator help ensure your markup is clean and interpretable. Incidentally, for a full list of available schema markup, head over to schema.org for a comprehensive community-supported repository. 

An example of WebSite Markup:

{

 "@context": "https://schema.org",
 
 "@type": "WebSite",
 
 "name": "Our Name is Mud",
 
 "url": "https://ournameismud.co.uk/",
 
 "potentialAction": {
 
   "@type": "SearchAction",
   
   "target": "https://ournameismud.co.uk/?s={search_term_string}",
   
   "query-input": "required name=search_term_string"
   
 }
 
}

This tells search engines (then used by LLMs):

• The site is called Our Name is Mud
• The homepage is ournameismud.co.uk
• Users can search the site using a built-in search bar

Adding this helps search engines understand your site as a whole entity, improving its representation in search results.

💡 Further reading: Web Design & SEO: Building a Website that Ranks in 2026

Does Schema Matter for SEO in 2026?

The internet would have you believe SEO is dead. It’s not. And if you strip away the hype surrounding AI and LLMs, you realise that the machines crawling your website still benefit from a little guidance. 

Schema remains the most reliable way to provide that assistance. And in 2026, schema’s relevance comes down to 3 core dimensions: understanding, visibility, and trust.

Understanding

Schema gives search engines and AI systems the context they can’t safely assume on their own. It clarifies entities, relationships, and meanings; the raw ingredients behind AI Overviews and entity-based search.

When you mark up a page with Article, Product, or LocalBusiness, you’re not pointlessly decorating the HTML. You’re disambiguating your content.

This is increasingly important as AI models rely on entity graphs to determine which sources to reference. Structured data helps position your website within those graphs.

It's worth noting that LLMs don't crawl schema markup directly. Instead, they rely on the structured understanding built by search engines and knowledge graphs, systems that use schema extensively. 

Google has clarified that its own AI features don't require schema specifically. But the context that schema helps establish - across search engines, knowledge graphs, and other AI platforms - still influences how confidently those systems can reference and cite your content.

In practice, this means schema still influences how confidently AI platforms can reference, interpret, and cite your content.

Visibility

When rich snippets initially rolled out back in 2009, they felt like a dramatic upgrade to Google’s SERPs. Although they have evolved since then, rich snippets have become part of the furniture of search.

But one thing everyone knows about furniture is that it’s essential. Your structured data still feeds countless systems that determine where and how your content appears. Schema supports:

• Knowledge Panels, giving Google and Bing confidence in who you are
• AI Overviews and generative results, where entity clarity supports - but doesn't directly drive - inclusion.
• Voice assistants, which rely on structured data to return accurate, contextual answers

• Product knowledge graphs, including Google Merchant Center

   

Trust & Authority

Trust has long been at the heart of SEO. Google’s E-E-A-T framework, backlink profiles, and author credibility have shaped organic performance for over a decade. Schema strengthens those signals.

By marking up authors, organisations, reviews, locations, and “about” details, you’re making your existing trust signals easier for machines to verify. Structured data helps reinforce:

• Who wrote this content
• Who owns this website
• What your business does
• Where you’re located
• What others say about you

In other words, schema acts like connective tissue between your website and the broader ecosystem of signals Google evaluates.

The importance of this information remains unchanged with AI search. 

LLMs depend on structured, cross-referenced information to reduce hallucination. When your business or content is clearly defined through schema, models can assign a higher internal confidence score, making you more likely to appear in and be cited by AI-generated answers. 

Schema alone doesn’t make you authoritative. But it reduces friction and makes clear your authority. 

The Real Role of Schema in 2026 

When schema was introduced nearly 15 years ago, it could have been accurately described as a "rich snippets booster". These days, it's far more integral - a key part of your metadata infrastructure. 

But it's worth being precise about what it actually does, because the AI hype has muddied the waters somewhat. 

Entity-Based SEO is the New Baseline 

Keywords have never been less valuable to search*, which is increasingly organised around entities. And that shift makes schema invaluable. It gives search engines clear definitions of who you are, what you offer, and how you relate to other people, places, and things. 

* They're still super important, just relatively less so. 

In practical terms, schema answers questions like: 

• Who is behind this website? 
• What type of content or product is this? 
• How should this be categorised or connected?

That entity-level clarity feeds Google's Knowledge Graph, which in turn underpins much of how Google's AI features assemble responses. 

So, even though Google has said schema isn't required for AI Overview visibility, the context that schema establishes still does indirect work in the background. 

What Does Schema Influence in 2026? 

So, we’ve talked a lot - too much, perhaps - about what schema, it turns out, doesn’t do in 2026. Now let’s talk about what it does do… 

Rich results in traditional Search. Star ratings, prices, FAQs, recipe metadata, event details. Schema is still the path towards enriching your search results. 

Voice assistants. Google Assistant, Alexa, and Siri lean heavily on structured data for accurate, contextual answers. 

Google Merchant Center and product knowledge graphs. For ecommerce, schema couldn’t be more valuable or essential. 

Local SEO. LocalBusiness markup feeds Google Business Profile and local pack visibility.

Knowledge panels. Organisation and Person schema help Google verify who you are. 

Other AI platforms. ChatGPT, Claude, Perplexity, and others don't follow Google's playbook. Schema still - though possibly indirectly - helps these systems parse your content reliably. 

That’s a long list of important stuff. Anyone telling you that schema no longer matters isn’t thinking outside the box. 

What’s Coming Next?

Schema.org continues to evolve with AI search in mind. We’re seeing growing emphasis on:

• Speakable markup for voice interfaces
• Author and Person schema to validate expertise
• Product and Review refinements for commerce

• FAQ and How-To that supports rich results and structured presentation

   

Is Schema Still Worth it in 2026?

Yes. But it’s important to be clear about the reasoning. 

Schema has recently been touted as a silver bullet for AI citations. That isn’t the case, especially where Google is involved. And Google has told you that much itself. 

The overriding reality continues to be that AI citation is driven by other factors: ranking well in traditional search (though not essential), having strong authority and trust, and producing genuinely useful first-hand content. 

This is all stuff Google has been saying for years, by the way. 

But none of that makes schema an optional extra. Our perspective, as web developers/designers, is that it’s important to have all of these best practice additions in place and accounted for on every build. 

If someone tells you otherwise, you’re going to want to inspect that red flag a little more closely. 

And, at the end of the day, implementing comprehensive schema isn’t hard at all. If your site's schema is incomplete, outdated, or missing entirely, that's almost certainly worth fixing before you worry about more exotic AI optimisation tactics.