Schema Markup in 2026: The Complete Guide to Structured Data

What schema markup actually is

Schema markup (also called structured data) is a small, standardised block of code you add to a web page that spells out, in a language search engines and AI assistants understand, exactly what the page is about. Instead of leaving Google to guess whether "Apple" means the fruit or the company, or whether a number on the page is a price, a phone number or a review score, schema labels each piece of information explicitly — this is the business name, this is the opening hours, this is the author, this is the rating.

The vocabulary itself comes from Schema.org, a shared standard backed by Google, Microsoft, Yahoo and Yandex. The recommended way to add it is JSON-LD — a script block that sits in your page's code, completely separate from the words your visitors read. That separation is the whole point: you describe your content for machines without touching how it looks for humans.

Get it right and you become eligible for richer, more eye-catching listings in Google — star ratings, prices, sitelinks, business panels — and, increasingly, you make it easier for AI assistants like ChatGPT, Gemini and Perplexity to read, trust and cite you. Get it wrong (or fake it) and you gain nothing, and occasionally you trigger a manual penalty. This guide walks through every part that matters in 2026, with UK businesses firmly in mind.

Why schema matters more in 2026, not less

For years the pitch for schema was simple: rich snippets get more clicks. That is still true for the formats Google supports. But 2026 added a second, bigger reason — AI search runs on structured data.

When ChatGPT, Google's AI Mode or Perplexity assemble an answer, they have to decide which sources are trustworthy and what each page is actually claiming. Clean structured data makes that job easier. SE Ranking's analysis of AI citations found that roughly 65% of pages cited by Google's AI Mode and 71% of pages cited by ChatGPT contained structured data — not proof that schema alone earns a citation, but a strong signal that the pages winning in AI search tend to be the well-structured ones.

There is an honesty test built into all of this. Modern AI systems compare your structured data against the visible content on the page. If your schema claims a 4.9 rating that does not appear anywhere a human can see, the model trusts the visible text and quietly marks your page as less reliable. The lesson for 2026 is blunt: schema describes reality — it does not invent it.

JSON-LD vs Microdata vs RDFa

There are three ways to write structured data, but in 2026 the decision is already made for you.

JSON-LD wins because it is self-contained. You can add, debug and update it without disturbing your page layout, and every major engine — Google, Bing, Perplexity and ChatGPT — reads it natively. If a plugin or developer suggests Microdata in 2026, push back.

What still earns rich results in 2026 (and what Google dropped)

This is where most older guides mislead you. Google has quietly narrowed the list of structured data types that actually produce a visual rich result, even though the underlying Schema.org types remain valid. The single biggest change of the year: Google deprecated FAQ rich results on 7 May 2026. FAQ snippets no longer appear in Search, the dedicated reports and Rich Results Test support are being retired, and the Search Console API support is winding down through the summer.

Crucially, FAQPage is still a valid Schema.org type and leaving it on your pages causes no harm — it simply won't draw a rich result anymore. The same pattern applies to HowTo, which Google removed back in 2023. And over the past year Google has also dropped support for several niche types, including Book Actions, Course Info, ClaimReview, Estimated Salary, Learning Video, Special Announcement and Vehicle Listing.

Google still lists roughly 30 supported structured data features in its Search Gallery. That document is the only list you should trust — bookmark it and check it before you invest time in any single type.

The five schema types most UK businesses actually need

You do not need all 30. For the vast majority of service businesses, local firms and content sites, five types cover the overwhelming majority of the value: Organization, LocalBusiness, Article (or BlogPosting), BreadcrumbList and Product or Service. Here is what each does and a stripped-back JSON-LD example for two of the most important.

Organization — tells Google who you are

This anchors your brand as an entity: your name, logo, URL and social profiles. It is the foundation for a knowledge panel and helps AI assistants attribute information to the right business.

LocalBusiness — the workhorse for local SEO

If you serve a town or region, this is the most important type you can add. It declares your name, address, phone, opening hours and area served — the exact details Google leans on for the Map Pack and that AI assistants quote when someone asks for a recommendation near them.

The other three

Article / BlogPosting marks up the headline, author and publish date of your content — vital for establishing authorship and freshness, both of which matter for AI citation. BreadcrumbList shows your site hierarchy in the results and helps machines understand structure. Product or Service describes what you sell, including price and availability for products, or service type and area for service businesses.

How to add schema, step by step

1. Audit what you already have. Run your key pages through Google's Rich Results Test and the Schema.org validator to see what's present and what's broken.
2. Pick the right type per page. Homepage and about page get Organization (and LocalBusiness if relevant); blog posts get Article; product pages get Product; every page can carry BreadcrumbList.
3. Generate the JSON-LD. Use a generator, an SEO plugin, or write it by hand. Keep every value truthful and matched to the visible page.
4. Add it to the page. Drop the <script type="application/ld+json"> block into the page. On WordPress, a plugin like Yoast or Rank Math handles much of this automatically.
5. Validate before you publish. Re-run the Rich Results Test. Fix every error and review the warnings.
6. Monitor in Search Console. The Enhancements reports flag structured data issues across your whole site over time.

Timeline: a single page takes minutes once you have a template. A small business site of 10–20 pages is realistically a day's careful work. Google then needs to recrawl — expect rich results to appear within days to a few weeks, not instantly.

What schema costs in the UK

Schema is one of the cheapest high-leverage jobs in SEO because the markup itself is free — you are paying for time and accuracy, not licences. As an indicative guide for UK work (ranges vary widely by site size and agency):

• DIY with a plugin: effectively £0 beyond your existing tools — most SEO plugins include schema features in free or low-cost tiers.
• Freelancer, one-off setup: a typical indicative range of roughly £150–£600 for a small business site, depending on page count and complexity.
• Agency, as part of a wider SEO retainer: usually bundled rather than billed separately.

These are clearly-labelled indicative ranges, not quotes — always get a firm price for your specific site.

Who schema suits, and who can skip it

It suits local service businesses (LocalBusiness is close to non-negotiable for the Map Pack), e-commerce stores (Product and Review snippets directly lift click-through), publishers and bloggers (Article and authorship signals), and any brand that wants to be understood as an entity by AI assistants.

It matters less for a brand-new site with no content and no rankings yet — you'll get more from publishing useful pages first — and for purely internal tools or pages you deliberately keep out of search. Even then, basic Organization markup is a sensible early win.

Tools to generate and validate schema

• Google Rich Results Test — the definitive check for whether a page is eligible for a Google rich result.
• Schema Markup Validator (schema.org) — validates against the full vocabulary, including types Google doesn't render.
• Google Search Console — the Enhancements reports surface structured data issues across your live site.
• SEO plugins (Rank Math, Yoast, Schema Pro) — automate common types on WordPress.
• JSON-LD generators — handy for one-off blocks you paste in by hand.

Inside The KillerEdge, members get a growing tool arsenal in The Armory — a live-SERP Content Optimizer and AI Humanizer today, with an SEO Auditor and more on the way — to take the guesswork out of on-page work like this.

Common schema mistakes

• Marking up invisible content. Your schema must reflect what a human can actually see on the page. Hidden ratings or fake review counts are the fastest route to a manual action — and AI models discount them too.
• Self-serving review markup. Google restricts site-level "self-reviews"; AggregateRating must come from genuine, on-page reviews following its guidelines.
• Leaving deprecated types as your strategy. Don't build your 2026 plan around FAQ or HowTo rich results — they're gone. Keep the markup if it aids clarity, but don't expect a snippet.
• Inconsistent NAP. For local businesses, your name, address and phone in schema must match your Google Business Profile and citations exactly.
• Never validating. One stray comma can break an entire JSON-LD block. Always test.
• Forgetting to update. Stale prices, old opening hours or a former author name undermine trust with both Google and AI engines.

Schema, AEO and getting cited by AI

The most forward-looking reason to take schema seriously in 2026 is answer-engine optimisation (AEO) — being recommended by AI assistants, not just ranked by Google. Structured data gives AI systems a clean, machine-readable summary of who you are, what you offer and why you're credible. Organization and Author markup help attribute your expertise; Product and LocalBusiness markup give assistants the concrete facts they quote back to users.

Schema is not a magic citation button — genuine expertise, clear writing and real authority still do the heavy lifting. But it tilts the odds. If you want the full playbook on earning AI citations, read our AEO playbook, and pair it with the wider technical SEO guide so your foundations are solid.

The 2026 view: where this is heading

The direction of travel is clear. Google is pruning rich results that became spam magnets (FAQ, HowTo and the niche types it dropped) while leaning harder on structured data behind the scenes to understand entities and feed AI experiences. In other words, schema is becoming less about decorative snippets and more about being correctly understood and trusted by machines.

That makes the honest, accurate use of schema a quiet competitive edge. The businesses that keep their structured data truthful, consistent and aligned with their visible content will be the ones AI assistants are comfortable recommending. The ones still chasing deprecated snippets or stuffing fake ratings will fall behind. If you serve a local market, start with LocalBusiness and Organization, get them right, and build from there — and lean on our local SEO guide for the bigger picture.

Want this done properly?

Schema is one piece of a winning marketing system — the rankings, the links and the leads all have to work together. Inside The Dojo we teach the whole thing in plain English, with the tools to do it. Join The Dojo — £67/mo, cancel anytime. Prefer it built and run for you? Take a look at our Done For You partnership.