Structured data

Structured Data: Why AI Search Lives or Dies by Your Schema

Schema markup is how you label your pages for machines. Here is what to add, what it does and, just as important, what it cannot do.

In plain English

What structured data is

Structured data is a set of labels that help machines understand what is on your page.

Without labels, a page is a block of text. A person can tell which number is your phone number and which line is your address. A machine has to guess. Structured data, written in a shared vocabulary called schema.org, removes the guessing. It is invisible to visitors and clear to machines.

It is one of the foundations of GEO and part of our technical SEO work, working hand in hand with the page structure we set up through on-page SEO.

What schema cannot do

Google says no special markup is needed to appear in its AI features, and schema does not guarantee a citation. It helps machines understand and trust your page. It works best on top of strong content and sound SEO, not instead of them.

The difference

A page with and without schema

The same business, described two ways.
Without schema??A blob of text the AI has to guess aboutWith schemanameUXIQtypeLocalBusinessaddressSomersettelephone07725 732164areaServedDevon, Somerset, DorsetClean, labelled facts the AI can trust

What to add

The six schema types that matter most

These six cover most of what matters. Google keeps a full list in its introduction to structured data.

FAQPage

Marks up questions and their answers.

Google now shows FAQ rich results only for a small set of authoritative sites. The markup stays valid and gives machines a clean question-and-answer structure.

LocalBusiness

Your name, address, phone number, hours and the area you serve.

The most important type for local businesses. It works alongside a complete Google Business Profile.

Article

The headline, author and dates of a guide or post.

Shows who wrote a page and when it was updated, which supports freshness and trust.

HowTo

The steps in a genuine step-by-step process.

Google no longer shows HowTo rich results, but the structure still reads cleanly. Use it only for real step-by-step content.

Speakable

The parts of a page best suited to being read aloud.

Google's support is limited, so treat it as optional rather than essential.

BreadcrumbList

Where a page sits within your site's structure.

Helps machines understand how your pages relate to each other.

For local businesses, start with LocalBusiness and keep it consistent with your Google Business Profile, which is part of local SEO. Article markup supports E-E-A-T by showing who wrote a page.

Real examples

What schema looks like

Both come from our own site. Swap in your own details, and only mark up what a visitor can see on the page.
FAQPage schemaJSON-LD
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "Do you guarantee rankings?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "No. Anyone who guarantees rankings is breaking Google's rules. We optimise for rankings, but we're honest about what we can and can't control."
      }
    }
  ]
}
One of the questions from our own homepage. The text must match the visible answer exactly.
LocalBusiness schemaJSON-LD
{
  "@context": "https://schema.org",
  "@type": "LocalBusiness",
  "name": "UXIQ",
  "url": "https://uxiq.co.uk",
  "telephone": "+447725732164",
  "foundingDate": "2011",
  "address": {
    "@type": "PostalAddress",
    "addressRegion": "Somerset",
    "addressCountry": "GB"
  },
  "areaServed": ["Devon", "Somerset", "Dorset"],
  "serviceType": ["SEO", "Generative Engine Optimisation", "Web Design"]
}
A cut-down version of our own business markup. Add opening hours, a full address and sameAs links to your profiles where you have them.

Answer-style content is where FAQ markup earns its place. See answer engine optimisation for how to write the answers.

What goes wrong

Common structured data mistakes

Four mistakes we find again and again when we audit schema.

No schema at all

Leaving every machine to guess what your pages are about and who you are.

The wrong type

Marking a service page up as an Article, or a business as a person. The label has to match the thing.

Schema that doesn't match the page

Marking up reviews, prices or questions that visitors cannot see. It breaks Google's guidelines and can lead to penalties.

Duplicate, conflicting schema

A plugin and a theme both adding LocalBusiness with different phone numbers. Machines cannot tell which is right.

Do you run a business in the South West? Let's talk.

Every project starts with a free SEO audit. We tell you what we find in plain English, with no obligation.

How we implement it

How we add structured data to your site

Five steps. We never add markup we have not tested.
  1. 1

    Audit what exists

    We find every piece of schema already on your site, including what plugins and themes add without telling you.

  2. 2

    Identify the gaps

    We compare it with what each page type should have, and flag anything wrong or conflicting.

  3. 3

    Implement

    We write clean JSON-LD for the right types and place it where it matches visible content.

  4. 4

    Validate with Google's Rich Results Test

    Every change is checked with Google's Rich Results Test and the Schema Markup Validator before it goes live.

  5. 5

    Monitor

    Search Console reports errors as it finds them, and we recheck after any template or plugin change.

Want to know what your site is saying right now? An AI SEO audit includes a full schema review. For how Google uses all of this, see AI Overview optimisation, and for the bigger picture our AI search readiness service.

Questions

Frequently asked questions

Structured data is a standard set of labels, from the schema.org vocabulary, that you add to a page to say exactly what is on it: a business name, an address, a question and its answer. It is usually written as JSON-LD and is invisible to visitors but easy for machines to read.

Not always. Many website platforms and plugins add the basics. The risk is accuracy: schema must match what is visible on the page and be kept up to date. We implement it properly, test it and fix errors, which is part of our technical SEO work.

No. Google says no special markup is needed to appear in its AI features. Schema helps search engines and AI tools understand a page, and it supports rich results where they apply, but it does not buy a citation. It works best alongside strong content.

LocalBusiness, with your correct name, address, phone number, opening hours and the areas you serve. It is the clearest way to tell machines who you are and where you work. FAQPage and BreadcrumbList are useful supporting types. Pair it with a complete Google Business Profile.

Use Google's Rich Results Test for types Google supports, and the Schema Markup Validator at validator.schema.org for everything else. Google Search Console also reports errors for the structured data it finds across your site. Test every time a template changes.

Incorrect schema is usually ignored, but markup that is misleading or does not match the visible page breaks Google's guidelines and can lead to a manual action that removes rich results. The safe rule is to mark up only what a visitor can actually see on the page.

Let's check what your site is telling machines.

Our free audit reviews your existing schema, finds the gaps and shows which fixes are worth doing first.