Craig Campbell SEO Training & Consultancy

Tutorials

Schema markup for SEO: choose, implement and check structured data

By Craig Campbell · 16 Nov 2026

6 min read

Schema markup helps search engines interpret what a page describes. For SEO, the useful work is choosing accurate markup, implementing it consistently and checking eligibility—not adding as many types as possible.

This guide covers a practical planning, implementation and maintenance workflow. The UK shop and code example are fictional and illustrative only, not client results or a ready-to-deploy template.

1. Understand what schema markup does

Schema markup describes the meaning of a page using a shared vocabulary from Schema.org. Structured data is the broader term for information organised in a machine-readable format. JSON-LD is one way to express that information; it is not a separate kind of SEO.

Google uses structured data to understand content and can use supported types to enable richer search appearances. That does not make every Schema.org type a Google search feature. Start with the page and its actual purpose, not a promise of extra stars or higher rankings.

Keep three questions separate: is the code valid, does it meet a supported feature’s requirements, and does Google actually show that feature? Passing one stage does not prove the next.

2. Choose markup that matches the page

Inventory the page templates that matter: articles, product details, business information and useful navigation. Check the current Google documentation for the feature you want before choosing a type or filling in properties.

  • Editorial article: describe the genuine headline, author and publication details.
  • Individual product: describe the actual product and supported offer information; do not treat a whole category as a single item.
  • Business information: use verified organisation or local business details that match the page.
  • Breadcrumbs: represent a sensible route through the site, not a list of target keywords.

Required and recommended properties vary by feature. A generic generator cannot decide which facts are true about your business. For shop-specific planning, read the ecommerce SEO guide; for business-page context, use the local location pages guide.

3. Check what your site already outputs

Before adding another plugin or code block, inspect what the website already publishes. A theme, SEO plugin, shop extension and custom template may all output structured data. Multiple blocks are not automatically wrong, but conflicting descriptions need attention.

  • Inspect a few real URLs from each template, not only the homepage.
  • Look for old names, incorrect authors, outdated prices and inconsistent addresses.
  • Check whether two tools describe the same entity differently.
  • Agree which tool or template owns each set of data.
  • Record the existing output before changing it.

If you use WordPress, the schema plugins article is a related resource; this guide focuses on choosing and checking markup rather than replacing that older article. Avoid installing another generator simply because a test shows a warning.

4. Build from visible, verified content

Google recommends JSON-LD for supported structured data. It usually appears in a script element with the type application/ld+json. The code itself is not reader-facing prose, but the information it describes must represent the page’s visible content.

Use the same reliable source of truth for the page and its markup. If a product price changes, the structured offer must not keep the old price. If an article has not genuinely changed, do not invent a new modification date to make it look fresh.

Fictional UK example — illustrative only.

A fictional Glasgow cycling shop has a buying guide called “Choosing a waterproof cycling jacket”, visibly credited to Alex Example. The simplified JSON-LD below describes those two facts. It is a teaching fragment, not a complete Google feature implementation or a claim about Craig’s work.

{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Choosing a waterproof cycling jacket",
  "author": {
    "@type": "Person",
    "name": "Alex Example"
  }
}

In a real implementation, replace the fictional facts, consult the relevant feature requirements and add only properties you can support. Do not paste this example across unrelated pages.

Never manufacture reviews, ratings, qualifications or business locations. Good markup describes useful content; it cannot make missing evidence true.

5. Test the code and the live page

Use tools for the question they actually answer. The Schema Markup Validator checks Schema.org-based markup and syntax. Google’s Rich Results Test focuses on supported Google rich-result types. A Schema.org type can be valid without being a feature that Google’s test recognises.

  1. Check a small code sample before changing a whole template.
  2. Use the Rich Results Test guide to interpret critical errors and non-critical issues.
  3. Publish a limited set of approved pages and test their live URLs.
  4. Check that the rendered content and extracted markup agree.
  5. Use Search Console’s URL Inspection to investigate whether Google can access the page.
  6. Retest other pages that share the changed template.

Automated validation does not check every quality guideline or prove that your claims are accurate. Keep a human content check alongside the tool results. If a page is blocked, unavailable or accidentally noindexed, investigate that with the technical SEO audit guide instead of adding more properties.

6. Monitor without promising results

Record the deployment date, affected templates and examples tested. Where Search Console provides an enhancement report for your markup, review new errors and affected URLs after Google processes the changes. Not every type has its own report.

For supported search appearances, examine the available performance filters alongside page and query data. The Search Console guide explains the wider evidence. Compare similar periods and account for seasonality, content changes and changes in search demand.

A valid test does not guarantee a rich result, a better ranking or more enquiries. A rise in clicks after implementation does not establish that schema caused it. Use the SEO testing guide when planning a more cautious evaluation.

Checklist and next step

  • Does the chosen type describe this page accurately?
  • Have you checked the current requirements for the intended Google feature?
  • Do the facts match visible content and genuine business records?
  • Have you checked existing output for conflicts?
  • Did both the sample and live-page tests answer the right questions?
  • Is someone responsible for keeping the data accurate?

Start with one important template and a few representative pages. Fix incorrect output before expanding coverage. Include schema maintenance in your content refresh workflow so later edits do not leave the markup behind.

Sources

Public documentation checked on 16 November 2026. Feature availability and requirements can change; check the current documentation before implementation.

Let's work together.

Whether you need an SEO partner, a consultant, a speaker for your event or training for your team — get in touch and let's chat.

Get in touch
Get in touch