Stop Adding Schema Markup You Cannot Verify

There's a version of the structured data conversation that goes: add schema markup, get rich results, get more clicks. The reality is messier, and worth understanding before you invest a lot of time in it.

Schema markup tells search engines things about your content that aren't always obvious from reading the page itself. It can help a search engine understand that a paragraph is a product review, that a section describes a recipe's ingredients, or that a question-and-answer exchange follows a particular format. Whether that information results in a rich snippet in search results is a separate question — and one that search engines decide, not you.

What Structured Data Actually Does

Schema.org markup is a vocabulary for annotating content. You add it to your HTML using JSON-LD (a script block in the page head), microdata embedded in your HTML, or RDFa. Most implementations today use JSON-LD because it doesn't require touching the visible HTML structure.

The markup describes relationships: this page is an Article, written by this Person, published on this date, part of this Organization. Or: this page is a Product with this name, this price, and this aggregate rating.

Search engines read that markup and use it to cross-check what they understood from reading the page directly. If the markup says "in stock" and the page text says "in stock," that's a consistent signal. If the markup says "5-star average" and the page has no visible reviews at all, that's a different kind of signal.

The Gap Between Markup and Rich Results

Rich results — those star ratings, FAQs, recipes, and event previews that appear directly in search results — require structured data, but structured data doesn't guarantee them.

Google's documentation is fairly clear about this: structured data makes a page eligible for rich results, not entitled to them. Whether a rich result actually appears depends on the quality of the content, the relevance to the query, the overall trustworthiness of the domain, and factors that aren't fully disclosed.

This matters because some teams implement schema markup, see no change in their search appearance, and conclude that it doesn't work. Sometimes that conclusion is right — the implementation was incorrect. But often the issue is that the page simply isn't a strong enough candidate for that type of rich result, regardless of the markup.

Types of Structured Data Worth Knowing

Article schema helps search engines identify content type, author, publication date, and organization. For news or editorial content, this is often the minimum sensible markup.

Product schema is particularly valuable for e-commerce because it enables price, availability, and review information to appear in search results. Getting this right requires keeping the markup synchronized with actual inventory and pricing — stale data in your schema is worse than no data.

FAQ schema can turn your FAQ sections into expandable question-and-answer pairs directly in search results. This has been tested and found to be useful for specific query types, though Google has shown signs of displaying it less broadly over time.

Review and Rating schema attached to products or local businesses can add star ratings to search listings. The markup needs to reflect genuine reviews that exist on the page, not ratings you've invented for the purpose.

Local Business schema helps Google understand location, hours, contact information, and service types. It's particularly relevant for businesses targeting local searches, where accuracy matters a great deal.

Breadcrumb schema tells search engines about your site's navigation hierarchy. It can cause the URL path shown in search results to be replaced with a more readable breadcrumb trail.

Getting the Implementation Right

The two most common implementation mistakes are markup that doesn't match the visible page content, and markup that's technically valid but applied to the wrong content type.

On the first problem: if your product schema says the price is £49 and the visible page says £59, Google will catch that inconsistency. The rich result won't appear, or if it does appear, it may be flagged as misleading. The fix is to generate your schema markup dynamically from the same data source that drives your visible content.

On the second problem: some publishers add FAQ schema to any page that has headers, even if those headers aren't actually frequently asked questions. Or they add Review schema to products that have no real reviews. These attempts don't work and can trigger manual actions if Google interprets them as attempts to manipulate search appearance.

Testing Before You Ship

Google provides a Rich Results Test tool that evaluates a URL or code snippet and shows you what types of rich results your markup is eligible for. It also identifies errors in the implementation.

This should be part of any deployment process for structured data. A single JSON-LD syntax error — a missing comma, an unclosed bracket — can make the entire markup unreadable.

Beyond the test tool, Search Console's Enhancements section shows structured data detected across your site, including errors and warnings. It won't show you whether your rich results are actually appearing, but it does show you whether your markup is being understood correctly.

When to Prioritize Structured Data

Not every page is worth marking up in detail. The calculation depends on traffic potential and the available schema types.

For a high-traffic product category on an e-commerce site, Product schema is almost always worth the effort because the visibility improvement in search results can directly affect click-through rates. For a general blog post, Article schema is low-effort enough to be standard practice. For a local service business, Local Business schema is close to mandatory.

For content that doesn't map cleanly to any established schema type, the effort is harder to justify. Creating unusual or stretched markup just to have markup rarely helps and can look odd in validation tools.

The Honest Version of the Promise

Structured data is one of those things that works exactly as advertised when implemented correctly and applied to the right content — and produces confusion when treated as a magic ranking ingredient.

The search benefit is real. Rich results do increase click-through rates in most studies. But the path from markup to click goes through content quality, page authority, and search engine interpretation.

Add the markup. Keep it accurate. Don't expect it to rescue pages that have other problems.