Most Websites Underuse Schema Markup. Here Is the Fix.

Schema markup has been around since 2011. Fifteen years. Google, Bing, Yahoo, and Yandex got together-practically holding hands-and created a shared vocabulary so websites could explain their content to machines in plain, unambiguous terms. They literally built a universal language for the web.

And after a decade and a half, many websites still don't use it, and most that do use only the basics. W3Techs finds JSON-LD on a little over half of websites, and far fewer implement it strategically with connected entities. That number is wild. Imagine if 69% of restaurants didn’t bother putting their name on the building. That’s essentially what two-thirds of the web is doing-running a business with no sign on the door, then wondering why the machines walking past can’t find them.

Why Schema Got Ignored (And Why That’s Changing)

Let’s be fair about why. Schema markup was traditionally seen as a “nice to have.” It wasn’t a direct ranking factor. Google’s Gary Illyes said as much in 2017, carefully phrasing it as: schema “will help us understand your pages better, and indirectly, it leads to better ranks in some sense.” Which is Google-speak for “it matters, but we’re not going to tell you how much.”

So SEOs prioritized things with more obvious ROI-content, backlinks, page speed. Schema got pushed to the “we’ll get to it” pile on the roadmap. And for years, that was arguably rational.

Then AI showed up and flipped the table.

Why Schema Matters More Now

Here’s what changed. AI systems-Google’s AI Overviews, Bing’s Copilot, ChatGPT’s web search, Perplexity-don’t read websites the way humans do. They don’t appreciate your beautiful hero image or your clever brand voice. They parse structure. They extract entities. They look for explicit, machine-readable relationships between concepts.

Schema markup is exactly that: explicit, machine-readable relationships between concepts.

Fabrice Canel, Microsoft Bing’s Principal Product Manager, has been beating this drum since 2023. At SMX Munich in March 2025, he restated it plainly: “Schema Markup helps Microsoft’s LLMs understand content.” Not “might help.” Not “could theoretically influence.” Helps. Present tense. Active voice.

Google’s documentation reinforces this: “You can help us by providing explicit clues about the meaning of a page to Google by including structured data on the page.” And Google’s Knowledge Graph-which feeds directly into AI Overviews and Gemini-is enriched by crawling schema markup across the web.

The Numbers That Should Get Your Attention

The data on schema’s impact is consistently compelling, even if it never got the headline attention that backlinks or Core Web Vitals did.

Pages that earn rich results (star ratings, prices, availability, event details) can see higher click-through rates; Google's published case studies include Rotten Tomatoes (25% higher CTR on pages with structured data) and Food Network (35% more visits after enabling search features). Videos using VideoObject markup saw 3x impressions and 2x clicks within a year.

And that’s just the traditional search impact. For AI visibility, the picture differs by engine: Microsoft has said schema helps its LLMs understand content, while Google says structured data isn’t required for its AI features and there’s no special markup for them. Treat schema as helpful support, not the deciding factor in whether you are cited.

What Google Deprecated (And What It Didn’t)

If you heard that Google killed schema, you heard wrong-but there’s a grain of truth in the noise. In June 2025, Google announced it would phase out seven structured data types, including Course info, Claim review, Estimated salary, SpecialAnnouncement, and Vehicle listing; Practice Problem followed in January 2026, and the Sitelinks Search Box was already retired in November 2024.

Google’s John Mueller clarified that this is a visual and functional refinement-not a signal that structured data is less important. Sites using deprecated types won’t see ranking drops. They just won’t get rich results for those specific implementations.

Meanwhile, the schema types that drive the most business value remain fully supported: Product, Article, LocalBusiness, Organization, Review, Event, Video, and Recipe. FAQ and HowTo markup is still valid, but Google no longer shows rich results for either. The deprecations pruned the edges. The core is stronger than ever.

Why It’s Still So Painful to Implement

If schema is this valuable and this established, why do so many sites still lack it? The honest answer: because implementing it properly is a tedious, developer-dependent, error-prone process that most SEO teams don’t have the bandwidth to prioritize.

Getting schema right means mapping your content to the correct schema.org types (there are over 800), generating valid JSON-LD for every relevant page type, testing against Google’s Rich Results Test, deploying through your CMS or via tag manager, and then maintaining it as your content changes. For a site with hundreds or thousands of pages across multiple templates, this is a significant technical project that requires coordination between SEO, development, and content teams.

And the kicker: most SEOs can write the schema. They just can’t get the development team to implement it before Q4. Or Q1 of next year. Or ever.

The Bottleneck Isn’t Knowledge. It’s the Dev Queue.

This is one of those tasks that perfectly illustrates the gap between knowing what to do and having the bandwidth to actually do it. Every experienced SEO knows schema matters. Precious few have the time to generate, validate, and deploy it across an entire client’s site-especially when the client has 47 product categories, 12 location pages, and a blog with 300 posts, none of which have Article schema.

Most schema tools solve half the problem: they generate the markup, then hand you a file and say “good luck getting your developer to implement this.” The SEO ends up in a ticketing queue behind three sprint cycles and a redesign. That’s why platforms like Silverbee take a different approach-the AI teammate for SEO reads the actual page content, determines the correct schema types, generates valid JSON-LD, validates against Google’s Rich Results requirements, and delivers the implementation-ready code directly to the SEO, ready to deploy via tag manager or CMS plugin without touching the dev backlog. It turns a multi-week cross-team coordination project into a same-day deliverable.

The Window Is Now

With so many websites still running without useful schema, there’s a real competitive gap.

The sites that close this gap first-especially with entity-rich, relationship-aware schema that builds proper Knowledge Graphs-will have a meaningful advantage