What’s the Deal with Schema

Subheader

The pros and cons of structured content

Lead-in

It’s not new and it’s not sexy, but Schema.org is getting a lot more attention these days. The reason is AI.

You’ve probably heard that Schema improves your AI search results — apply it to your site and voilà, better results. But what is it? How does it help with AI? And, more controversially, does it actually help at all?

Featured Image

Sections

Schema.org was created in 2011 in a joint venture between the major search engines. They wanted a universally accepted set of terms to describe and categorize content on the web.

By labeling content with terms like “Job Title” or “Service Area,” search engines could understand what a page is actually about, instead of guessing from the surrounding text. The result is more accurate search results — for both search engines and AI platforms.

Schema.org wasn’t the first attempt at this, but at the time the web lacked a single, unifying vocabulary for structured content. Today it’s the primary vocabulary for describing online content, and it’s still actively updated and supported.

The benefit to search engines is clear. Applying it to every website is another matter.

The first challenge is time. It’s already a lot of work optimizing a site for SEO, accessibility, performance, and design. One more thing to worry about isn’t a welcome ask — especially on a large site with hundreds or thousands of pages.

The second is need. While the reasons behind Schema are logical, the truth is that search engines don’t seem to require it! Google’s crawlers are already good at reading and interpreting a page, whether Schema is present or not. It’s more like a “nice to have.”

Asking people to do more work for a benefit they can’t see isn’t a great pitch. That’s why I think Schema has largely been ignored for the past 15 years.

So why the renewed attention? AI. Everyone wants their content cited by ChatGPT and other platforms, and Schema is being sold as a kind of cheat code — apply it and your AI visibility goes up. But is that true? The answer, unfortunately, is a bit muddled.

Let’s burst the biggest bubble right away. Adding Schema won’t automatically increase your AI citations.

It’s an idea repeated so often that even our team started to believe it. But a recent study from Ahrefs was a much-needed reality check. Across 1,885 webpages that added structured data, they found no meaningful increase in AI citations.

The study isn’t meant to be the final word in the matter. Almost all the pages tested were already being cited by AI, so it couldn’t measure whether Schema helps sites that aren’t yet visible. Ahrefs did find that AI-cited pages were roughly three times more likely to use structured data (great!). But they attribute that to overall site quality, not Schema itself (hmm).

Here are my takeaways. AI prefers structured content, because it’s easier to crawl, understand, and index. The better you structure your content, the better the odds AI finds it useful. That’s not a guarantee your citations will jump, but it could help, especially if your content is invisible today. Treat it as a best practice that offers incremental gains.

Again, that’s not a strong case for pouring time and money into Schema. But the good news is it doesn’t have to be complicated to implement.

Schema isn’t the magic bullet many are hoping for, but it does have some real upsides.

The first is SEO. Yes, Google can index your site without it, but structured data makes your pages easier to crawl and surface in results. It’s hard to quantify, but for sites that lacked exposure before, it can make your most valuable content more visible.

The second is rich results. On Google, some listings show more than a blue link and text summary. A search for restaurants may include a star rating next to a result, for example. A search for “how to cook asparagus” may return some nicely formatted recipe cards. Those richer entries are powered largely by Schema data.

A third, much more subtle benefit is helping AI understand your organization better. Some Schema types let you link your content to known entities, such as connecting your organization to its Wikidata listing, so AI can tell you apart from a similarly named company. It won't stop hallucinations entirely, but those connections feed the knowledge graphs AI is trained on, and can sharpen how it represents your company over time.

Getting Schema onto your content comes down to three steps: (1) identify the types of content on your site, (2) match each one to an existing term in Schema’s vocabulary, and (3) apply the markup to each page.

On a site with thousands of pages, that sounds daunting. But breaking your site into patterns and categories isn’t as hard as it looks. Most sites already have structure, whether it was intentional or not.

Think about a higher ed website. It has events, programs, staff bios, news, and forms for applications and contact. If you run a content management system (CMS) like Drupal, those are already defined as content types with specific fields and media — and each can be matched to a term in Schema.org’s vocabulary.

We work primarily in Drupal, one of the best systems for managing structured content. Several modules can help get you started with Schema, each with a different approach:

Schema.org Metatag — this is the most widely used option, and the one we used on our own site. The module adds fields that let you tag your existing content types and pages with the appropriate Schema terms.
Schema.org Blueprints — the most comprehensive and complex option. Its best use case is when you’re building a new site. Start by identifying the Schema.org terms your content needs, such as “Event,” and then generate your content types and fields from that vocabulary. This is different from how most people apply it, which is to content that already exists.
Schema.org Metatag AI — the newest option. It relies on AI to identify and tag each page for you. It’s the least amount of manual work, but at the cost of accuracy and clean structure (with the usual caveats about AI errors). This is more one to keep an eye on, versus deploying on a live site, but it depends on your tolerance for risk.

In every case, the markup lands in your page code, where crawlers and AI can read it. Verify your work with a tool like the Schema Markup Validator or Google’s Rich Results Test, and note how your page content gets picked up in Google Search Console (which you should be using if you are not).

Based on what we know today, we don’t think everyone needs to rush out and add Schema. For many organizations the payoff is small and incremental. You should focus on well-written, relevant content first.

The best case is large, content-rich organizations like universities, cities and state agencies. Higher ed sites could tag course listings to reach prospects researching schools through AI. Government agencies could tag services and FAQs to be more visible to the residents and businesses they serve.

But don’t spend your effort where you’re already cited. AI and Google already know that these big organizations exist. What they may lack is the full picture of your content. Better-structured content is often how an organization gets seen, regardless of size.

You don’t need a big investment to start. Run an experiment on one part of your site that isn’t getting traction and see whether Schema helps. Bite off a little at a time and scale your effort to the results you see. If you’re planning a full redesign, budget some time for it — even 30 to 40 hours could be enough to make a difference.

Not sure where to start? Reach out to our team to talk through options — or subscribe to our monthly newsletter for more tips like these.

About the Author

About the author