How Schema Markup Boosts AI Visibility by 55% (Case Study)

What’s your website’s AI visibility score?

If you’re wondering how to get seen and cited by LLMs like ChatGPT and Perplexity, you’re in the right place.

Lacrosse Marketing Co., a boutique agency for growing sports brands, increased their AI Visibility Score by 55% in less than one business day (not citations, more on that below).

The biggest driver? Structured data (also known as schema markup).

By improving the site’s schema across 10 key pages, we helped AI systems better understand the content, resulting in stronger discoverability and the first tracked AI referral visit.

Don’t worry, it’s not as technical as it sounds. You can optimize your website’s schema too.

Stick with us and we’ll show you exactly what we did (it’s easy).

The Premise - Schema Influences AI Website Visits

The first place we explored is traffic analytics in HubSpot. As of March 2, 2026, HubSpot reported zero AI referrals sending traffic to the Lacrosse Marketing Co. As a leader in their niche, they should have a place at the table. So why are the LLMs not citing them? Let’s take a look at why.

After running their AI visibility baseline in SuperSchema with a simple URL scan, we could easily see the problem.

The site scored a 60 out of 100 (D grade) due to missing schema markup on most pages. And none of the pages contained alternate files (essential context for LLMs).

So while the site looks fantastic to human visitors, AI tools had very little structure to help them scan, understand, and confidently cite the content.

Which means when someone asks ChatGPT something like:

“What are the best marketing agencies for a lacrosse company?”

This site, despite being extremely relevant, wasn’t showing up.

Let’s fix that.

The Process - Optimizing Your Website for AI with Schema Markup

Time to optimize this website for AEO.

HubSpot has a built-in structured data feature in blog settings that will apply BlogPosting schema to all blog posts, which was toggled on for Lacrosse Marketing Co. We took one of the pages with preexisting HubSpot schema and ran it through SuperSchema’s free schema markup grader.

The result?

It got a D grade. Woof.

While the schema was technically schema.org compliant, it wasn’t complete enough to provide the context an AI bot would need to understand the content of the post, author, recency (when the post was updated), and more.

Before generating the new and improved schema, we turned off the Hubspot default by going to settings → Blog → and toggle the structured data off.

Once HubSpot structured data setting was toggled off, we created new schema in SuperSchema by pasting your URL and clicking a few buttons. In just a few seconds, out popped brand new optimized structured data, complete with alternate files. Then we simply pressed the cute “push to hubspot” button and the schema magically appeared live on the page (no update or publish step needed).

Then we ran the page through the grader again.

WOW.

The score jumped from 66 (D) to 88 (B+) in less than five minutes.

Pardon me while I pick my jaw up off the floor.

Quick HubSpot Schema Gotcha:

If you’re running a HubSpot blog, there’s one weird quirk worth knowing.

Blog listing pages can’t receive pushed schema like normal pages.

Instead, you need to manually paste the schema into:

Settings → Blog → Listing Page Header

It takes about two seconds, but it’s easy to miss if you don’t know where to look.

The Proof - Schema Boosts Website Grade and Earns First AI Referral

The rest of Lacrosse Marketing’s pages had no preexisting schema, so we generated schema markup and alternate files for each page using SuperSchema and pushed it directly into HubSpot.

After optimizing roughly 10 website pages across the site, we ran their AI Visibility Score again.

And oh. my. goodness: the results were dance-party inducing.

Lacrossemarketing.co jumped 33 points from 60 (D grade) to 93 (A grade). Woohoo!

Yep, you read that right. Just by adding schema and alternate markdown files.

Once the schema went live, we ran a simple test.

We asked ChatGPT a question relevant to their services.

ChatGPT cited lacrossemarketing.co.

To confirm that AI referral tracking was working as expected in their HubSpot account, we clicked the citation link and checked HubSpot.

Traffic reports can take a few hours to refresh their data so we had our virtual dance party and checked back an hour or two later to discover that sure enough, the AI Referral visit appeared in HubSpot’s traffic sources.

Success.

And while schema alone doesn’t guarantee AI citations, structured data dramatically improves how AI systems interpret and trust web content.

30 Days of (Somewhat Surprising) AI Visibility and Traffic Results

While the schema improvements and test AI referral were nearly instantaneous, website optimization is iterative. In the first 30 days since these initial AEO improvements, we've seen AI bots beginning to crawl the pages on the Lacrosse Marketing site.  Some of these bots are gathering training data, others are saving the info to serve up in search queries.

SuperSchema's bot tracking shows 37 AI crawlers visiting the site within 30 days of optimization, a strong early signal

And while over 30 bot crawls is an exciting leading indicator of AEO optimization, we haven't seen any AI referral traffic yet in HubSpot. It's possible that Lacrosse Marketing is such a niche that it simply doesn't have a high traffic volume for LLM search.  Or, potentially, their main audience hasn't fully adopted tools like ChatGPT for search purposes.  Just like traditional SEO, real impact happens over time, and this story is just beginning.

As mentioned above, schema doesn't guarantee AI search mentions, but not having any is a risk that businesses can't afford to take in our shifting marketing landscape.

What this Schema Case Study Means for Your AI Visibility

We’re already seeing bot crawls that can lead to future AI referral traffic.

Structured data gives AI systems the context they need to understand your content and cite your pages when answering questions.

Which means your website can start showing up inside AI answers, not just traditional search results.

For Lacrosse Marketing, this could mean attracting more of their dream clients simply because their expertise is now easier for AI tools to discover.

If optimizing your site for AI discovery only takes a few minutes per page, is it worth it?

Just imagine the ROI.

Now I’m curious:

What grade does your website get? Run your free scan and find out (takes 30 seconds).

How to Get Your Website Cited by AI Tools Like ChatGPT

Here’s the TL:DR:

AI systems like ChatGPT, Perplexity, and Claude discover and cite websites when the content is easy for machines to interpret and trust. One of the most effective ways to improve AI discoverability is by implementing structured data.

Structured data signals like schema markup help AI interpret and cite web content.

Here are three steps that help websites appear in AI-generated answers:

1. Add structured data (schema markup).
Schema markup helps AI systems understand what your content represents including articles, companies, services, and authors. Try a tool like SuperSchema to get started and push schema across your full site.

2. Provide structured alternate content formats.
Alternate files like markdown versions of pages make it easier for AI systems to parse and summarize your content.

3. Maintain clear authorship and content recency.
Signals like author attribution and last-updated timestamps help AI systems evaluate credibility and relevance.

When these signals are present, AI tools are more likely to understand, trust, and cite your content when answering user questions.

Want to know how AI interprets your website? Get your AI Visibility Grade here.