Skip to main content
Schema.org structured data — usually written as JSON-LD — is a machine-readable label you attach to your own content: “this is an Organization,” “this is a Product,” “this is an Article by this author, published on this date.” It doesn’t change what a visitor sees on the page. It changes how confidently a crawler can parse what your page actually is.

​
Why it matters for AI engines

Treat schema as hygiene, not a growth lever. It helps a crawler correctly identify your business, your products, and your content without guessing from prose — Bing has said explicitly that it uses schema to help its systems understand a page. But no major AI platform has confirmed that adding schema increases how often you’re cited, and controlled testing has found no measurable citation uplift from adding it. The value is confidence and eligibility, not a ranking boost. The one rule that actually matters: your markup must match what’s visibly on the page. Structured data that claims something your visible content doesn’t back up is worse than no markup at all — it damages trust with the platforms reading it.
This is also why Surfais never treats missing schema as a critical failure on your AI Readiness page — it’s shown as an opportunity to tidy up, not a red flag holding back your visibility.

​
What Surfais looks for

When you run a readiness check, Surfais reads the structured data actually present on your homepage — across the common formats (JSON-LD, microdata, and RDFa) — and lists every schema.org type it finds. There’s no fixed checklist you’re being marked against; the page simply reports back what’s genuinely there; an empty result means no structured data was detected at all. In practice, the types worth having depend on what kind of business you are:

Organization

Your company identity — name, logo, and links to your other verified profiles. The foundation almost every site should carry.

Product

For e-commerce and retail: what you sell, with pricing and availability, paired with an Offer.

LocalBusiness

For any business with a physical presence or service area: address, opening hours, and contact details.

Article

For published content and blogs: author, publish date, and headline — helps establish who wrote what, and when.
Software businesses typically look to a SoftwareApplication type instead of Product; publishers and recipe-led content often add Article or Recipe on top of the basics. Match the type to what the page genuinely is, rather than reaching for every type available.
FAQ-specific markup is no longer worth adding — major search engines have retired the rich-result treatment it used to unlock, and no AI platform ties it to citation eligibility. Visible question-and-answer content on the page is still genuinely useful; it’s the dedicated markup for it that’s no longer worth your time.

​
Implementing it

1

Pick the type that matches the page

Don’t force a Product type onto a services page, or an Organization type onto every individual product. Each page should carry the type that actually describes it.
2

Write it as JSON-LD

This is the format search and AI platforms consistently prefer — a single <script type="application/ld+json"> block in the page’s HTML, kept separate from your visible layout. It’s also the easiest format to maintain, since it doesn’t need to be woven into your markup.
3

Keep it in sync with the visible page

If a product’s price or availability changes, the schema needs to change with it. Stale or contradictory markup is a bigger problem than no markup.
4

Validate before you ship it

Run it through a schema validator to catch malformed JSON or missing required properties before it goes live — a broken script tag means nothing gets read at all.

​
Verifying it worked

Once you’ve added or updated schema, head back to your AI Readiness page and click Run Check. It reads your homepage fresh and lists the schema types it finds within moments — if your new markup shows up in the results, it’s valid and detected. If it doesn’t appear, double-check the JSON is well-formed and that the script tag made it into the page’s actual HTML rather than only a client-rendered version of it (see AI crawler access for why that distinction matters). A full audit goes further than the readiness check — it looks at specific properties within your markup, not just which types are present, and flags gaps like an Organization type missing its logo or verified profile links.

The AI Readiness page

Where your detected schema types are shown, alongside the rest of your readiness check.

AI crawler access

The other half of on-site technical readiness — making sure crawlers can reach your pages at all.

What audits check

The deeper property-level review behind your Readiness Score.

AIS Score vs Readiness Score

How this technical groundwork relates to your measured visibility.