{"id":10028,"date":"2026-07-20T18:17:54","date_gmt":"2026-07-20T16:17:54","guid":{"rendered":"https:\/\/www.al-hostings.com\/datos-estructurados-seo-ia\/"},"modified":"2026-07-21T10:03:30","modified_gmt":"2026-07-21T08:03:30","slug":"structured-data-seo-ai","status":"publish","type":"post","link":"https:\/\/www.al-hostings.com\/en\/structured-data-seo-ai\/","title":{"rendered":"Structured data: how it helps SEO and AI"},"content":{"rendered":"
\n

<\/p>\n

\n Quick answer<\/span><\/p>\n

Structured data (schema) is code that tells machines what each thing on your page is: who you are, what service you offer, who signs an article. It helps SEO because it disambiguates your content and your entity for Google, and it helps AI indirectly: it feeds the knowledge graph engines use to decide who to trust. It’s not a button to “appear in AI”, but the identity layer that reinforces everything else.<\/p>\n<\/p><\/div>\n

<\/p>\n

Structured data has been surrounded by hype in both directions for years. One camp promises that with schema you magically appear in ChatGPT; the other says it’s no longer good for anything. Both are wrong, and understanding the nuance is exactly what gives you an edge.<\/p>\n

The data: an Ahrefs study of 1,885 pages (2026)<\/a> found that adding schema did not significantly increase citations in ChatGPT or Google’s AI Mode<\/strong>. At the same time, Google confirms that in its AI Overviews and AI Mode it does use structured data as context and a trust signal when generating answers. Contradictory? No: schema simply doesn’t work the way most people think.<\/p>\n

This guide explains what structured data is, how it really helps SEO and AI, and which types deserve your time in 2026.<\/p>\n

<\/p>\n

What structured data is<\/h2>\n

Structured data is standardized code (usually in JSON-LD format, following the Schema.org vocabulary) that labels your page’s content so machines understand it without guessing. Where a person reads “ALHOSTINGS, web design in Barcelona, from \u20ac599” and gets it instantly, a search engine sees ambiguous text. Schema translates it: this is an Organization, this is its service, this is its price, this is its location<\/em>.<\/p>\n

That translation is what lets Google build its knowledge graph (the map of entities and relationships behind brand panels and AI answers) with your information inside it and free of errors.<\/p>\n

<\/p>\n

\n

How it really helps: 5 real functions<\/h2>\n

<\/p>\n

\n
1<\/div>\n
\n

It disambiguates your entity (who you are)<\/h3>\n

This is the most powerful function in 2026 and the most underused. Organization<\/code> or Person<\/code> schema, with the sameAs<\/code> property pointing to your LinkedIn, Wikipedia or Wikidata, tells Google exactly who you are and connects it to already-known records. A resolved entity gets more trust when AI decides who to cite.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

<\/p>\n

\n
2<\/div>\n
\n

It declares what you know (your topical authority)<\/h3>\n

The knowsAbout<\/code> property in your Organization schema lists the topics you have real expertise in. It’s an explicit way of telling Google “I’m a reliable source on WordPress, SEO and web maintenance”, which is exactly what it assesses before citing you in an answer on those topics.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

<\/p>\n

\n
3<\/div>\n
\n

It gives context to Google’s AI<\/h3>\n

In AI Overviews and AI Mode, Google uses structured data as one more signal (alongside content, authority and relevance) to verify claims and assess credibility. It’s not a switch that turns you on in AI, but it does remove the ambiguity that would otherwise lower the confidence with which they select you.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

<\/p>\n

\n
4<\/div>\n
\n

It still produces rich results (the ones that remain)<\/h3>\n

Although Google removed FAQ rich results in May 2026, many others are still alive: review stars, product prices, recipes, events, breadcrumbs. A well-placed Product<\/code>, Review<\/code> or BreadcrumbList<\/code> improves how you look in the SERP and your CTR.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

<\/p>\n

\n
5<\/div>\n
\n

It helps Google classify your content correctly<\/h3>\n

Marking an article as Article<\/code> with its author and dates, or a service page as Service<\/code>, helps the search engine understand what each page is and how it fits into your site. It’s technical hygiene that’s rarely seen, but adds clarity.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

\n

The key idea: schema amplifies other quality signals; it doesn’t replace them<\/strong>. If the content underneath is weak, marking it up won’t improve it. You’re decorating an empty room.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

<\/p>\n

What’s said versus what the data shows<\/h2>\n

Two common exaggerations, and the realistic point in between.<\/p>\n

\n\n\n\n\n\n\n\n\n
What’s said<\/th>\nWhat the data shows<\/th>\n<\/tr>\n<\/thead>\n
“Schema gets you into ChatGPT”<\/td>\nMost third-party AI engines strip the JSON-LD on retrieval and read the visible HTML. Schema isn’t what gets you cited directly.<\/td>\n<\/tr>\n
“Schema is no longer good for anything”<\/td>\nIt is: it disambiguates your entity, feeds the knowledge graph, strengthens organic and produces rich results. Google uses it as context in its AI Overviews.<\/td>\n<\/tr>\n
“Adding schema lifts your AI citations by 44%”<\/td>\nThe Ahrefs study (1,885 pages) found no significant increase in ChatGPT or AI Mode from adding schema. Distrust round figures with no study behind them.<\/td>\n<\/tr>\n
“The more schema, the better”<\/td>\nMarking up what isn’t visible on the page is considered structured-data spam. The rule is parity: schema must always mirror the visible content.<\/td>\n<\/tr>\n<\/tbody>\n<\/table><\/div>\n

Sources: Ahrefs study on schema and AI citations (2026) and Google Search Central documentation (2026).<\/p>\n

<\/p>\n

Which structured data to prioritize in 2026<\/h2>\n

Not all of it is worth the same. This is the order I’d follow on a freelancer or SME service website:<\/p>\n

\n
1<\/div>\n
\n

Organization and Person (with sameAs and knowsAbout)<\/h3>\n

The foundation for everything. Define your brand and the person behind it, link them to your real profiles (LinkedIn, social) and declare your topics. It’s the highest-impact implementation and the most neglected.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

\n
2<\/div>\n
\n

LocalBusiness (if you serve an area)<\/h3>\n

With complete NAP (name, address, phone) and opening hours. It’s what lets voice assistants and Google’s local block retrieve your contact details correctly. Essential for local SEO.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

\n
3<\/div>\n
\n

Article (on every blog post)<\/h3>\n

With author, publish date and modified date. Connect the author to their Person schema to reinforce E-E-A-T and help AI know who signs it and when it was updated.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

\n
4<\/div>\n
\n

Service and, if you sell products, Product<\/h3>\n

Mark each service with its provider and service area. In e-commerce, Product<\/code> with price and reviews still produces valuable rich results.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

\n
5<\/div>\n
\n

FAQPage and BreadcrumbList (supporting)<\/h3>\n

They no longer produce FAQ rich results, but they still clarify structure for machines. Useful as reinforcement, never as the main lever. We covered this in how to structure FAQs for SEO and AI answers<\/a>.<\/p>\n<\/p><\/div>\n<\/p><\/div>\n

<\/p>\n

\n

Checklist to implement schema without headaches<\/h2>\n

Six rules so your structured data adds up rather than subtracts.<\/p>\n