Most guides to the types of schema markup hand you an alphabetical list of 30-plus schema.org entities and leave you to guess which ones matter. That list is worse than useless, because it treats a type that triggers a star rating in the search results as equal to one that does nothing visible at all — and it never tells you the two that Google quietly killed. The real question isn’t “which types exist.” It’s “which types change what Google shows for my page, and which are pure overhead.” This guide sorts every schema type worth your attention into that framework, then shows you exactly how to nest them the way a crawler reads them.
Schema markup makes you eligible — it does not rank you
Before any list of types makes sense, you have to fix the mechanism in your head. Schema markup is structured data — a machine-readable JSON-LD block in your page’s HTML that describes what the content is: a product, a recipe, an organization, an event. Google reads it to understand the entity and, for certain types, to render a richer result: a price, a star rating, a breadcrumb trail, a video thumbnail. That richer result is called a rich result, and eligibility for one is the entire prize.
Here is the part almost every article gets wrong: schema is not a ranking factor. Adding valid markup does not push you up the results. It makes you eligible for enhanced display, which lifts click-through rate on the position you already hold — and higher CTR can feed rankings indirectly over time. So the correct filter for every schema type below is blunt: does this type produce something visible, does it build the entity graph Google uses to understand your site, or does it do neither? If it does neither, it is a waste of your implementation time.
Tier one: the types that earn a visible rich result
These are the schema types with the highest return, because Google renders them directly in the search results and they change how your listing looks. If your content matches one of these categories, implement it first:
- Product — price, availability, and (with valid review data) star ratings on ecommerce and comparison pages. The single highest-impact type for anyone selling something.
- Recipe — cook time, calories, ratings, and a photo carousel. Near-mandatory in the food niche; a recipe page without it is invisible in the recipe carousel.
- Event — dates, venue, and ticket status, eligible for the events experience in Search and Maps.
- VideoObject — thumbnail, duration, and key moments; the way to surface video in the main results and the Video tab.
- JobPosting — salary, location, and posting date, feeding the Google Jobs experience.
- LocalBusiness — hours, address, and phone, reinforcing your knowledge panel and local pack presence.
- BreadcrumbList — replaces the raw URL in your listing with a clean navigational trail. Trivial to add, universally useful, and one of the few types that helps almost any site.
The trade-off with tier one is eligibility, not guarantee. Valid Product markup makes you eligible for a price and rating snippet; Google still decides per-query whether to show it. You do the work; Google keeps the discretion. That is not a reason to skip it — it is a reason to get the required properties exactly right so you are never disqualified on a technicality.
Tier two: the types that work invisibly but matter
The second tier of schema types never produces a snippet you can screenshot, which is exactly why people skip them and why skipping them is a mistake. These types build the entity graph — Google’s model of who you are, what your site is, and how your pages relate. That model feeds knowledge panels, sitelinks, and increasingly the AI Overviews and chatbots that cite sources.
- Organization — your name, logo, and official social and contact profiles. This is the backbone of your knowledge panel and the type most sites should implement sitewide first.
- WebSite — enables the sitelinks search box and tells Google the canonical name of your site as a whole.
- Article / BlogPosting / NewsArticle — author, publish date, and headline, establishing authorship and freshness signals that matter more every core update.
- Person — for author entities and the individuals behind your content, increasingly important as Google leans on demonstrable expertise.
None of these light up your listing tomorrow. All of them compound. In a search landscape where Google and AI answer engines are trying to decide whether your brand is a real, trustworthy entity, machine-readable Organization and Person data is how you tell them directly instead of hoping they infer it.
Tier three: the types that are effectively dead
This is the tier no alphabetical list will tell you about, and it is the most valuable information here. Two once-popular schema types no longer earn rich results for the vast majority of sites:
- FAQPage — Google restricted FAQ rich results in 2023 to well-known, authoritative government and health sites. For everyone else, marking up FAQs no longer produces the expandable accordion in search. You can still use the markup for structural clarity, but do not implement it expecting the visible payoff.
- HowTo — deprecated across the board. The step-by-step rich result is gone. Implementing HowTo schema today is pure effort with zero display return.
If your schema strategy from two years ago leaned on FAQ and HowTo markup for visibility, that visibility is gone and you likely never got the memo — the markup validates fine, it just does nothing. This is precisely the kind of silent decay a real site audit surfaces. SEO Rocket’s crawler-based site audit flags structured-data issues across your whole site, so you find the schema that quietly stopped working before you spend another quarter maintaining it.
Review and rating markup: allowed, but fenced
Star ratings are the most coveted rich result and the most abused, so Google fenced them. Review and AggregateRating markup is supported, but only when it is attached to a valid host type — Product, Recipe, LocalBusiness, and a short list of others — and only when the reviews are genuinely on your page and about the item itself. “Self-serving” reviews (a business rating its own services) are explicitly ineligible. Bolting AggregateRating onto arbitrary pages to farm stars is exactly the pattern manual actions target. Use it where it is legitimately earned; never fabricate the numbers.
A worked example: nesting types the way Google reads them
The real skill isn’t picking one type — it’s nesting several so the crawler reads a coherent entity. A blog post should not carry a lone Article block. It should express Article, its Person author, and the publishing Organization as one connected graph. Here is the shape that matters, trimmed to the load-bearing properties:
{
"@context": "https://schema.org",
"@type": "BlogPosting",
"headline": "The Types of Schema Markup That Actually Earn Results",
"datePublished": "2026-07-28",
"author": {
"@type": "Person",
"name": "Your Author",
"url": "https://yoursite.com/about"
},
"publisher": {
"@type": "Organization",
"name": "Your Site",
"logo": {
"@type": "ImageObject",
"url": "https://yoursite.com/logo.png"
}
}
}Notice three things. First, author nests a full Person object with a url that points to an author page — that URL is how Google connects the byline to a real entity. Second, publisher nests the Organization and its logo, which Google needs for the article rich result. Third, every required property is present; a missing logo or datePublished can silently disqualify the whole block. Getting the nesting right is the difference between markup that validates and markup that Google actually uses.
How many types should one page carry?
More is not better. The rule is one primary type that describes the page’s main entity, plus the supporting types that legitimately apply. A product page carries Product as primary, with BreadcrumbList and Organization supporting. A blog post carries BlogPosting as primary, with Person and Organization nested in. Stacking every type you can think of onto a single page doesn’t multiply your rich-result chances — it creates conflicting signals and validation errors, and it can look like an attempt to game eligibility.
Match the type to what the page genuinely is, not to what you wish it ranked for. A thin FAQ page dressed in Product markup will not earn product snippets; it will earn a mismatch flag. The decision rule is one sentence: mark up what the page truly is, at the tier that actually pays off.
Choosing types for your specific site
Collapse the whole decision into your site archetype:
- Ecommerce — Product and BreadcrumbList first, Organization sitewide, Review only where reviews are real.
- Publisher or blog — Article/BlogPosting with nested Person and Organization, plus WebSite for the search box.
- Local service business — LocalBusiness with complete hours and address, plus Organization.
- SaaS or B2B — Organization and WebSite are the whole game; there is rarely a tier-one rich result to chase, so don’t manufacture one.
- Events, recipes, jobs, video — the matching tier-one type is close to mandatory in these niches, not optional.
Validate, monitor, then stop touching it
Schema is not a set-and-forget-forever asset, but it is close. Validate every template with Google’s Rich Results Test and the schema.org validator before you ship, then monitor the structured-data reports in Search Console for errors after Google recrawls. The most common failures are boring and fatal: a missing required property, an invalid date format, a price without a currency, a logo URL that 404s. Fix those, confirm eligibility, and then leave it alone — endlessly tinkering with working markup is how you introduce the error that breaks it.
At scale, hand-checking every template is where this falls apart. This is the unglamorous, mechanical work SEO Rocket is built to carry: the site audit catches broken and outdated structured data across every URL, rank tracking tells you whether the rich results you earned are lifting click-through, and the AI-visibility tracking shows whether the entity data you fed Google is actually getting you cited in AI answers. It is the same playbook proven across 1,000,000+ ranking pages — implement the types of schema markup that pay off, kill the ones that don’t, and measure instead of guessing.
Frequently asked questions
Is schema markup a Google ranking factor?
No. Schema markup does not directly raise your position. It makes you eligible for rich results — richer, more clickable listings — which lifts click-through rate and can feed rankings indirectly over time. Treat it as a display and eligibility lever, not a ranking hack.
Which schema type should I add first?
Organization sitewide, then BreadcrumbList, then the single tier-one type that matches your content (Product, Recipe, Event, LocalBusiness, and so on). Those three cover the highest-return cases with the least effort.
Does FAQ schema still work?
For almost everyone, no. Google restricted FAQ rich results in 2023 to authoritative government and health sites. The markup still validates but no longer produces the expandable accordion in search for typical sites, so don’t implement it expecting visibility.
Can too much schema markup hurt me?
Yes, indirectly. Stacking mismatched or fabricated types (self-serving review stars, Product markup on non-product pages) creates conflicting signals, validation errors, and in the worst case a manual action. Mark up what the page genuinely is, and nothing it isn’t.
