Article Schema for Blog Posts That Machines Can Parse
Contents
Article schema is a small block of structured data, usually written in JSON-LD, that tells Google and AI systems the headline, author, and publish date of a blog post instead of leaving them to guess from the page's layout. Getting it right is one of the few technical steps a blog can take that both classic search and AI answer engines read the same way.
That matters more than it looks. Google treats Article as one of only a handful of structured-data types with a dedicated eligibility guide, which signals how much weight it carries for rich results. JSON-LD itself, the format Article schema is written in, has grown fast: it now appears on 41% of crawled pages, up from 34% in 2022 (HTTP Archive Web Almanac, 2024 chapter). Yet content-specific markup like BlogPosting has not kept pace with that broader adoption curve: BlogPosting schema appears on just 1.40% of crawled pages, and Article schema on 0.18%, both far below foundational types like WebSite (12.73%) or Organization (7.16%) (HTTP Archive Web Almanac, 2024 chapter). That gap between "widely recommended" and "rarely implemented correctly" is exactly what this article closes, with a decision table, a complete template, and the AI-crawler mechanics most guides never touch.
What Is Article Schema, and Why Does It Matter for a Blog Post?
Article schema is a JSON-LD block that tells Google and AI systems what a blog post is, who wrote it, and when it was published or updated, instead of leaving that information to be inferred from the page's visible text. A human reader can look at a byline and a date stamp and understand both instantly. A crawler cannot make that same inference reliably from raw HTML and CSS, especially when a theme renders the date as an icon plus a string with no semantic markup around it.
This is separate from general on-page SEO in one specific way: it does not change what a page ranks for, it changes what a system can confirm about a page it has already decided to consider. That distinction is why Article schema sits inside the broader discipline of schema markup for AI, which covers four other types (Organization, FAQPage, HowTo, Product/Review) that each label a different kind of page. Article is the one type that applies to nearly every blog post a site publishes, which is exactly why getting the implementation right, rather than copy-pasting a half-finished snippet, is worth a dedicated pass.
Article vs. BlogPosting vs. NewsArticle: Which Type Should You Actually Use?
Use BlogPosting for standard, recurring blog content, Article only when nothing more specific fits, and NewsArticle exclusively for time-sensitive journalism, because Google's own guidelines say to pick the most specific type available. Most business blogs, SaaS content sites, and personal blogs should default to BlogPosting rather than the generic Article type, since BlogPosting is a more specific subtype built for exactly this content shape.
The three types share the same core properties (headline, author, datePublished, and so on), so switching between them later is a low-effort change. What differs is the signal each name sends about the content itself: a schema-aware system reads NewsArticle as "this content is time-sensitive and journalistic," BlogPosting as "this is a recurring editorial post," and plain Article as "this is written content that doesn't fit a more specific subtype." The table below resolves the decision in one pass.
| Type | When to use it | What it signals |
|---|---|---|
| BlogPosting | Recurring blog content on a company site, personal blog, or content marketing hub | "This is a regular editorial post, part of an ongoing publication" |
| Article | Written content that doesn't clearly fit BlogPosting or NewsArticle (an evergreen resource page, a long-form explainer outside a blog structure) | "This is written content, use the most generic label available" |
| NewsArticle | Time-sensitive journalism: breaking news, event coverage, press-style reporting | "This content is time-bound and journalistic, treat it as news" |
If a site runs a single blog and nothing resembling a newsroom, BlogPosting is the correct default for essentially every post, and there's rarely a reason to reach for plain Article instead.
The Article Schema Properties Google and AI Systems Actually Check
Google's own structured-data documentation states there are no strictly required properties for Article: the guidance is to add whichever properties genuinely apply to the content. In practice, Google names five properties in its own recommended list, headline, image, author, datePublished, and dateModified, and two more, publisher and mainEntityOfPage, are worth adding on top even though Google's own table doesn't name them, because both AI crawlers and other schema consumers still read them for authorship and canonical clarity. That same consistent Organization object, name and logo used identically across every post, also feeds the broader entity signals covered in how to get a Google Knowledge Panel.
| Property | Required or recommended | What it does |
|---|---|---|
| headline | Recommended (named in Google's own guidance) | The article's title, must match the visible on-page headline exactly |
| image | Recommended (named in Google's own guidance) | At least one representative image, used for rich results and social previews |
| author | Recommended (named in Google's own guidance) | A Person (or Organization) object identifying who wrote the piece |
| datePublished | Recommended (named in Google's own guidance) | ISO 8601 date the article first went live |
| dateModified | Recommended (named in Google's own guidance) | ISO 8601 date of the most recent substantive update |
| publisher | Recommended (schema.org best practice, not named in Google's own list) | An Organization object with name and logo, identifying who runs the site |
| mainEntityOfPage | Recommended (schema.org best practice, not named in Google's own list) | The canonical URL, confirming this markup describes this specific page |
Each row here is self-contained enough to answer a single question on its own: what does headline do, what does author require, what does dateModified track. That is deliberate. A properties table structured this way is exactly the kind of content an AI system can lift as a complete, accurate answer without needing the surrounding paragraph for context, the same self-contained-answer principle behind BLUF writing.
A Complete JSON-LD Template You Can Copy Into a Blog Post
Below is one complete, valid BlogPosting template with every recommended property filled in, ready to adapt with real values. Replace every bracketed placeholder, keep the structure otherwise unchanged, and place the whole block inside a <script type="application/ld+json"> tag in the page's <head> or just before the closing </body> tag; either location works, as long as it renders in the page's initial HTML rather than being injected only after client-side JavaScript runs.
{
"@context": "https://schema.org",
"@type": "BlogPosting",
"headline": "[Exact Page Headline, Matching the Visible H1]",
"image": "[https://yourdomain.com/path/to/featured-image.jpg]",
"author": {
"@type": "Person",
"name": "[Author Full Name]",
"url": "[https://yourdomain.com/authors/author-slug]"
},
"datePublished": "[YYYY-MM-DD]",
"dateModified": "[YYYY-MM-DD]",
"publisher": {
"@type": "Organization",
"name": "[Your Company Name]",
"logo": {
"@type": "ImageObject",
"url": "[https://yourdomain.com/logo.png]"
}
},
"mainEntityOfPage": {
"@type": "WebPage",
"@id": "[https://yourdomain.com/blog/post-slug]"
}
}
Two details are worth double-checking after pasting this in: the headline value must match the on-page title character for character, and dateModified needs to change every time the post's substantive content changes, not just its formatting.
This article, Blue could have written it for you: content optimized for Google + AI, without you writing a single word.
Try mentionLABHow Do AI Crawlers Like GPTBot and PerplexityBot Read Article Schema Differently Than Googlebot?
Googlebot primarily uses Article schema to decide rich-result eligibility, while AI crawlers such as GPTBot, PerplexityBot, and ClaudeBot lean on the same fields, author, datePublished, dateModified, and headline, to establish freshness and authorship when deciding whether a page is citable at all. That's a different job than ranking a page in ten blue links, and it changes which fields carry the most weight.
Googlebot's rich-result eligibility check is largely binary: the required properties are present and valid, or a page doesn't qualify for the enhanced search appearance, full stop. An AI crawler assembling an answer is doing something closer to source triage across dozens of candidate pages at once, and a clear, machine-readable author name plus a recent dateModified value are cheap, verifiable signals that a page is maintained and attributable, rather than an abandoned or anonymous scrape. This is the same logic covered in more depth in which AI crawlers to allow: a crawler first has to be allowed in, and once it's in, Article schema is one of the fastest ways it confirms what it's looking at.
The stakes of getting this right have grown. Organic click-through fell 61%, from 1.76% to 0.61%, on queries that triggered an AI Overview, while pages cited directly inside the AI Overview saw 35% higher organic click-through than pages that ranked organically but weren't cited (independent CTR research across 25.1 million search impressions, as reported by Search Engine Land, 2026). Separately, Pew Research Center found that users click a traditional organic result in just 8% of searches where an AI summary appears, versus 15% when no summary appears, and 26% abandon the search session entirely after a page with a summary, versus 16% without one (Pew Research Center, 2025 fieldwork). Article schema alone won't secure a citation, but it removes one of the easiest reasons a crawler has to skip a page in favor of a competitor with cleaner markup, and it's just one input into what makes a page get cited in AI Overviews. For the levers beyond schema that actually drive citation, see how to get cited by AI.
Common Article Schema Mistakes That Quietly Break Rich Results
The most common Article schema mistakes come straight from Google's own author-markup guidance, restated here as a plain checklist rather than scattered across a documentation page. Each one is easy to introduce by accident and easy to miss during a routine content update.
- Merging multiple authors into one string. When a post has more than one author,
authorshould be an array of Person objects, not a single string like "Jane Doe and John Smith" stuffed into one name field. - Using Organization type for a person. If a named individual wrote the post,
authorshould be"@type": "Person", not"Organization", even if the byline also credits the company. - Adding Article schema to non-article pages. A pricing page, a category listing, or a landing page isn't an article, and forcing Article markup onto it can create a mismatch between what the schema claims and what the page actually is.
- Letting the visible headline drift from the markup's headline field. Editors often tweak an on-page title after publishing without updating the schema's
headlineproperty, creating a silent mismatch that can trigger a rich-result error. - Forgetting to bump dateModified after an update. A page can go through a substantial content revision while
dateModifiedstill shows the original publish date, which undercuts exactly the freshness signal Article schema exists to provide.
How to Validate Article Schema Before You Publish
Validate Article schema by running the live URL or raw code through Google's Rich Results Test, confirming the page is recognized as Article- or BlogPosting-eligible with no critical errors, and rechecking after any future content update. The check takes under a minute and catches the majority of implementation mistakes before they reach a live page.
- Paste the URL, or the raw JSON-LD code, into Google's Rich Results Test. This tool renders the page and reports exactly which structured-data types it detected.
- Confirm the page shows as Article or BlogPosting eligible with zero critical errors. Google does not treat any Article property as strictly required, but critical errors on the fields it names in its own guidance (headline, image, author, datePublished, dateModified) still need fixing before the markup is trusted.
- Recheck after any substantive content update. Since
dateModifiedandheadlineare two of the fields most likely to drift out of sync with the live page, a quick revalidation after edits catches problems before they accumulate across dozens of posts.
Frequently Asked Questions
Does Article schema improve search rankings?
No, Article schema does not directly improve search rankings. It makes a blog post eligible for certain rich-result treatments and gives both Googlebot and AI crawlers a clearer, faster way to confirm headline, author, and date information, but ranking position is still driven primarily by content quality, relevance, and the site's overall authority. Think of it as removing friction for machines that are already evaluating the page, not as a ranking signal in its own right.
What is an example of Article schema for a blog post?
A complete example includes the headline, image, author (as a Person object), datePublished, dateModified, publisher (as an Organization object with a logo), and mainEntityOfPage properties, all wrapped inside a "@type": "BlogPosting" JSON-LD block. The full copy-paste template earlier in this article shows every one of these fields filled in with placeholder values ready to swap for real data.
Can Article schema be combined with FAQ schema on the same page?
Yes, a single blog post can carry both BlogPosting schema and FAQPage schema at the same time, typically bundled together inside one @graph array in the same JSON-LD block. This pairing is common on explainer-style posts that end with a genuine question-and-answer section, since it lets one page describe both what the article is and what specific questions it answers.
Can AI tools generate Article schema automatically?
Yes, AI-assisted content tools can generate a complete Article or BlogPosting JSON-LD block automatically by pulling the headline, author, and publish and update dates directly from the post itself, which removes the manual copy-paste step most publishers currently rely on. This is one of the smaller, more mechanical steps inside a broader GEO content workflow, alongside the larger editorial decisions, like sourcing and structure, covered in LLM SEO, that determine whether a page gets cited in the first place.
Blue handles your SEO and your GEO. On autopilot.
You approve, she produces content optimized for Google + AI.
Join the Lab · 5-day trial