Article Schema Markup: Add Accurate JSON-LD for Articles and Posts

Article schema is structured data that tells search engines the basic facts about a written piece: the headline, who wrote it, when it was published and updated, and which images belong to it. It is one of the most useful and least risky kinds of markup for blogs, news sites and guide libraries, because every fact in it is already on the page. The job is to state those facts accurately in a machine-readable form.

This guide is for site owners, editors and developers. You will learn which article type to use, which properties Google recommends, how to format dates, authors and images, and how to test and maintain the markup. It also sets realistic expectations. Google's article documentation describes structured data as a way to help it understand the page and possibly show richer results. It does not promise a result or a ranking change.

Before you begin

You need edit access to the template that renders your articles (or an SEO plugin that outputs the markup), a sample of published articles, and a clear idea of who the author is on each one. Plan about an hour for a template change and testing. If a plugin already outputs article markup, view the page source first and decide which source you will keep. Two competing blocks are confusing.

Have these facts ready for each article, and make sure they are visible on the page or in the page's normal metadata: the headline, the author name (and a profile or about page URL, if there is one), the first publication date, the last meaningful update date, and the main image.

What Google says about Article markup

From Google's article structured data documentation:

No properties are required. The page says to add the properties that apply to your content. Complete and accurate data matters more than a long list.
Supported types are Article, NewsArticle and BlogPosting, based on schema.org.
Recommended properties include author (with author.name and author.url), datePublished, dateModified, headline and image.
Dates use ISO 8601. Google recommends including timezone information; otherwise it defaults to Googlebot's timezone.
Authors: include every author shown on the page, each as a separate entry. Use Person for people and Organization for organizations. Put only the name in author.name; job titles belong in jobTitle, and honorifics in honorificPrefix or honorificSuffix. Add a url or sameAs value to help Google identify the author.
Images: image URLs must be crawlable and indexable. Provide multiple high-resolution images in 16x9, 4x3 and 1x1 aspect ratios. The minimum is 50K pixels, calculated as width multiplied by height. Images should represent the marked-up content, not a logo or a caption, and use a format Google Images supports.

Schema.org defines Article as a written piece such as a news story or report, with properties including headline, author, datePublished, dateModified and articleBody. Google's general structured data policy applies on top: markup must describe content that is visible to readers, must be a true representation of the page, and should be up to date.

Which type should you use?

BlogPosting for blog posts and guides written in a blog style.
NewsArticle for news reporting.
Article for other written pieces that are neither, such as a long-form report or a general article.

All three are supported by Google. The choice matters less than accuracy. If you are unsure, Article is a safe general choice. Do not label a product page or a category page as an article.

Step-by-step instructions

Step 1: Gather the facts for the template

Why: the markup should come from the same fields as the page, so it cannot drift. List the CMS fields or data sources for headline, author, publish date, modified date and image. If a field does not exist (for example no modified date), decide whether to add it or leave the property out. Google says to include properties that apply to your content, so leaving one out is better than guessing.

How to confirm: for each property you plan to output, you can point to the field that supplies it.

Step 2: Choose the headline

Why: headline is the title of the article. Use the title readers see on the page, usually the same as the H1. Do not stuff keywords or add promotional text. Keep it consistent with the on-page headline so there is no disagreement. How Google chooses the title shown in search is a separate matter, covered in Why Google Rewrites Title Tags and How to Write Titles That Stick.

How to confirm: the headline value equals the visible article heading.

Step 3: Mark up authors correctly

Why: authors are where most mistakes occur. Follow Google's rules:

One entry per author. Do not merge two names into one string.
@type of Person for a human, Organization for a company.
Only the name in name. Move "Dr.", "PhD" or "Senior Editor" into the proper properties.
A url or sameAs link to a page about the author helps Google identify them.

If you publish under a team name, as an organization-level byline, use an Organization author and link to a page that describes the team. Do not invent a person.

How to confirm: the author names and links in the markup match the byline on the page.

Step 4: Format the dates

Why: wrong date formats are a common validation error. Use ISO 8601, for example 2026-10-29T09:30:00+00:00, and include the timezone offset. Set datePublished to the first publication date and dateModified to the date of the last significant update. Do not change dateModified for trivial edits or to make an old article look new. The date shown on the page should agree with the markup.

How to confirm: the dates parse as ISO 8601 and match what the page displays.

Step 5: Prepare images

Why: Google's guidance asks for crawlable, indexable images that represent the article. Supply one or more images, ideally at 16x9, 4x3 and 1x1, each at least 50K pixels by Google's calculation (width times height). Use a format Google Images supports. Do not use your logo as the article image. For more on image handling and crawl access, see Image SEO: Alt Text, File Names and Formats That Help Search.

How to confirm: every image URL opens in a private window and is not blocked by robots.txt.

Step 6: Write the JSON-LD

Why: JSON-LD is Google's recommended format because it is easiest to implement and maintain. Generate it from your fields with a JSON encoder, so quotes in titles are escaped. Here is a complete example using only properties Google lists as recommended:

{
  "@context": "https://schema.org",
  "@type": "BlogPosting",
  "headline": "Choosing a Standing Desk for a Small Room",
  "image": [
    "https://example.com/images/standing-desk-1x1.jpg",
    "https://example.com/images/standing-desk-4x3.jpg",
    "https://example.com/images/standing-desk-16x9.jpg"
  ],
  "datePublished": "2026-10-01T08:00:00+00:00",
  "dateModified": "2026-10-20T14:30:00+00:00",
  "author": [
    {
      "@type": "Person",
      "name": "Alex Rivera",
      "url": "https://example.com/team/alex-rivera"
    }
  ]
}

Place this JSON inside a script element with the type application/ld+json.

Output the block in the server-delivered HTML, in the <head> or the <body>.

How to confirm: view the page source, copy the block into a JSON validator, and check that it parses.

Step 7: Test the page

Why: Google recommends the Rich Results Test to validate structured data during development, and Search Console reports after deployment because template or serving problems can break markup later. Test one article from each template. Fix any errors first, then review the warnings and decide whether the missing recommended properties apply to your content. For test-tool details, see JSON-LD Validation: Test and Debug Structured Data and Rich Results Test Errors: Fix the Common Errors and Warnings.

How to confirm: the test detects your article type with no errors, and the values match the page.

Step 8: Keep it accurate over time

Why: markup that no longer matches the page is misleading. When an article is updated, make sure dateModified changes only for real updates. When an author leaves or a byline changes, update the template. Review the markup after any redesign.

Crawvix's structured data check parses JSON-LD blocks on sampled pages and, for Article, NewsArticle and BlogPosting, expects at least a headline and datePublished. That is a minimum used by the check, not the full list of Google's recommended properties.

How to confirm: after each significant template change, you re-run a test on one article.

Final verification checklist

Each article page has one JSON-LD block of type Article, NewsArticle or BlogPosting.
headline matches the visible article heading.
Each author is a separate Person or Organization entry, with only the name in name.
datePublished and dateModified are ISO 8601 with a timezone, and match the page.
Image URLs are crawlable, relevant and not logos.
The JSON parses with no errors and appears in the delivered HTML.
No second plugin or theme block conflicts with yours.
A Rich Results Test run on a sample article shows no errors.

Common mistakes to avoid

Marking up content that is not on the page. Markup must describe what readers can see.
Combining several authors into one name. Use separate entries.
Putting job titles or credentials in author.name. Use jobTitle and the honorific properties.
Updating dateModified on every build. Change it for real updates only.
Using a logo or a tiny image. Choose a representative image with at least 50K pixels.
Leaving two markup sources active. Keep one.
Using Article markup on non-articles. Do not label product, category or landing pages as articles.

Frequently asked questions

Is Article schema required?

No. Google's page says there are no required properties. Add the ones that apply to your content.

Does Article schema improve rankings?

Google does not say it does. It uses structured data to understand the page and, where eligible, to show richer results. Do not expect ranking changes.

Should I use Article, NewsArticle or BlogPosting?

All three are supported by Google. Use BlogPosting for blog-style posts, NewsArticle for news and Article for other written pieces.

Do I need a publisher property?

Google's list of recommended properties for article markup does not include publisher. You may add other schema.org properties if they are accurate, but they are not part of the recommended set.

How do I mark up an article with several authors?

Give each author a separate entry in the author array. Use Person for people and Organization for organizations.

How should I mark up an article that was updated?

Keep datePublished as the original date and set dateModified to the last significant update, both in ISO 8601 with a timezone.

Official references

Next step: See how Crawvix checks a website and capture a baseline once scanning opens. This guide provides technical information, not a guarantee of rankings, traffic or index inclusion.