Article schema: required fields and common mistakes

• ✓headline
• ✓image

• +datePublished
• +dateModified
• +author
• +publisher
• +mainEntityOfPage

These are hints, not blockers. Our validator marks them non-critical for exactly that reason: a missing optional field does not make the item invalid, and treating it as an error would make every report noisy and therefore ignorable.

• A headline that differs from the visible h1. The markup describes the page; it does not replace it.
• A dateModified that changes on every deploy. A timestamp that moves without the content moving tells readers and engines nothing.
• author as a bare string where a Person with a url is what the feature expects.
• image URLs that are not crawlable or are blocked in robots.txt — the rich result needs to fetch the image.

Structured data describes the page; it does not make the page true. Google also requires the visible content to match, and it decides eligibility on its own published requirements — which change without notice. A valid block that matches the page is the whole of what you control.

Test it on a real URL with the free schema validator — no account, and it reports eligible, ineligible, or not applicable without inventing a score.

• Product — Marks a page as a product, with its price and availability.
• LocalBusiness — Marks a page as a physical business, with an address and opening hours.
• Organization — Marks a page as representing an organisation rather than a person.
• BreadcrumbList — Marks a page as a position in a hierarchy, for the breadcrumb in a result.
• Person — Marks a page as being about, or written by, a person.
• WebSite — Marks the site itself, usually on the homepage.
• FAQPage — Marked a question-and-answer page. Google stopped showing these as a rich result.
• HowTo — Marked a set of steps. Google stopped showing these as a rich result.