Article Schema and Structured Data: Making News Machine-Readable

A page can read perfectly to a human and still be a blank wall to a machine. That gap is what structured data closes. For news publishers it is not optional anymore — it is how you qualify for rich results, how Google understands recency, and increasingly how AI systems decide what to quote.

Start with Article, not the kitchen sink

You do not need forty schema types. You need Article (or NewsArticle) done correctly: headline, author with a real profile, datePublished and dateModified, publisher, and a primary image. Get those five right and you have covered most of what surfaces in search.

Dates are a trust signal

News lives and dies on freshness. If your markup omits or fudges publish dates, you teach the crawler to distrust your timestamps. Keep dateModified honest — bump it when you genuinely update, not on every cache refresh.

Authors need an identity

“By Admin” is a missed opportunity. Tie each piece to a named author with a bio page and, where possible, links to their profiles elsewhere. This is the raw material for E-E-A-T, and it is what separates a credible blockchain news section from an anonymous content farm.

Validate, then watch

Run every template through the Rich Results Test, then watch Search Console’s enhancement reports for warnings. Schema that throws errors is worse than none — it signals sloppiness. Fix the template once and every article inherits it.

Structured data is the cheapest authority you can buy. It costs a developer an afternoon and pays out every time a machine has to decide whether your page is worth showing.