No Author Identity in Article Schema

Generative engines lean on a named, verifiable author as one of the few signals available for whether an article's advice can be trusted. Crawl Cove's ai-author-authority check flags article-family pages whose structured data either names no author at all, or names one with no way to confirm who they are.

What this finding means

This check evaluates every page that resolves at least one Article, BlogPosting, NewsArticle, TechArticle, ScholarlyArticle or Report node in its JSON-LD, and inspects that node's author property. A page with no article-family node in its structured data is skipped entirely: this check has no opinion about authorship on a pricing page, a tool, or a category hub.

Two distinct issues can fire on an article node:

Article structured data has no author

The author property is missing, empty, or an empty array, so there is no author signal at all.

Article author has no identity link (url/sameAs/@id) for entity disambiguation

An author is present, either a bare name string or an object carrying a name, but it has no url, sameAs or @id a generative engine could follow to confirm which specific person or organisation wrote the piece.

A page can carry several article nodes with different issues; the distinct messages combine into exactly one finding per page rather than one finding per node.

Why it matters

An unnamed byline reads the same to a generative engine as an anonymous one: there is nothing to check the claims against, so a model weighing two similar answers has a real reason to favour the one it can attribute to a specific, verifiable source. A named author with a linked profile, an author page on your own site, a LinkedIn profile, a Wikidata entry, gives the engine something to resolve the name against. Neither guarantees a citation, but an anonymous or unlinked byline removes a signal the engine could otherwise have used in your favour.

How to fix it

1. Add an author object to the page's Article JSON-LD, not only a visible byline in the HTML. A name in the page text is invisible to this check; it only reads structured data.
2. Give that author object a name and at least one of url, sameAs or @id. A link to an author bio page on your own site is enough; a sameAs array can add LinkedIn, a personal site, or a Wikidata entry on top.
3. For organisation-authored content, use an Organization author with the same name plus url/sameAs shape, rather than inventing an individual byline the article does not really have.
4. If several people contributed, list them as an array and make sure at least one carries a full name and an identity link.
5. Re-crawl the page to confirm the finding clears once the author identity is in place.

False positives and edge cases

• Pages with no article-family JSON-LD are skipped, not flagged. A missing author on a product or landing page is not what this check evaluates.
• A bare author name string, no object, no link, always raises the identity-link issue. A plain string cannot carry a url, sameAs or @id, so naming an author without a linkable object clears the "no author" issue but not this one.
• An array of authors passes as a whole once one entry is both named and linked. Co-authored pieces do not need every contributor identified to clear the check.
• Non-200 pages produce no finding. A page that cannot be fetched is not evaluated for authorship at all.

Related reading

For the check that scores a page's overall structural readiness for AI answer engines, including author and date signals alongside answer-first openings, see Low AI Extractability Score. For the sibling check on the same Article schema, whether it names when the page was last updated, see No dateModified in Article Schema.