Author bios and credentials

An author box is the easiest E-E-A-T signal to bolt on and the easiest to make meaningless.

The difference is whether the name identifies somebody who'd actually answer for the content being wrong.

Does it have to be a person?

No, and this trips people up.

Google's guidance asks who's responsible for the content. Plenty of good publications answer that with an organization — a newsroom, a research outfit, a firm. What matters is that somebody real is accountable and that a reader can work out who.

A person's name adds something a company name doesn't: individual experience, an individual reputation, a specific career you can go and check. If you've got someone who genuinely did the work, use them.

But a company byline that's honest beats a person's name attached to something they never read. We use the firm's name on this site for exactly that reason.

Relevance beats prestige

Twenty years running law firm operations qualifies you to write about running law firms. It doesn't qualify you to write about attorney advertising rules in Ohio.

That distinction has to be allowed to stop you publishing, or it isn't a real distinction. We cut a planned legal playbook rather than write it without a lawyer, and one page in this section is still held back for the same reason. The byline only means something if it's permitted to constrain what goes out.

The common failure is a credential mismatch — a genuinely accomplished person bylined on a subject outside what they know. That's weaker than no byline at all, because it invites a check that fails.

One identity, everywhere

Someone who exists as three different identifiers across three sites is three weaker authors.

Pick one canonical identifier, declare it once, reference it everywhere. Same discipline entity consistency applies to companies, for exactly the same reason. Keep the profiles you claim consistent too, so somebody checking resolves you instead of splitting you.

In practice: one canonical bio page, the same identifier in your structured data across every property, and the same list of profiles everywhere.

What goes in it

Enough to establish standing, and nothing that can't be checked:

• What they do now, and for whom.
• Directly relevant experience. Years, roles, scale.
• Credentials that bear on this subject.
• Where else they can be found, so a reader can verify it without taking your word.
• Where their competence stops, if you publish anywhere near that line.

That last one is rare and weirdly convincing. A bio that marks its own limits is making a checkable claim instead of a sales pitch.

Markup

Person or Organization structured data, referenced by @id from the articles, states the relationship outright instead of leaving a machine to match name strings and hope.

It isn't a ranking lever and nobody should sell it to you as one. It removes guesswork. That's the whole argument for structured data in the first place.

The failure that actually costs you

Putting a real person's name on content they didn't write and haven't read.

It looks like the cheap version of this signal. It's the expensive one. It's a false statement about authorship, on a page asking to be trusted, and it falls apart the first time somebody asks that person about it.

If AI helped draft it, name whoever is accountable for it being right and say how it was made separately. That's what we do here, and the field recording it isn't something we can edit to get past our own build.