Schema untuk Author dan Reviewer: Kapan Membantu Trust

Di banyak website, byline muncul karena template.

“Ditulis oleh Admin.”

“Reviewed by Tim Ahli.”

“Expert verified.”

Kelihatannya trust-building.

Lalu kita klik “Tim Ahli”.

Tidak ada profile.

Tidak ada nama.

Tidak ada bidang expertise.

Tidak ada tanggal review.

Tidak ada penjelasan siapa yang melakukan apa.

Schema bisa menandai author.

Schema.org juga memiliki relationship seperti `reviewedBy` pada WebPage untuk orang atau organisasi yang telah meninjau content dari sisi accuracy atau completeness.

Tetapi markup tidak bisa mengubah label kosong menjadi trust.

Author dan reviewer membantu ketika mereka merepresentasikan accountability yang nyata.

Bukan ketika nama ditempel untuk kosmetik.

Author adalah Factual Relationship

Google Article structured-data documentation secara eksplisit merekomendasikan author markup.

Google meminta site memasukkan semua author yang ditampilkan sebagai author pada page dan menyarankan memberi `type` serta `url` atau `sameAs` yang membantu mengidentifikasi author.

Ini useful.

Article:
ditulis Person A.

Schema:
author Person A.

Profile:
Person A.

Now relationship clear.

If Organization is actual author:
Organization can be author.

Google supports Person or Organization for Article author.

Important:
choose actual authorship.

Not whoever has biggest title.

CEO Bukan Default Author

Article tentang AI.

CEO company ahli business.

Writer adalah research editor.

Marketing berpikir:

“Kalau CEO jadi author, E-E-A-T lebih tinggi.”

Lalu byline CEO.

CEO never saw draft.

That's misrepresentation.

Structured data should not manufacture authorship.

If CEO reviewed:
maybe reviewer relation can be represented in content/internal workflow.

If CEO contributed:
credit contributor.

If wrote foreword:
credit foreword.

Role precision.

Trust comes from real accountability.

False authorship destroys trust if discovered.

“Admin” adalah Masalah Identitas

Can “Admin” technically be author name?

A system can output.

But user learns nothing.

If article is institutional editorial:
use Organization or named editorial team if that team is a real identity and page explains.

If individual wrote:
name Person.

Google author best practices emphasize identifying author and linking to a page that uniquely identifies them.

So “Admin” is low-information.

Don't create fake Person named Admin with photo logo.

Model correct type.

Author Page Harus Berguna

Profile page:

Name.

Current role.

Organization.

Expertise.

Bio.

Selected public work.

Author archive.

Professional identity links.

Contact channel if appropriate.

No need private data.

For author disambiguation, enough.

If Person name common:
profile link becomes valuable.

Multiple authors:
each own Person.

Don't merge names into one author string.

Google explicitly recommends listing each author separately in markup.

This is one area where structured data clearly mirrors good editorial practice.

Reviewer Adalah Role yang Berbeda

Writer:
creates text.

Reviewer:
checks accuracy/completeness.

Editor:
shapes narrative.

Fact checker:
verifies claims.

Medical reviewer:
reviews clinical accuracy.

Legal reviewer:
reviews legal interpretation.

These roles are not same.

If site says “Reviewed by Dr X,” it should mean Dr X actually reviewed the published version or defined revision.

Not:
Dr X works at company.

Not:
Dr X approved content strategy months ago.

Not:
Dr X is advisor and name is reused globally.

Review relationship must be real and scoped.

Schema.org `reviewedBy` can express people or organizations that reviewed a WebPage for accuracy/completeness.

But Google Article rich-result documentation focuses on `author`; `reviewedBy` is not a magic Google Article trust field.

This distinction matters.

Schema.org semantic capability ≠ documented ranking feature.

Trust Is Not a Markup Property

You cannot write:

“trusted: true.”

Reviewer schema doesn't force search engine trust.

What it can do:
make role explicit to consumers that understand it.

Visible reviewer information can help human readers assess accountability.

The trust effect is human/editorial and possibly machine-understanding related.

But don't claim:

“Add reviewedBy and rankings increase.”

No official universal basis.

If reviewer program improves content accuracy, quality can improve.

That's meaningful.

Markup is representation.

Not source of expertise.

Skenario Meeting

SEO:
“Kita tambahkan reviewedBy ke 800 articles.”

Editor:
“Siapa yang review?”

SEO:
“Dokter kita.”

Doctor:
“Saya cuma review 12 artikel.”

Silence.

Exactly.

Don't scale reviewer markup beyond actual workflow.

If 12 reviewed:
mark 12.

For rest:
no reviewer.

Empty field is better than fake accountability.

Schema governance should enforce evidence:
review event exists.

Reviewed version.

Date.

Reviewer.

Scope.

Status.

Reviewer Expertise Harus Match Topic

Tax lawyer reviews tax article.

Relevant.

Doctor reviews medical article.

Relevant.

Cybersecurity engineer reviews bot-verification guide.

Relevant.

Celebrity founder reviews nutrition article.

Not necessarily.

A reviewer isn't trust decoration.

They should have competence related to claims.

Visible profile explains.

No need inflate credentials.

No fake “expert”.

Use actual role.

If reviewer only checks grammar:
call editor.

Don't imply technical verification.

Review Date Matters

Article published January.

Reviewed February.

Updated June.

Is February review still valid after June update?

Maybe not.

If major content changed, re-review.

Store:

reviewed_at.

content_version.

reviewer.

scope.

If minor typo update:
maybe no re-review.

Define policy.

For high-stakes content:
stricter.

Reviewer relationship is temporal.

A stale “medically reviewed” badge after major rewrite can mislead.

Schema doesn't know unless workflow updates.

Editorial system must.

Author vs Publisher

Article by Person A.

Published by GEO.or.id.

Schema:
author Person A.
publisher Organization.

Don't put publisher name in `author.name`.

Google author best practices explicitly warn against putting publisher in author name and recommend using publisher property appropriately.

This seems small.

But semantic clarity matters.

Publisher is responsible for publication.

Author creates work.

Both can build accountability.

Different roles.

Organization as Author Is Legitimate

Not every piece needs named person.

Institutional reports.

Policy statements.

Product documentation.

Press release.

Team research.

Organization may be actual author.

Google Article markup supports Organization as author.

Use it when truthful.

Don't invent Person because industry says “Google prefers people.”

The documentation doesn't require that simplification.

If content really belongs to institutional editorial process:
Organization author can be more honest.

Then profile/about explains organization.

Authorship policy visible.

Multiple Authors Need Real Credits

Research report:

Analyst A.

Researcher B.

Editor C.

Who is author?

Depends contribution.

Use authors for substantive creators.

Editor as contributor/reviewer maybe.

Don't list every employee.

Byline should match markup.

Google says all visible authors should be in markup.

Conversely, adding hidden “SEO authors” not visible can violate visible-content principle.

One source of truth:
CMS contributor roles.

Render visible.

Generate schema.

Clean.

Reviewer Profile Should Not Expose Private Data

Same privacy principle.

Name.

Professional role.

Expertise.

Organization.

Credentials that are public/relevant.

Official profiles.

No home address.

No personal phone.

No sensitive data.

A reviewer page doesn't need CV dump.

Machine identity can be strong with professional attributes.

Person entity governance.

Especially regulated experts:
license information only if public and appropriate.

Keep current.

Credentials Expire

Professional licenses.

Board certification.

Employment.

Academic affiliation.

Reviewer profile can become stale.

If trust badge relies on credential:
monitor.

Don't keep:
“Practicing doctor at Hospital A”
if person moved.

Maybe credential remains.

Role changed.

Trust infrastructure needs maintenance.

This is why generic “expert reviewed” feels easier.

But generic is less accountable.

Good trust costs operations.

Author Schema Helps Disambiguation

Person named “Ari Wibowo.”

Many people.

Article author URL:
`/authors/ari-wibowo-data-engineer/`

Profile:
organization, expertise, work.

sameAs:
official LinkedIn or ORCID if same person.

Now identity clear.

Without URL:
name alone ambiguous.

Google author docs specifically recommend `url` or `sameAs` for better understanding.

This is tangible schema use.

Not ranking guarantee.

Identity clarity.

Reviewer Schema Can Be Useful Internally Even if Search Doesn't Surface It

An institutional site can maintain:

WebPage → reviewedBy → Person.

Even if Google Search doesn't create a special rich result for reviewer.

Why useful?

Internal graph.

Audit.

Content governance.

Potential machine consumers.

CMS queries:
Which pages reviewed by expert?

Which review expired?

Which reviewer left?

Which topics need re-review?

Structured semantics can support this.

But if page does not show reviewer, be cautious with Google's visible-content guideline for Search-targeted markup.

Better:
display reviewer transparently.

Trust should be public.

Not hidden.

High-Stakes Content Needs Stronger Review Process

Health.

Legal.

Financial.

Tax.

Safety.

Cybersecurity.

Policy.

No schema property can replace domain review.

For such content:

Qualified author/reviewer.

Source references.

Update date.

Method.

Limitations.

Jurisdiction.

Version.

Correction policy.

Then structured data.

This is where reviewer can genuinely help trust.

Because process exists.

Markup just exposes.

For low-stakes lifestyle article:
formal reviewer maybe unnecessary.

Don't bureaucracy everything.

Kapan Author Schema Membantu?

Almost whenever Article authorship is meaningful and visible.

Especially:

News.

Blog.

Research.

Technical guide.

Opinion.

Long-form analysis.

Multi-author.

Named experts.

Author archive.

It helps identity structure.

Kapan Reviewer Relationship Membantu?

When review is a real editorial control.

Medical/technical/legal/fact-check review.

Research QA.

Institutional accuracy review.

High-impact content.

Not all articles need.

If no review workflow:
don't add.

Kapan Tidak Membantu?

Fake expert badge.

Hidden reviewer.

Same reviewer sitewide without actual review.

Reviewer unrelated field.

CEO used as author for all.

Admin pseudonym with no identity.

Author profile copy-pasted.

Credentials stale.

Markup only, no visible accountability.

This can decrease trust if user investigates.

Trust is fragile.

Schema can expose inconsistency faster.

Content Trust Stack

Layer 1:
Accurate content.

Layer 2:
Sources/evidence.

Layer 3:
Real author.

Layer 4:
Review process where needed.

Layer 5:
Profiles and credentials.

Layer 6:
Structured data that mirrors roles.

Wrong order:

Layer 6 first.

Then hope trust appears.

No.

Markup belongs last.

A practical CMS model:

author_ids.

editor_id.

reviewer_ids.

reviewed_at.

review_scope.

content_version.

published_at.

modified_at.

Then page renders:

Written by.

Reviewed by.

Updated.

Sources.

Schema maps.

This is scalable.

No manual JSON-LD editing.

No fake role.

Data comes workflow.

GEO Measurement Should Not Use Reviewer as Citation Hack

Could adding reviewer correlate with better AI visibility?

Maybe.

But reviewer deployment often comes with:

Better source citations.

Content updates.

Improved expertise.

Profile pages.

Editorial QA.

Which factor matters?

Unknown.

Measure.

Don't say:
“reviewedBy causes citation.”

Say:
“Editorially reviewed content provides clearer accountability; structured data can express authorship/review relationships where appropriate.”

That's enough.

Human Reader Is Final Test

Imagine user asks:

“Siapa yang bertanggung jawab atas informasi ini?”

Can page answer?

Who wrote?

Who reviewed?

When?

Why qualified?

Sources?

If yes:
trust architecture works.

If schema says Person A but page gives no clue:
machine-first.

Fix.

Author and reviewer markup is most valuable when it formalizes something the organization already does well.

Real authors.

Real experts.

Real review.

Real accountability.

If those don't exist, don't manufacture them in JSON-LD.

Trust isn't generated by property names.

It's generated by people willing to put their name behind work they actually did.