The Entity Authority Checklist

ClickRadius Institute · Published July 6, 2026

This is the capstone of our entity-authority series: every check from ten articles, consolidated into one working document. The context that makes it urgent is by now familiar — AI Mode is Google's default experience, AI Overviews appear on roughly 48% of queries per industry tracking, zero-click searches sit near 60% (about 93% inside AI Mode), and the #1 organic position's click-through rate has fallen from roughly 27% to about 11%. The businesses that get named in AI answers are the ones machines can resolve, verify, and trust. Industry estimates suggest a large majority of brands still have zero AI-search mentions — so this checklist is, for most categories, still a first-mover's document. Work it top to bottom; each section depends on the ones above it.

Two quotes frame why every item below exists. The architectural one:

"We've been working on an intelligent model — in geek-speak, a 'graph' — that understands real-world entities and their relationships to one another: things, not strings."

— Amit Singhal, then SVP of Search at Google, introducing the Knowledge Graph (2012)

And the evaluative one:

"Trust is the most important member of the E-E-A-T family because untrustworthy pages have low E-E-A-T no matter how Experienced, Expert, or Authoritative they may seem."

— Google Search Quality Rater Guidelines

Everything below either makes you a resolvable thing or makes that thing verifiably trustworthy.

Section 1: Foundation (checks 1–5)

1. Canonical fact sheet exists — one document with your exact public name, legal name, address, phone, URL, category, founding date, founders, service list, geography, and 25/100-word descriptions. Every later check reconciles to this. (Playbook, step 1)
2. One public name form, everywhere, forever. No LLC-sometimes, no abbreviations-sometimes.
3. Entity home designated — homepage or About page that states in plain visible text who you are, what you do, whom you serve, where.
4. The entity home says it in the first paragraph — resolvable language, not slogans. "Powering tomorrow's growth" resolves to nothing.
5. A named owner for entity facts — one person accountable for every propagation when facts change.

Section 2: Declaration (checks 6–12)

6. Organization/LocalBusiness JSON-LD on the entity home, populated from the fact sheet: name, legalName, url, logo, description, address, telephone, foundingDate, founder, areaServed.
7. Stable @id on the Organization node, referenced (not re-declared) by every other schema block on the site. Re-declaration drift manufactures ambiguity. (sameAs guide)
8. Curated sameAs array — every URL passes the test: does this page unambiguously identify exactly this entity? Platform profiles, knowledge bases, authoritative directories, registries. No category pages, no dead profiles, no namesakes.
9. Bidirectional loops closed — every sameAs destination links back to your domain. Reciprocity converts assertion into evidence.
10. Markup matches the visible page — no JSON-LD asserting facts the page no longer shows.
11. Article schema on content pages, with author referencing a Person @id and publisher referencing the Organization @id.
12. Structured data validates — run the site through a validator; broken JSON-LD is silent nullification of checks 6–11.

Section 3: Corroboration (checks 13–21)

13. Google Business Profile complete and canonical — every field, specific primary category, description written as the sentence you want an AI to repeat, holiday hours current. (GBP guide)
14. Bing Places and Apple Business Connect complete and matching.
15. Major horizontal directories (Yelp, BBB, national peers) — accurate, complete, closed-loop.
16. 3–5 industry-authoritative directories — the ones your market and the engines actually use; verify by asking the engines your customers' questions and noting what gets cited. (Directory guide)
17. Data aggregators corrected — fix the wholesale layer before the retail layer.
18. Stale-data hunt completed — searched old addresses, old phones, old names; corrected or removed every hit. The graveyard contradicts everything you build. (NAP guide)
19. Duplicates merged — one listing per platform, reviews consolidated.
20. Knowledge-base presence where eligible — Wikidata item with neutral description, referenced statements, external identifiers, and official-website loop; Wikipedia only if you genuinely clear notability. (Wikidata guide)
21. Substantive-conflict audit — no source anywhere asserts a different phone, address, or category. Formatting variance is tolerable; factual conflict is not.

Section 4: People (checks 22–27)

22. Expert inventory done — every human with citable expertise listed; legacy bylines consolidated to one canonical name per person. (Author entities)
23. Real bylines on every substantive page — "Admin" retired.
24. Author pages as entity homes — bio with checkable facts (degrees, licenses, years, work), photo, full article list, stable URL.
25. Person JSON-LD — jobTitle, worksFor pointing at the Organization @id, knowsAbout, and a person-level sameAs array (LinkedIn, registries, speaker pages).
26. Person loops closed — each expert's LinkedIn names the company and links the site.
27. Experts pointed outward — podcasts, panels, press commentary; every appearance is retrievable corroboration.

Section 5: Reputation and mentions (checks 28–33)

28. Systematic, policy-compliant review cadence — built into job completion; no gating, no incentives, no fakes (FTC's fake-review rule carries civil penalties, and engines cross-check).
29. Owner responses within days — including, especially, to negative reviews.
30. Retrieval map maintained — the specific roundups, directories, and community threads engines cite for your query space, refreshed quarterly.
31. Presence pursued on those surfaces — accurately and legitimately; sentences about you in retrieved documents are the asset. (Mentions vs backlinks)
32. Canonical description used wherever you control copy — mentions only build association when the language carries category and geography.
33. Mention quality over link count — acquisition question is "what will the page say about us?", not "is the link followed?"

Section 6: Citable content (checks 34–37)

34. Statistics in every substantive page — concrete, attributed numbers. The Princeton-led GEO study (KDD 2024) found statistics, quotations, and citations raised generative-engine visibility by up to 40%.
35. Attributed quotations — real voices with names, in blockquotes machines can parse.
36. Source citations — factual claims link to primary sources; verifiability is the citation lever.
37. First-hand specificity — numbers, dates, outcomes, and process detail only you could write. Engines cite expertise they cannot replicate. (E-E-A-T guide)

Section 7: Monitoring (checks 38–40)

38. Monthly answer-layer testing — ask ChatGPT, Gemini, Perplexity, Claude, and Grok your identity questions ("What is [brand]?"), recommendation questions ("Best [category] in [geo]?"), and topical questions. Log presence, accuracy, sentiment.
39. Wrong answers traced upstream — every inaccurate AI statement about you has a fixable source: a stale listing, a conflicting fact, a missing declaration.
40. Re-audit rhythm scheduled — declaration after any business change; corroboration sweep twice yearly; this checklist annually. Calendar entries, not intentions: monitoring that depends on someone remembering is monitoring that stops within a quarter, and the answer layer will drift silently while nobody is looking.

The five failure patterns this checklist prevents

Having audited many sites against these forty checks, we can tell you where real programs actually break — and each failure maps to a section above:

• The unresolvable brand (fails Section 1–2): the company that never decided what it is called or what it does in plain words. No amount of downstream work compounds onto an entity machines cannot lock onto.
• The self-contradicting record (fails Section 3): the most common pattern by far — years of accumulated listings quietly asserting two addresses, three phone numbers, and four descriptions. Engines respond exactly as the trust hierarchy predicts: they say less, or say wrong things, or recommend someone cleaner.
• The anonymous expert shop (fails Section 4): genuinely skilled practitioners publishing under "Admin," forfeiting the attribution signal that the entire E-E-A-T apparatus — and the Princeton-verified quotation lever — is built to reward.
• The link-era holdover (fails Section 5): budget still flowing to link counts while the surfaces engines actually retrieve for the category contain no sentence about the brand at all.
• The fire-and-forget project (fails Section 7): everything above done once, in a sprint, two years ago — then drift, unmonitored. Entity authority decays like any reputation; monitoring is what converts a project into an asset.

If you recognize your business in one of these, you also now know exactly which section to start with — that is the point of working the list in order.

How to use this document

A realistic pass for a typical established business: Sections 1–2 are a one-to-two-week sprint. Section 3 is two to six weeks including recrawl lag, dominated by waiting on third-party platforms rather than by your own effort. Section 4 is days of work gated only by having real people willing to put their names on their expertise. Sections 5–6 are permanent operating rhythm — budget them as recurring line items, not projects with end dates. Section 7 is an hour a month that tells you whether everything else is working, and it is the hour most programs skip first and regret longest. None of it requires a platform — this page is the complete recipe, and we mean it to be usable as-is. What a platform adds is persistence and scale: ClickRadius scores the on-site checks automatically across its six categories, auto-fixes declaration issues, and runs Section 7 continuously across all five engines — which matters because engine behavior shifts and manual spot-checks decay. Either way, the deciding factor is the same one Google's rater guidelines named years ago: trust, made machine-readable, one unglamorous check at a time.

The deeper references, section by section

Each section of this checklist compresses a full article in this series; use them when a check needs the reasoning, the examples, or the exact markup. According to the reading order that makes the dependencies clearest: start with the entity primer and the build playbook (Sections 1–2), then the sameAs guide for declaration mechanics. For Section 3, the trio of NAP reconciliation, directory presence, and the GBP deep-dive, plus the Wikidata guide where you qualify. Section 4 is author entities; Section 5 is mentions vs backlinks and the off-site evidence; Section 6 is grounded in the E-E-A-T translation. Ten articles, one architecture.

Frequently asked questions

In what order should I work through the entity authority checklist?

Top to bottom: foundation, declaration, corroboration, people, reputation, content, monitoring. Each section depends on the ones above it — corroboration without canonical declaration fragments, and mentions without consistent facts fail to fuse to your entity.

How often should the checklist be re-run?

Foundation and declaration after any business change; the corroboration sweep twice a year; reputation, mentions, and content continuously; answer-layer monitoring monthly at minimum, because engine behavior and retrieved sources shift.

Can I do all of this myself, or do I need a platform?

You can do it all manually — this checklist is the complete recipe. A platform adds automated scoring, auto-fixing of declaration issues, and continuous five-engine monitoring that manual checks can't sustain. By hand or with tooling: do it, because most competitors currently do neither.

Next step: see your starting point in minutes instead of auditing for a week — the free AI Readiness Score runs the measurable checks across six categories and five AI engines, and plans cover execution, auto-fixes, and continuous monitoring.