The E-E-A-T Checklist: 27 Signals Google Actually Evaluates

If you've searched for an E-E-A-T checklist, you've found a dozen articles claiming various numbered lists — 12 signals, 18 signals, 30 signals, no two the same. That's because Google has never published an official numbered list. Every checklist online is the author's synthesis of what Google's documentation, rater guidelines, and public statements actually say.

This is mine. The 27 signals below come from three primary sources: Google's Search Quality Rater Guidelines (last updated September 2025), Google's Creating Helpful, Reliable, People-First Content page, and on-the-record statements from Googlers including Danny Sullivan and John Mueller. Where a signal isn't directly stated by Google but is widely accepted in the SEO community, I've flagged it.

I run a digital PR and link building agency. We work with about 500 clients across healthcare, SaaS, eCommerce, and B2B services. So the part of E-E-A-T I have the most direct experience with is the Authoritativeness pillar — the off-site signals that come from earned mentions, citations, and links. That's where most of the under-served signals on this checklist live.

What E-E-A-T Stands For

E-E-A-T is Google's framework for evaluating content quality, made up of four pillars: Experience, Expertise, Authoritativeness, and Trust. It's used by Google's human Quality Raters to assess search results — not as a direct ranking factor, but as a guide that shapes how the algorithms are tuned.

Why E-E-A-T Hits Differently After the March 2026 Core Update

Experience signals carry more weight after the March 2026 core update than at any point since Google added the second "E" to the framework. Sites with named cases, original screenshots, and specific implementation details held or gained rankings. Sites with generic, surface-level content lost ground — even on queries where they previously dominated.

The framework itself isn't new. The Search Quality Rater Guidelines have used a quality framework since 2014 — first as E-A-T (Expertise, Authoritativeness, Trust), then in December 2022 with the addition of Experience. What changed in 2026 is the weight Google appears to give Experience signals specifically.

The March 2026 core update completed on April 8, 2026. Google's own documentation for this update emphasized "experience signals beyond all previous indicators." If you've read a generic E-E-A-T checklist that treats all four pillars equally, that's no longer accurate. Experience is doing more work than it used to. Trust is still the foundation. Authoritativeness is the most movable through deliberate work. Expertise alone — without Experience — gets you less than it did three years ago.

The Full Checklist: 27 Signals Across 4 Pillars

Each signal below is one a human Quality Rater would check. Trust is listed last, but it's the most important — Google explicitly states this in the Quality Rater Guidelines. I've kept the descriptions short on purpose. The full picture is the combined pattern, not any single signal.

For the quick reference version, here's the full set in one table. The detailed walkthrough follows.

Experience (5 signals)

The five Experience signals all answer the same underlying question: did the author actually do the thing they're writing about?

1. First-person language tied to the topic. "I tested," "we ran," "in our campaign," "when I visited." Not just first-person opinion, but first-person action. Google's December 2022 announcement on Experience specifically called out content "demonstrating actual use of a product, having visited a place, or communicating personal experiences."
2. Original photos, screenshots, or visuals from the author's own work. Stock photography signals nothing. A screenshot of a tool's interface as it looked when the author actually used it — with the author's own data visible — signals genuine use.
3. Named cases, clients, or projects. "Our client X saw Y" beats "case studies show." The named version is verifiable. The unnamed version is hand-waving.
4. Specific implementation details. Pricing, version numbers, interface descriptions, settings used. The kind of detail you can only produce after actually using something. Surface-level reviews skip this; experienced users include it.
5. Original data or test results. Even small samples (10 sites, 20 emails, one campaign) beat zero. The act of generating the data is itself an Experience signal — it's evidence the author engaged with the topic beyond reading.

Expertise (5 signals)

Expertise is about knowing. Experience is about doing. The two overlap — Google says so directly — but they're distinct dimensions. Expertise without Experience reads as a research summary. Experience without Expertise reads as a personal blog without rigor.

6. Author byline with visible credentials. Name, role, company, and one line of relevant qualification. Not a generic "Marketing Team" byline. Google's Helpful Content guidance explicitly asks: "Do bylines lead to further information about the author or authors involved?"
7. Topical depth beyond the obvious. Does the content address edge cases, exceptions, contradictions? Or does it stay at the level a five-minute search could produce? Google's self-assessment questions explicitly ask whether content "provides insightful analysis or interesting information that is beyond the obvious."
8. Correct technical terminology used appropriately. Wrong terminology is the fastest way to identify content written by someone outside the field. Right terminology, used in correct contexts, is one of the cheapest credibility signals.
9. Citations to primary or authoritative sources. Linking to the original study, the official documentation, the regulatory body — not the secondhand summary. This signals the author understood the source material directly.
10. Topical concentration in the author's body of work. Authors who write across the same topic across multiple articles signal expertise more than authors who jump between unrelated topics. This is checked at the author level, not the site level.

Authoritativeness (6 signals)

Authoritativeness is the off-site reputation pillar. It's also the one most generic E-E-A-T guides barely cover, because the people writing about E-E-A-T usually aren't the ones running off-site campaigns. This is the pillar where digital PR work moves the needle most directly.

11. Backlink profile quality. Not just count — quality and topical relevance of the linking domains. Google's Quality Rater Guidelines explicitly tell raters to consider what other sites say about the page being evaluated.
12. Brand mentions in established publications, linked or unlinked. Both count. Google can identify brand mentions without anchored links. A brand mentioned in the New York Times without a link still signals reputation.
13. Author has been quoted, interviewed, or featured externally. The author's name appearing in third-party publications — as a source, a contributor, or a subject — builds personal authority that follows them across content.
14. Topical concentration of the site itself. Sites that publish 80%+ of content within a single topical cluster are read as authorities on that cluster. Generalist content farms — sites publishing across health, tech, finance, and travel — are penalized for the spread.
15. External authority signals beyond the website. Author's LinkedIn presence, conference speaking, podcast appearances, books, professional certifications. These don't directly rank a page, but they're checked by Quality Raters when evaluating the author's authority.
16. Bylines on tier-one third-party publications. When the same author has bylines on Forbes, Inc., Search Engine Journal, or industry-equivalent publications, that authority transfers back to their primary site. This is one of the most impactful signals to actively build.

Trust (7 signals)

Trust is the foundation. Google's exact wording in the Quality Rater Guidelines: "Untrustworthy pages have low E-E-A-T no matter how Experienced, Expert, or Authoritative they may seem." If your site fails on Trust, the other three pillars don't save you.

17. HTTPS on every page. The most basic signal, and the one most sites get right. Mixed-content warnings (HTTPS site loading HTTP resources) still count against Trust. Check with a browser-level audit, not just the URL bar.
18. Real, multi-method contact information. An email form by itself is weak. Email plus phone plus physical address signals a real business. PO boxes count for less than street addresses.
19. About page with real, named team members. Photos, names, roles, and bios — not stock images and made-up titles. The About page is one of the first things Quality Raters check when assessing Trust.
20. Editorial standards, fact-check policy, or correction policy. A linked policy describing how content is reviewed and corrected. This signal is increasingly important post-2024 as Google emphasized editorial review for AI-assisted content.
21. Reviews and ratings on third-party platforms. Trustpilot, G2, Clutch, BBB — whatever's relevant to the industry. Quality Raters can and do check third-party reputation sources to evaluate Trust.
22. Recent updated dates that reflect actual changes. Updating the published date without changing content is one of the practices Google explicitly warns against. Fresh dates need to reflect fresh content. The "dateModified" field in your schema must match real edits.
23. NAP consistency across the site and external sources. Name, Address, Phone — consistent across the website footer, Google Business Profile, social profiles, and citation sources. Discrepancies signal an unmaintained or possibly fraudulent business.

YMYL Add-ons (4 additional signals)

YMYL — Your Money or Your Life — is Google's classification for content where bad information could cause real harm: health, finance, legal, safety. As of September 2025, Google expanded the YMYL category to explicitly include elections and civic information. If your site falls in YMYL, the four pillars apply with higher rigor, plus these four additional signals.

24. Professional credentials displayed with verifiable specificity. "MD," "CPA," "JD," license numbers, board certifications. Generic claims of expertise don't count — Quality Raters specifically check whether credentials match standard naming conventions for the field.
25. Disclaimers and disclosures where relevant. Affiliate disclosures, sponsored content disclosures, "this is not medical advice" notices on health content, "consult a financial advisor" on finance content. Missing disclaimers on YMYL content is a fast Trust failure.
26. Citations to regulatory or scientific primary sources. FDA, CDC, NIH, SEC, official government bodies, peer-reviewed journals. Health content citing WebMD instead of the underlying study is a downgrade signal.
27. Documented expert review process. "Reviewed by [Dr. Name, MD]" with the reviewer's credentials. The review process needs to be a real, named human — not a generic "medically reviewed" badge with no attribution.

How Backlinks Actually Build the Authoritativeness Pillar

Backlinks build the Authoritativeness pillar through three distinct off-site signal patterns: linked editorial mentions, unlinked brand mentions, and bylines on tier-one publications. Each contributes to authority differently, and the highest-leverage strategy stacks all three.

Most E-E-A-T checklists treat "authority" as something a site develops by publishing more content. That's wrong. Authority is conferred by external recognition — being mentioned, cited, and linked by other authoritative sites in your space.

The two services that most directly map to this pillar are Digital PR — earning placements through reactive expert-source pitching — and Full-Feature Articles, which place dedicated editorial pieces on tier-one publications like USA Today, VentureBeat, and Fast Company. Both build the kind of mentions and citations Quality Raters check when evaluating Authoritativeness.

For more on what makes a strong overall backlink profile, we wrote a dedicated breakdown on what to track and how the patterns differ across industries.

The 5 Signals That Actually Move Rankings

Five signals do most of the work: first-hand experience, named author bylines, backlink profile quality, tier-one publication bylines, and a real About page. The other 22 signals still matter — and a Quality Rater would still check them — but the gap between sites that nail these five and sites that don't is where rankings get won and lost.

This ranking comes from observing what changes in our clients' rankings when specific signals get fixed. It's not from a published study — it's from running campaigns since 2017 and watching what correlates with movement.

1. First-hand experience signals (Signals #1-5). The Experience pillar collectively. Post-March 2026 update, this is the most amplified set. A page with no first-person language, stock photos only, no named cases, and no original data is at a structural disadvantage no matter how authoritative the site otherwise looks.
2. Author byline with credentials (Signal #6). Almost free to fix, almost universally underdone. Sites that switch from generic "Marketing Team" bylines to named authors with real bios consistently see rankings improve over the next 60-90 days. The signal is cheap; the omission is expensive.
3. Backlink profile quality (Signal #11). Not just count — quality and topical relevance of linking domains. A site with 50 referring domains from topically relevant authorities outperforms a site with 500 referring domains from generic directories. Google has gotten better at devaluing irrelevant links since the 2024 spam updates.
4. Bylines on tier-one publications (Signal #16). The single highest-leverage signal an individual author can build. One byline on a high-authority third-party publication does more for the author's E-E-A-T signal than ten on the author's own site. We see this most clearly in B2B SaaS — founders with bylines on Inc., Forbes, or Fast Company rank their own content meaningfully better than equally-credentialed founders without external bylines.
5. About page with real team members (Signal #19). Trust failures here cascade. Quality Raters check the About page early in their evaluation. Sites with stock-photo team pages, no real names, or "team" pages that don't actually list anyone fail Trust at the foundation level. Once Trust is failing, the rest of the framework can't recover the page.

What's not on the top five list is also worth calling out. HTTPS gets too much attention — almost every site has it, so it provides almost no differentiation. NAP consistency matters but is often pre-fixed for any business with a real Google Business Profile. Editorial standards pages are widely treated as a magic Trust booster; in practice, they help, but not as much as fixing author bylines first.

The 5 Most Common E-E-A-T Mistakes We See

The five most common E-E-A-T gaps across client audits are generic author bylines, stock photography in product content, hollow About pages, citations to summaries instead of primary sources, and dates updated without content updated. In rough order of how often we see them:

1. Generic author bylines on every post. "Marketing Team," "Editorial Staff," "Admin." This is the single most common gap and the single easiest fix. Even one named author across the site beats no named author. Add real names with real bios.
2. Stock photography in places where original photography would be cheap. A SaaS company writing about their own product using stock photos of generic dashboards instead of screenshots of their actual product. The original visual exists. Using it costs nothing. Most sites just don't.
3. About pages that don't actually describe the team. Mission statements, values, and "Founded in [year]" copy without ever showing the humans involved. The About page is supposed to tell Quality Raters who you are. Most About pages don't.
4. Citations to summaries instead of primary sources. Linking to a HubSpot article that summarizes a Google study, instead of linking to the Google study directly. Quality Raters notice; algorithms approximate the same pattern.
5. Dates updated without content updated. Republishing 2022 articles with 2026 dates and no actual changes. Google explicitly warns against this. The freshness signal is meant to reflect freshness — when it doesn't, it eventually gets devalued or treated as a deceptive signal.

How to Improve Your E-E-A-T: 30, 60, and 90 Day Plan

Improving E-E-A-T breaks into three 30-day phases: foundation Trust signals first, on-site content audit second, off-site authority work third. None of this is something you fix in a week, but the moves with the biggest payoff come early.

Days 1-30: Foundation

The cheap, fast Trust signals first. Most sites have gaps here that take a day or two to close.

• Audit your About page. Are real team members listed with photos and roles? If not, fix this first.
• Audit your Contact page. Is there a real phone number, real email, and real address? Not just a contact form.
• Audit your byline strategy. Are articles attributed to named authors with real bios? Or to generic team accounts?
• Verify HTTPS is clean across the site (no mixed-content warnings).
• If you're YMYL, add disclaimers and disclosures where missing.

Days 31-60: Content Audit

Now look at what's already published. Most sites have older content that fails the Experience and Expertise tests.

• Identify your top 20 highest-traffic pages. For each, check: does it use first-person language anywhere? Are there original visuals? Are named examples used?
• Pick 5-10 pages to actively rewrite with stronger Experience signals — pages where adding first-hand observation would genuinely improve the content.
• For YMYL content specifically, add expert review attributions where missing.
• Replace stock images with original visuals where the original visual exists and using it would be cheap (especially for product or service content).

Days 61-90: Off-Site Authority

By now the foundational Trust signals are solid and the on-site Experience signals have been improved. Time to work on the off-site Authoritativeness pillar — the slowest signal to build but the one with the highest ceiling.

• Build a digital PR pipeline focused on earning expert-source mentions in tier-one publications.
• Pursue 2-3 byline placements per quarter for your founder or lead authors on industry-relevant publications.
• Audit unlinked brand mentions — sites mentioning your brand without linking — and pursue link reclamation where natural.
• For sites with weak topical concentration, decide which 2-3 topical clusters to consolidate around and prune or de-emphasize the rest.

E-E-A-T for YMYL Sites

For YMYL sites, the four pillars and 23 base signals all still apply — the difference is rigor, plus four YMYL-specific signals (24-27) that aren't required on other sites. A signal that's nice-to-have on a recipe blog is required on a healthcare site.

YMYL (Your Money or Your Life) is Google's classification for content where errors can cause real-world harm. Health, finance, legal, safety — and as of the September 2025 Quality Rater Guidelines update, also elections and civic information.

The four YMYL-specific signals are particularly load-bearing for sites in healthcare SEO and financial services verticals. We've seen healthcare sites with strong content lose rankings simply because their author bios didn't list MD or DO credentials, even when the actual authors held those credentials. The credential needs to be visible, not just held.

Documented expert review processes (Signal #27) are increasingly the differentiator for YMYL sites. Generic "Medically Reviewed" badges with no named reviewer are weaker than specific attributions like "Reviewed by Dr. [Name, MD], Internal Medicine, [Institution]." Specificity is the signal.

For YMYL sites, the off-site Authoritativeness work also has higher minimum standards. Mentions in industry-trade publications matter; mentions in unrelated lifestyle blogs matter less. The relevance of the linking domain to the YMYL topic carries more weight than raw domain authority.

The Trust pillar is non-negotiable for YMYL. White hat link building practices are required, not optional — any sign of manipulated authority signals can put a YMYL site at higher penalty risk than a non-YMYL site doing the same thing.

Does E-E-A-T Affect AI Search Visibility?

Yes. The same signals that build E-E-A-T for traditional Google rankings increasingly correlate with citation rates in AI search engines like ChatGPT, Gemini, and Perplexity. Investments in Experience signals, named authors, and tier-one bylines pay off in both Google rankings and AI citation rates.

This isn't surprising once you think about it. LLMs are trained on web content and tuned to surface what humans rate as helpful. The framework Google's Quality Raters use to evaluate helpfulness is similar to the framework an LLM is implicitly optimizing for when it decides what to cite.

Practically, this means E-E-A-T work has compounding returns. We're tracking citation patterns across our own client portfolio, and the correlation is strong enough that we now treat E-E-A-T improvements as part of GEO (generative engine optimization) work, not just traditional SEO.

Frequently Asked Questions

Sources

Google Search Quality Rater Guidelines (September 2025 update) — services.google.com/fh/files/misc/hsw-sqrg.pdf
Google Search Central, Creating Helpful, Reliable, People-First Content — developers.google.com/search/docs/fundamentals/creating-helpful-content
Google Search Central Blog, "E-A-T gets an extra E for Experience" (December 2022) — developers.google.com/search/blog
Search Engine Journal, "Google Confirms March 2026 Core Update Is Complete" (April 2026)
Search Engine Land, March 2026 core update completion coverage
John Mueller, Search Central Live NYC remarks on E-E-A-T as a framework
Reporter Outreach client campaign data, 2017-2026