Article 13 min read
E-E-A-T: What Google Wrote, and What the Industry Sells Under the Acronym
by Daniel Sócrates published
The spreadsheet arrived along with the proposal. One of its rows carried a score: E-E-A-T, 6 out of 10. Next to it sat the plan to raise the score. An author page, a bio, badges in the footer.
Before you accept the number, run the search. On September 14, 2026, we searched “e-e-a-t” on Google in Brazil, where we are based. An AI Overview opened the page. It called the acronym a set of criteria Google uses to evaluate content. At the end, it offered help optimizing your site for E-E-A-T. 1
The screen leaves three questions open. Who wrote this yardstick, and for whom? Does a score like that exist anywhere inside Google? And which parts of that plan have a document behind them? This piece separates what Google wrote from what the industry sells under the acronym.
If you run a medical practice or an accounting firm, the yardstick for your field has its own piece: SEO for doctors and SEO for accounting firms. This one covers the whole acronym, from its origin to its limits.
A manual written for the people who test search
Picture a coffee chain with a hundred locations. Headquarters hires mystery shoppers. Each one carries a checklist: the coffee was hot, the cashier said hello, the change was right. One mystery shopper’s checklist never closes a store. Added up, the checklists show whether the new training worked.
Google has a similar team. They are people who receive search results and rate them, following a public manual. This is called a search quality rater.
The manual is called the Search Quality Rater Guidelines. At the official address, on September 14, 2026, the version posted was the one from September 11, 2025, at 182 pages. 2 E-E-A-T lives there, in a section that runs from page 26 to page 28. 2
Page 6 says what the ratings do. No single rating directly changes how a page appears in Search. None of them moves a site up or down. 2 The manual gives the reason: humans could never rate every page on the open web one by one. The ratings measure whether search is working and serve as examples of good and bad results. 2
Google has been saying this for at least eleven years. In 2015, it wrote on its developer blog that ratings do not determine where individual sites rank. 3 In 2022, it wrote that the guidelines do not directly influence ranking. 4 Today, its page on helpful content says raters have no control over how pages rank. 5
The spreadsheet in the proposal gave your site a score. A rater’s checklist measures a test of search, and that rating moves no one up or down.
The four letters, in the manual’s order
The acronym packs four words. The manual defines each one with a question and an example. 2
Experience is how much the content creator has lived the subject firsthand. The manual’s example is a product review. Would you trust someone who used the product, or someone who never held it? 2
Expertise is the knowledge or skill the topic calls for. Here the example is rewiring a house. On one side, a skilled electrician. On the other, an antique homes enthusiast who knows nothing about electrical wiring. 2
Authoritativeness is how well the author or the site is known as a go-to source on the topic. For renewing a passport, the official government page is the unique, official source. 2 Topical authority, with the three meanings it carries in the industry, has a piece of its own. The A in the acronym is about being recognized as a source.
Trust is how accurate, honest, safe, and reliable the page is. The manual puts this letter at the center and calls it the most important member of the family. 2 Its example has a dry humor to it. A financial scam is still untrustworthy, even when the person behind it is a highly experienced, expert scammer. 2 An untrustworthy page has low E-E-A-T, whatever the reason. 2
The other three letters hold up the fourth. The manual says they can overlap, and that the useful combination changes with the topic. 2 The helpful content page adds that content does not have to demonstrate all four. 5
The E that arrived in 2022
Until 2022, the acronym had three letters: E-A-T. On December 15, 2022, Google’s blog announced the E for experience, in a post signed by Elizabeth Tucker. 4
The post’s example comes from everyday life. To fill out your tax return, you want content from an expert in accounting. To pick the software that helps you file, you might prefer a forum discussion with people who have used it. 4 The same post says these are not fundamentally new ideas. 4
The manual is also older than it looks. Two Portuguese-language texts on the first page of that search say it appeared in 2019. 6 Google’s 2015 post says the guidelines were published in 2013. 3
What Google writes about ranking
The sentence that settles almost every meeting sits on the helpful content page. Google’s systems identify a mix of factors that help recognize content with these four qualities. 5 E-E-A-T itself is not a specific ranking factor. 5 The SEO Starter Guide raises the same question and answers no, in three words. 7
The same page says where the weight goes up. These are topics that could affect people’s health, financial stability, or safety. On them, the systems give even more weight to content with strong E-E-A-T. 5 Google calls these topics YMYL, short for Your Money or Your Life. 5
In my book SEO de Entidade, published in Portuguese, I recorded a silence that confirms the sentence. The internal documentation that leaked in 2024 holds thousands of attributes, and none of them is an E-E-A-T score. 8 The acronym describes the target. The measures that chase the target are other things, and none of them has a published weight. 8 That bridge between target and measure is my reading, and the book flags it as such. 8
What the manual calls YMYL
The manual splits sensitive topics into four groups. Health or safety. Financial security. Government, civics, and society. And a group for other harms to people’s well-being. 2
It treats YMYL as a spectrum. Tsunami evacuation routes are clearly YMYL. A weather forecast might be. Music award winners, probably not. 2 How often to replace a toothbrush also falls outside. It is a casual health topic people talk over with friends, and slightly imperfect advice is unlikely to cause harm. 2
The sentence that most reassures a site owner is on page 11. Many topics, maybe most, are not YMYL. They do not require a high level of accuracy or trust to prevent harm. 2
Inside YMYL, a table in the manual separates experience from expertise. A review of a retirement savings service, written by someone who used it, counts as experience. Advice on how much to save and where to invest for retirement is better coming from an expert. 2 The same yardstick, applied to medical practices and accounting firms, is in the two pieces linked at the start.
What the industry sells as optimizing for E-E-A-T
On September 14, 2026, I read the first page for “e-e-a-t” on Google in Brazil and in Semrush’s Brazil database. 1 6 Together, the two readings gave twelve addresses. Two belonged to Google itself. One was a Brazilian maker of high-voltage test equipment that happens to share the acronym. That left nine industry texts, all in Portuguese. 6
The nine look alike in how they define the acronym. When it comes to ranking, they split. Four repeat that E-E-A-T is not a direct factor. 6 One says applying E-E-A-T directly impacts your rankings. Another says the guidelines exist to determine how pages rank. A third quotes Google’s statement and concludes, in the very next sentence, that the acronym influences rankings. 6
The plan repeats too. Five of the nine ask for a bio or an author page. Three ask for HTTPS and a privacy policy. One tells you to prove your reputation with badges and awards. 6 Each item is worth checking against the documents.
Byline and author page
This item has a document behind it. The helpful content page asks three questions about who wrote the content. Is it self-evident to your visitors? Do pages carry a byline where one might be expected? Does the byline lead to more information about the author? 5 And Google writes that it strongly encourages accurate authorship information, where readers might expect it. 5
The documentation for article structured data goes further. For the author field, it asks for a link to a page that uniquely identifies that person, like a bio page. And it says Google uses that address to tell one author from another. 9 The page talks about understanding and representing the author. Ranking does not come up in it. 9
Bio
The manual tells raters where to look for signs of E-E-A-T. There are three places. What the site and its authors say about themselves. What others say about them, in independent sources. And what shows on the page itself, with comments and reviews. 2
The About page and the author profile belong to the first place, as a starting point. 2 The same passage warns about conflicts of interest. A review written by the product’s own manufacturer counts for less, for that reason. 2 A bio the site writes about its own people is the site talking about itself. It opens the conversation, and the confirmation comes from outside.
Badges and awards
I searched the manual for the words badge and seal. Neither one appears in its 182 pages. 2 Awards show up somewhere else. In reputation research, the manual tells raters to look for news articles and other pieces written about the business. Awards and other forms of recognition can appear inside those texts. 2
The difference lies in who is talking. An award reported by a newspaper is someone outside confirming. A badge you put in your own footer is the site talking about itself.
HTTPS and privacy policy
The manual cites security as an example of the kind of trust each page needs. An online store needs secure payment and reliable customer service. 2 The manual ties trust to the type of page. Nothing I read for this piece turns a technical item into an E-E-A-T score. 2 5
And the small site?
Here the manual says something none of the nine texts say. A small site may have little or no reputation information on the web. That does not indicate high or low quality. 2 Many small local businesses live on word of mouth. When reputation is missing, the rater pays extra attention to the other criteria, especially on YMYL topics. 2
Where E-E-A-T meets entity
Picture the byline on a newspaper column. Under the headline sits a name. At the bottom, a line says the columnist is an engineer who writes on Tuesdays. If you don’t know her, you type her name into a search engine. In two minutes, you find out whether she exists outside that newspaper.
The machine walks the same path, with other tools. A loose name in a byline is only text. A name that leads to a single page, linked to records outside the site, becomes an identifiable thing. This is called an entity.
Three documents point to this path, each from its own side. The manual tells raters to check what others say about the author. 2 The article documentation asks for the address that identifies the author, and says it helps tell authors apart. 9 And the internal documentation leaked in 2024 has attributes that tie documents to authors treated as identifiable things. 8 The leak shows that the attributes exist. Their weight, nobody outside Google knows. 8
In practice, the path has four stops. The byline on the page. The author page inside the site. The person’s official home, at an address that does not change, following the rules of the entity home. And the outside sources that confirm who she is.
This piece follows the same path. It goes out under my name. My page on this site describes what I do at SEO Genome. It points to my official home, at an address of its own, on a site written in Portuguese.
What Google wrote, side by side with what gets sold
| What gets sold | What Google documents | Where it ends |
|---|---|---|
| An E-E-A-T score | Not a specific ranking factor 5 7; no such score in the leak 8 | There is no score to raise |
| A rater who judges your site | No single rating changes a position 2 3 4 5 | The ratings measure whether search works |
| Byline and author page | Recommended where readers expect them; the author’s address helps disambiguate 5 9 | Weight not published |
| Bio | The rater’s starting point, added to what others say 2 | On its own, the site talking about itself |
| Badge in the footer | The manual has no badges; awards appear in third-party coverage 2 | A badge of your own is no independent source |
| High E-E-A-T everywhere | Many topics are not YMYL; the trust needed depends on the page 2 | The bar rises with the risk |
The question for the next spreadsheet
Go back to the spreadsheet with the 6 out of 10. Ask where the number came from. If the answer is Google, ask for the page. Google writes that this factor does not exist. 5 7
Then ask what changes with each item in the plan. Byline and author page have a document behind them. A bio on its own is the site talking about itself. A badge in the footer does not appear in the manual. 2 5 9
Now try an exercise that depends on no proposal at all. Open the three most important pages on your site. For each one, write three things on a sheet of paper. Who signs it. Where the byline leads. And one source outside your site that confirms who that person is.
A blank line is work to do, in the order it appears. The rest of this section starts on our semantic SEO page.
From now on, when someone mentions an E-E-A-T score, you will ask for the document. And the three-page sheet starts today, before the next meeting.
Frequently asked questions
Is E-E-A-T a ranking factor?
Google writes that E-E-A-T itself is not a specific ranking factor. Its systems use a mix of factors that help identify content with these qualities, with more weight on YMYL topics. 5 The SEO Starter Guide answers no. 7
Can a Google rater tank my site?
The manual says no single rating directly changes how a page appears, or moves a site up or down. 2 Google uses the ratings to learn whether its systems are working. 5
Does every site need high E-E-A-T?
The manual says the type and amount of trust depend on the page. It also says many topics, maybe most, are not YMYL. 2 The bar rises with the risk of the topic.
Does a small site with no press coverage get rated poorly?
The manual says a small site may have little or no reputation information, and that this does not indicate high or low quality. 2 In that case, the rater pays extra attention to the other criteria, especially on YMYL topics. 2
Does an author page improve my ranking on Google?
Google recommends accurate authorship information where readers expect it. It uses the author’s address in the markup to tell authors apart. 5 9 How much that weighs in ranking, it does not publish. 8
Notes and sources
- Live reading of the Google results page in Brazil for “e-e-a-t,” September 14, 2026, at https://www.google.com.br/search?q=e-e-a-t&hl=pt-BR&gl=br&pws=0, in the author’s own browser tab. At the top, the AI Overview, labeled “Visão geral criada por IA” in the Portuguese interface, with the passages “é um conjunto de critérios usado pelo Google para avaliar a qualidade e a credibilidade de conteúdos na web” (is a set of criteria used by Google to evaluate the quality and credibility of content on the web), “Embora não seja um fator direto de ranqueamento algorítmico isolado, o E-E-A-T orienta os sistemas automatizados do Google a premiarem páginas” (Although it is not a direct, standalone algorithmic ranking factor, E-E-A-T guides Google’s automated systems to reward pages) and the offer “Como otimizar o seu site para o E-E-A-T” (How to optimize your site for E-E-A-T). Organic results, in order: Conversion, Google’s helpful content page in Portuguese, EEAT Brasil (a Brazilian maker of high-voltage testing systems, unrelated to SEO, https://eeat.com.br/), Mailchimp, Google’s December 2022 post, Fresh Lab, Semrush, Rankme, and a video from liveSEO. The “People also search for” box listed “Eeat saneamento” (sanitation), “Eeat sabesp” (Sabesp is São Paulo’s water utility), and “Eeat estação elevatória” (pumping station). Accessed September 14, 2026.
- Google, “Search Quality Rater Guidelines” (General Guidelines), version of September 11, 2025, 182 pages, https://static.googleusercontent.com/media/guidelines.raterhub.com/en//searchqualityevaluatorguidelines.pdf. Accessed September 14, 2026. Passages, with the PDF page: p. 6, “No single rating can directly impact how a particular webpage, website, or result appears in Google Search, nor can it cause specific webpages, websites, or results to move up or down on the search results page” and “humans could never individually rate each page on the open web”; p. 11, the four categories “YMYL Health or Safety”, “YMYL Financial Security”, “YMYL Government, Civics & Society”, and “YMYL Other”, and “Many or most topics are not YMYL and do not require a high level of accuracy or trust to prevent harm”; p. 12, the table with “Evacuation routes for a tsunami”, “Weather forecast”, “Music award winners”, and “How often to replace a toothbrush”, explained as “a casual health topic people commonly discuss with friends”; p. 23, reputation research in “articles, references, recommendations by experts”, which may include “awards and other forms of recognition”; p. 25, “small websites may have little or no reputation information. This is not indicative of high or low quality”, “Many small local businesses or community organizations have a small ‘web presence’ and rely on word of mouth”, and “pay extra attention to other PQ considerations, especially when assessing pages on YMYL topics”; p. 26, “The most important member at the center of the E-E-A-T family is Trust”, “The type and amount of Trust needed depends on the page”, “Online stores need secure online payment systems and reliable customer service”, the definitions of Experience (product review), Expertise (“home electrical rewiring advice from a skilled electrician or from an antique homes enthusiast”), and Authoritativeness (“known as a go-to source for the topic”, with the passport example), and “Experience, Expertise, and Authoritativeness may overlap”; p. 27, “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”, the example “a financial scam is untrustworthy, even if the content creator is a highly experienced and expert scammer”, the three sources “What the website or content creators say about themselves” (with the “About us” page or the author profile “as a starting point”), “What others say about the website or content creators”, and “What is visible on the page”, the warning about “a clear conflict of interest” with the manufacturer’s reviews, and “If a page is untrustworthy for any reason, it has low E-E-A-T”; p. 28, the table “YMYL Topics: Experience or Expertise?”, with “Saving for retirement”. A literal search for “badge” and “seal” in the PDF text: no occurrences. The manual guides human raters and is not a list of ranking factors.
- Mimi Underwood, “Updating Our Search Quality Rating Guidelines,” Google Search Central Blog, November 19, 2015, https://developers.google.com/search/blog/2015/11/updating-our-search-quality-rating. Accessed September 14, 2026. Passages: “Ratings from evaluators do not determine individual site rankings” and “In 2013, we published our human rating guidelines”.
- Elizabeth Tucker, “Our latest update to the quality rater guidelines: E-A-T gets an extra E for Experience,” Google Search Central Blog, December 15, 2022, https://developers.google.com/search/blog/2022/12/google-raters-guidelines-e-e-a-t; Portuguese version, “Nossa última atualização das diretrizes para avaliadores de qualidade: o E-A-T recebe um E adicional de Experiência,” https://developers.google.com/search/blog/2022/12/google-raters-guidelines-e-e-a-t?hl=pt-br. Accessed September 14, 2026. Passages: “Now to better assess our results, E-A-T is gaining an E: experience”; the tax return example, “content produced by an expert in the field of accounting”, and the tax software example, “a forum discussion from people who have experience with different services”; “These are not fundamentally new ideas”; and “they don’t directly influence ranking”. The Portuguese version says the same: “não influenciam diretamente a classificação” (do not directly influence ranking).
- Google Search Central, “Creating helpful, reliable, people-first content,” https://developers.google.com/search/docs/fundamentals/creating-helpful-content, page last updated December 10, 2025; Portuguese version, “Como criar conteúdo útil, confiável e que prioriza as pessoas,” updated December 18, 2025. Accessed September 14, 2026. Passages from the section “Get to know E-E-A-T and the quality rater guidelines”: “they identify a mix of factors that can help determine which content demonstrates aspects of experience, expertise, authoritativeness, and trustworthiness, or what we call E-E-A-T”; “Of these aspects, trust is most important. The others contribute to trust, but content doesn’t necessarily have to demonstrate all of them”; “While E-E-A-T itself isn’t a specific ranking factor, using a mix of factors that can identify content with good E-E-A-T is useful”; “our systems give even more weight to content that aligns with strong E-E-A-T for topics that could significantly impact the health, financial stability, or safety of people”; “We call these “Your Money or Your Life” topics, or YMYL for short”; and “Search raters have no control over how pages rank. Rater data is not used directly in our ranking algorithms”. From the section “Who”: “Is it self-evident to your visitors who authored your content?”, “Do pages carry a byline, where one might be expected?”, “Do bylines lead to further information about the author or authors involved”, and “We strongly encourage adding accurate authorship information, such as bylines to content where readers might expect it”. The Portuguese version gets the acronym wrong. Its list reads “aspectos da experiência, autoridade e reconhecimento, o que chamamos de E-E-A-T” (aspects of experience, authority, and recognition, what we call E-E-A-T), and “expertise it shares” becomes “experiência compartilhada” (shared experience), so expertise turns into experience. That is why this page cites the English.
- SEO Genome survey, September 14, 2026: the top ten organic results in Semrush, Brazil database, for “e-e-a-t” (Conversion, eeat.com.br, Google’s 2022 post in Portuguese, Mailchimp, Fresh Lab, HubSpot, Olivas Digital, Adtail, Semrush, and SEO Happy Hour), added to those in note 1, for a total of twelve addresses. Two are Google’s (notes 4 and 5), and one is EEAT Brasil, unrelated to SEO. The nine industry texts, all in Portuguese, were downloaded and read the same day: https://www.conversion.com.br/blog/o-que-e-eeat/ (March 10, 2025, updated February 22, 2026); https://mailchimp.com/pt-br/resources/google-eeat/; https://www.freshlab.com.br/blog/eeat-mais-autoridade-no-google/ (May 13, 2025); https://br.hubspot.com/blog/marketing/eeat (September 14, 2023); https://www.olivas.digital/blog/marketing-digital/eeat-google-como-funciona/ (February 11, 2026); https://www.adtail.ag/post/google-eeat; https://pt.semrush.com/blog/eat-ymyl-diretrizes-google/ (September 22, 2023); https://www.seohappyhour.com/glossario/eeat-seo/ (April 9, 2024); and https://www.rankme.com.br/seo/e-e-a-t/ (April 9, 2024). Saying it is not a direct factor: Mailchimp, Semrush, SEO Happy Hour, and Rankme. Fresh Lab: “impactando diretamente no seu ranqueamento” (directly impacting your rankings). HubSpot, on the guidelines: “Elas servem para avaliar a qualidade das páginas na internet e determinar o seu ranqueamento” (They serve to evaluate the quality of pages on the internet and determine their ranking). Mailchimp: “Embora o Google tenha declarado que o E-E-A-T não é um fator de classificação direto, ele influencia os rankings de busca” (Although Google has stated that E-E-A-T is not a direct ranking factor, it influences search rankings). Bio or author page: Conversion, Mailchimp, Fresh Lab, Adtail, and Semrush. HTTPS and privacy policy: Conversion, Mailchimp, and Fresh Lab. Conversion: “comprove a sua boa reputação, através de selos e/ou premiações” (prove your good reputation through badges and/or awards). Dating the manual to 2019: Conversion (“Em 2019, o buscador apresentou o Search Quality Evaluator Guidelines,” In 2019, the search engine introduced the Search Quality Evaluator Guidelines) and Fresh Lab (“documento que, desde 2019,” a document that, since 2019). None of the nine cites page 6 of the manual, the three sources on page 27, the small-site rule on page 25, or the author documentation for article structured data. Accessed September 14, 2026.
- Google Search Central, “SEO Starter Guide,” https://developers.google.com/search/docs/fundamentals/seo-starter-guide, page last updated December 10, 2025. Accessed September 14, 2026. In the list of things not worth focusing on: “Thinking E-E-A-T is a ranking factor”, with the answer “No, it’s not.”
- Daniel Sócrates, SEO de Entidade (published in Portuguese), 2026, chapter 7, “Confiança não é opinião. É cálculo” (Trust is not an opinion. It is a calculation). The source for the absence of an E-E-A-T score in the internal documentation leaked in 2024, the family of author signals that ties documents to authors treated as identifiable things, the rule that none of these measures has a known weight, and the reading of E-E-A-T as a target chased by computable measures, flagged in the chapter as the author’s inference.
- Google Search Central, “Article (Article, NewsArticle, BlogPosting) structured data,”
https://developers.google.com/search/docs/appearance/structured-data/article, page last updated
September 8, 2026. Accessed September 14, 2026. Passages: on
author.url, “A link to a web page that uniquely identifies the author of the article. For example, the author’s social media page, an ‘about me’ page, or a bio page” and “Google can understand both sameAs and url when disambiguating authors”; and, in the best practices, “To help Google better understand who the author is, we strongly recommend using the type and url (or sameAs) properties”. The page does not address position in the results.
Want to see SEO Genome more often on Google? Add it as a preferred source.
Let us look at what the machine understands about your company
A diagnostic conversation, with no slide deck. You bring your website address. We bring what Google and the AI assistants already know about it today.