Definition
E-E-A-T stands for experience, expertise, authoritativeness and trust. Google uses the term in the guidelines for its search quality raters to describe what good content looks like. E-E-A-T itself isn’t a specific ranking factor. Google says trust is the most important of the four.
Key takeaways
- E-E-A-T stands for experience, expertise, authoritativeness and trust. Google says trust is the most important of the four.
- Google says E-E-A-T itself isn’t a specific ranking factor. Its systems use a mix of factors meant to identify good content.
- Search quality raters assess pages against this framework. Their ratings don’t move any page up or down; Google uses them as feedback.
- On topics that affect health, money or safety (YMYL), Google’s systems give even more weight to content with strong E-E-A-T.
- Who, How and Why help you check whether it’s clear who created a piece of content, how it was made and why it exists.
What do the four letters in E-E-A-T stand for?
E-E-A-T stands for experience, expertise, authoritativeness and trust. Google uses it to ask whether someone knows a topic firsthand, has the necessary knowledge and is a go-to source. Trust asks whether content is accurate, honest, safe and reliable. Google calls this fourth part the most important.
The acronym comes from the Search Quality Evaluator Guidelines (external site), the instructions Google gives its search quality raters. Google also calls them the search quality rater guidelines. The current edition is dated September 11, 2025.
On December 15, 2022, Google added experience to E-A-T (external site). For some questions, Google says, what you value most is content from someone with firsthand experience of the topic. To fill out a tax return correctly, you’d probably want an accounting expert. To choose tax software, you might look for what other users have experienced.
The spelling EEAT means the same thing, and Google has also written “Double-E-A-T.” The final T stands for Trust in the guidelines and for trustworthiness in Google’s documentation on helpful content (external site).
The examples in the table are ours; the questions come from the guidelines. The numbers in the “Sources” column refer to the list of sources at the end of this article.
| Part | What raters ask | Example from a small business | Sources |
|---|---|---|---|
| Experience | Does the person know the topic firsthand? | A roofer posts photos of a repair she did and walks through the job. | 2, 3 |
| Expertise | Does the person have the necessary knowledge or skill? | A tax firm explains a filing deadline and cites the legal basis. | 2 |
| Authoritativeness | Is the person or website known as a go-to source for this topic? | A business’s own website is the definitive source for its opening hours. | 2 |
| Trust | Is the page accurate, honest, safe and reliable? | The address and service details are correct, and checkout is secure. | 1, 2 |
Is E-E-A-T a ranking factor?
No. Google says E-E-A-T itself isn’t a specific ranking factor. Its systems use a mix of factors meant to identify content with good E-E-A-T. On topics involving health, money or safety, they give that content even more weight. Google doesn’t mention any single E-E-A-T score.
Google doesn’t list which factors make up that mix. So all you can check are the qualities themselves: Are your details correct, and can people tell who’s behind the site? Google says reading the guidelines may help you self-assess your content. It says they can also help you align it conceptually with the signals its automated systems use.
Some guides call E-E-A-T a ranking factor, and one even says Google names trust as the most important ranking factor. That blurs two separate statements. Google says trust is the most important part of E-E-A-T. Neither Google’s documentation on helpful content nor the guidelines call trust or E-E-A-T a ranking factor.
Our guide to generative engine optimization (GEO) compares how each system picks the sources for its AI answers.
How do Google’s quality raters assess E-E-A-T?
External quality raters assess pages against the published guidelines. They look at what a website says about itself, what others say about it and what’s visible on the page. Their ratings don’t change where individual pages rank. Google uses them as feedback on its systems.
Google says it works with external Search Quality Raters (external site). They have no control over how pages rank, and their data isn’t used directly in the ranking algorithms. Google compares that data to the feedback cards a restaurant collects from diners.
Section 3.4 of the guidelines names three kinds of evidence:
- What the website or its authors say about themselves. The starting point is the “About us” page and the author’s profile.
- What others say. That includes independent reviews, references and news articles.
- What’s visible on the page. Often the main content alone shows how much experience and expertise went into it.
A site’s own claims are only a starting point. A source with a clear conflict of interest may not be trustworthy, and a manufacturer reviewing its own product is one example.
Deception earns the lowest rating. That covers made-up author profiles and false claims about credentials or expertise. On Your Money or Your Life (YMYL) pages and other pages that require trust, a complete lack of information about who is responsible for the website and its content also earns the lowest rating.
Does E-E-A-T apply to AI answers?
Google has no separate E-E-A-T rules for AI Overviews or AI Mode, and it says there are no additional requirements to appear in them. Both features retrieve pages through Search’s core ranking systems. Here too, Google recommends helpful, reliable content that draws on your own experience.
According to Google’s page on AI features (external site), the SEO best practices for Search remain relevant for AI features. Google’s guide to generative AI features (external site) contrasts a firsthand review with a mere summary: the review offers a unique perspective, while the summary only restates what’s already out there.
E-E-A-T is Google’s term. It doesn’t appear in the Bing Webmaster Guidelines (external site). There, Microsoft asks for content that’s original and authoritative. It warns that thin, ad-heavy or affiliate-only pages may lose ranking eligibility and visibility in AI answers.
Step 4 of our how-to shows how to build pages for those answers: Do your pages answer your customers’ questions up front?
When does E-E-A-T matter for a small business?
E-E-A-T matters most when your pages touch on health, finances, safety or the well-being of society. Google calls these topics YMYL. In our view, that’s why a tax firm needs more evidence than a café. On any website, though, it has to be clear who’s behind it and whether the details are correct.
YMYL stands for “Your Money or Your Life.” Google uses it for topics that could significantly affect people’s health, financial stability or safety, or the well-being of society. Not all content has to show all four qualities. Some content is helpful because of the experience it shows, other content because of its expertise.
These steps work for a small business too:
- Name the author. Put a real name on every article and link it to a page about that person’s background and areas of work. At a small business, that’s often the owner.
- Show your own work. Describe real jobs you’ve done, with photos, figures and dates. Google specifically asks about firsthand expertise.
- Cite your sources. Link to the laws, standards and manufacturer specifications you rely on.
- Keep dates honest. Change a date only when the content has changed. Google calls a new date without a substantial change a warning sign.
- Get an outside opinion. Have someone unaffiliated with your business read your content. Google recommends that too.
- Don’t make anything up. Google treats made-up names, AI-generated headshots and false credentials as deception.
If a tool offers an “E-E-A-T score,” that’s the vendor’s own estimate. No such metric appears in Google’s documentation or in the guidelines.
What are Google’s Who, How and Why questions?
Who, How and Why are three self-assessment questions from Google’s documentation. Who asks who created a piece of content. How asks how it was produced and what role AI played. Why asks what it’s for. Google says Why is perhaps the most important, because content should primarily help people.
- Who: Readers should be able to see right away who wrote the content. Google encourages bylines that lead to more information about the author.
- How: Readers should be able to find out how a piece was produced. For product reviews, that means the number of products tested, the results and how the tests were run. Google says an AI disclosure is useful where someone might wonder how the content was created.
- Why: Content should primarily help people and be useful to visitors who come to your site directly. Using AI mainly to manipulate rankings violates Google’s spam policies.
To describe your company and its people unambiguously for machines as well, see our article on Entity SEO.
Google allows AI content. Appropriate use isn’t against its guidelines (Google’s post on AI content (external site)). Even so, Google says it’s critical to fact-check and review all AI-generated content by hand for accuracy and trustworthiness before publishing (Google’s guidance on generative AI (external site)).
How does E-E-A-T differ from YMYL, page quality and PageRank?
E-E-A-T describes how trustworthy a piece of content and its creator are. YMYL describes how sensitive the topic is. Page quality is the overall rating raters assign, and E-E-A-T feeds into it. PageRank is a link analysis method and one of Google’s core ranking systems.
The numbers in the “Sources” column refer to the list of sources at the end of this article.
| Term | What it is | Who applies it | Role in ranking | Sources |
|---|---|---|---|---|
| E-E-A-T | four qualities of content and its creator | raters assess them; Google’s systems use a mix of related factors | not a specific ranking factor in itself | 1, 2 |
| YMYL | topics that could significantly affect health, financial stability, safety or the well-being of society | raters classify the topic | Google’s systems give strong E-E-A-T even more weight here | 1, 2 |
| Page quality rating | overall rating of a page on a scale from Lowest to Highest | raters only | data isn’t used directly in the ranking algorithms | 1, 2 |
| Helpful content | a separate system until it became part of the core ranking systems in March 2024 | Google’s systems | part of the core ranking systems | 5 |
| PageRank | a method that analyzes how pages link to each other | Google’s systems | part of the core ranking systems | 5 |
What’s specific to Germany and the EU?
Google has no separate E-E-A-T rules for Germany or the EU, but it tells raters to represent the cultural standards of their locale. In Germany, the legal notice (Impressum) identifies who runs a commercial website. It doesn’t say who wrote a particular article. For that, you need an author name and a profile page.
Legal notice: In Germany, providers of commercial digital services, typically paid ones, must keep their name, address and other details easy to find (Section 5 of the German Digital Services Act, in German (external site)). This usually goes in a legal notice called an Impressum. Google’s guidelines mention pages like “about us” and “contact us.” The word Impressum doesn’t appear in them.
Employee names: Talk to your employees before you put their names, photos or work histories on your website.
AI disclosure: Article 50(4) of the EU AI Act (external site) adds a separate disclosure duty for certain AI-generated text on matters of public interest. Among other exceptions, the duty doesn’t apply when the text has gone through human review or editorial control and someone holds editorial responsibility for it. Whether your content is covered is a question for a lawyer. Google’s How question is a separate matter.
From our work: how we show Who, How and Why on datafalke.de
We hold our own website to Google’s three questions. The table covers the 32 English articles in our sitemap at 12:45 a.m. German time on November 9, 2026. You can check everything in it without signing in; Mohammad Alboush’s profile and our article on website costs are two examples. The 32 German articles published at that time use the same author line, editorial note and change log.
| Google’s question | What the articles show | Limit |
|---|---|---|
| Who created it? | The header of each article names Mohammad Alboush as the author and links to his profile, which lists his role, focus areas, working languages, location and articles. | The profile is his own account of himself. It’s no substitute for what others say. |
| How was it made? | The editorial note on each article says: “Researched and drafted with AI assistance.” | The same note says he has not yet checked the sources, figures and claims. |
| How: what is it based on? | The articles end with a list of sources showing publisher, title, date and access date. Source lines give the publisher and date of a figure, for example “Source: Vercel, Dec 17, 2024 · vendor study.” | 31 of the 32 articles have a source list; the statistics page names the source of each figure in its table instead. 25 of the 32 have source lines. |
| How: what has changed? | The editorial note includes a log with the date, type and description of each change. Below it, you’ll see how to report a mistake. | The logs go back no further than Nov 7, 2026. Six of the 32 already show an update; the others list only the first publication and corrections. |
| Why does it exist? | The articles answer questions about websites and AI search. Each one ends with a related service and a suggested next step. | We sell services in these areas. No disclosure makes that conflict of interest go away. |
Measured · Python 3.12, 32 English articles from sitemap-en.xml fetched and analyzed
The second row is our biggest gap. Google says it’s critical to review AI content by hand. Yet all 32 articles still say: “Mohammad Alboush has not yet checked the sources, figures and claims.” That note will stay until he has. So the machine-readable part of our pages claims no review either: no article carries reviewedBy or lastReviewed.
The structured data lists the author under author, which points to the Person node for his profile.
{
"@type": "Article",
"@id": "https://datafalke.de/en/insights/llms-txt/#artikel",
"author": { "@id": "https://datafalke.de/team/mohammad-alboush/#person" },
"publisher": { "@id": "https://datafalke.de/#organization" },
"datePublished": "2026-11-08",
"dateModified": "2026-11-09"
}We don’t know whether any of this helps Google or AI systems rate datafalke.de more highly. The site has been online since November 7, 2026, and there’s no measurement log with data yet. And this page can’t show what others say about us.
To try this on your own site, open one of your articles and answer three questions. Does the author’s name lead to a profile? Does the article say how it was produced and what it’s based on? Would you still have written it if search engines didn’t exist?