E-A-T (expertise, authoritativeness, trustworthiness)
The superseded three-letter form of what Google's rater guidelines now call E-E-A-T, after Experience was added in December 2022, and which Google states is not a ranking factor.
Karl-Gustav Kallasmaa, Founder & CEOLast updated E-A-T stands for expertise, authoritativeness and trustworthiness: the criteria Google's search quality raters were trained to assess when judging page quality. The term is superseded. In December 2022 Google added a second E, for Experience, and the concept is now written E-E-A-T in its own documentation and rater guidelines.
Two things about it are widely misstated, and Google contradicts both in writing.
It is not a ranking factor
Google's guidance on creating helpful content says it plainly: "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."
There is no E-E-A-T score inside the ranking system, and nothing on your page sets one. E-E-A-T is the target that Google's ranking systems are built to approximate — a description of the outcome, not a lever inside the machine. An audit that reports "your E-E-A-T score is low" is describing a number that does not exist.
Raters do not rank pages
Quality raters read search results and score them against the guidelines. Google states that "rater data is not used directly in our ranking algorithms. Rather, we use them as a restaurant might get feedback cards from diners."
The consequence is a timing one. Rater judgements are evidence about whether a change to the system made results better in aggregate. They never touch an individual URL's position, and no amount of writing for the guidelines causes a rater to intervene on your behalf.
The four aspects, and which one carries the weight
- Experience — first-hand or life experience with the subject. Added December 2022, and the aspect a purely researched page most often lacks.
- Expertise — the depth of knowledge behind the content.
- Authoritativeness — the degree to which the creator or site is a recognised source on the topic.
- Trustworthiness — accuracy, honesty, safety and reliability.
Google is explicit about the hierarchy: "Of these aspects, trust is most important. The others contribute to trust, but content doesn't necessarily have to demonstrate all of them." A page written by a genuine expert that hides who wrote it, carries no date, and asserts numbers with no source has demonstrated expertise and failed at trust — and trust is the one that matters.
What it means for a page trying to be cited by an agent
Nothing, mechanically. A retrieval step selects passages by similarity and access; there is no trust score in the pipeline, and Google's own AI-features documentation states there are no additional requirements or special optimizations for appearing in AI Overviews or AI Mode.
What does transfer is the substance the acronym points at, because it is the same substance a passage needs to survive being lifted out of your site:
- A named author with a real identity. A quoted passage with no attributable source is harder to reuse than one with a byline behind it.
- A visible date. A model deciding between two contradictory passages has almost nothing else to break the tie.
- A [source](/glossary/source-citation) next to every number. A statistic with its origin in the same sentence stays true when the sentence travels alone. A statistic whose source sits three paragraphs above does not.
- Qualifiers written into the claim. The same rule, applied to caveats: what is not in the passage does not go with the passage.
- [Structured data](/glossary/schema-markup) stating author, publisher and date in a machine-readable form, so the byline is not only prose.
That is the honest version of "optimising for E-E-A-T": it is not a score to raise, it is a set of publishing habits that happen to serve both a quality rater and the retrieval systems behind generative engine optimization, for different reasons.
Failure modes
- Treating it as a checklist. Adding an author box to a page whose claims are unsourced satisfies the form and not the substance. Trust is the aspect being assessed, and it is the one a checklist cannot fake.
- Citing the acronym as a factor. Any strategy premised on "raising E-E-A-T" is premised on a number Google says does not exist as a ranking input.
- Assuming raters can be appealed to. They score samples for system evaluation; they do not adjudicate individual pages.
- Using the old acronym as if nothing changed. Experience has been part of the framework since December 2022, and it is the aspect most often missing from research-assembled content.
Frequently asked questions
Is E-A-T still the right term?
No. Google added Experience in December 2022; the current form is E-E-A-T.
Is it a ranking factor?
No. Google states it is not a specific ranking factor, though systems try to identify content that has it.
Which aspect matters most?
Trust. Google says the other three contribute to it.
Do quality raters change my rankings?
No. Google states rater data is not used directly in ranking algorithms.
Terms related to E-A-T (expertise, authoritativeness, trustworthiness)
The Schema.org vocabulary and the JSON-LD you embed with it, what it is genuinely good for, and the limits of what it can make happen.
How an AI answer attributes what it says to the pages it read, and why a citation is a distinct outcome from a click or a mention.
Google's AI-generated summary at the top of a results page, and the snippet controls that decide whether your page can appear inside one.
The practice of getting a source reproduced inside an AI-generated answer, introduced as a named paradigm in a 2023 research paper.