Pillar · Hub
Trust: the most important E-E-A-T dimension
Systems that detect manipulation and verify reliability
The question
Can a reader act on this, and is it clear who stands behind it?
What Google goes by
Trust is the foundation of E-E-A-T. Google's own Quality Rater Guidelines, the handbook for its human raters, state that Trust is the most important member of the E-E-A-T family. Trust is earned through reliable behavior — accurate content, honest linking, transparent authorship, and the absence of spam signals. More systems contribute primarily to Trust than to any other dimension.
What this means for your page
Straight from the system pages: what each system rewards or punishes in practice.
NavBoost cannot be manipulated directly, because it reflects genuine user satisfaction. In practice, titles and snippets in search results must deliver on their promise. A page about general climate served for 'storm Berlin tomorrow' risks immediate bounces and a worse ranking as a result. The best optimization is a page that genuinely answers the user's need.
Content should be prepared so that Google can include it in its features: structured data (Schema.org), clear widgets, timely reports with news value. Appearing in carousels and PAA boxes and generating clicks there earns the same ranking signals as organic web clicks.
Page optimization alone cannot improve this metric. It arises from genuine brand recognition and return-visit habits. App usage, newsletters, browser bookmarks and direct visits strengthen the Chrome signal far more than any SEO work on individual pages.
Q* does not react to individual page improvements. It requires consistent quality over months and years. Long-term content quality beats short-term SEO tricks. Pages with persistently low quality scores drag down the entire site's Q* value.
In the Wild
Not a lab: 8 brand advice pages from one industry, measured where they actually live, on September 16, 2026. Observations, not verdicts — anyone can repeat them.
2 of 8 meet all 2.
All 16, one by one
Patagonia — A date is visible to the reader ↗ — A date is visible to the reader: no
JSON-LD Article carries datePublished and dateModified 2026-09-08. No <time datetime>, no date in the text.
2026-09-16 · Gerendertes DOM
Patagonia — A date is machine-readable ↗ — A date is machine-readable: yes
JSON-LD Article carries datePublished and dateModified 2026-09-08. No <time datetime>, no date in the text.
2026-09-16 · Gerendertes DOM
Vaude — A date is visible to the reader ↗ — A date is visible to the reader: no
BlogPosting with datePublished 2023-01-20 and dateModified 2026-06-09. No visible date, no <time datetime>.
2026-09-16 · Gerendertes DOM
Vaude — A date is machine-readable ↗ — A date is machine-readable: yes
BlogPosting with datePublished 2023-01-20 and dateModified 2026-06-09. No visible date, no <time datetime>.
2026-09-16 · Gerendertes DOM
The North Face — A date is visible to the reader ↗ — A date is visible to the reader: no
No date: no <time datetime>, no datePublished, none in the text.
2026-09-16 · Gerendertes DOM
The North Face — A date is machine-readable ↗ — A date is machine-readable: no
No date: no <time datetime>, no datePublished, none in the text.
2026-09-16 · Gerendertes DOM
Jack Wolfskin — A date is visible to the reader ↗ — A date is visible to the reader: no
No date anywhere: no <time datetime>, no structured date, none in the text.
2026-09-16 · Ausgeliefertes HTML
Jack Wolfskin — A date is machine-readable ↗ — A date is machine-readable: no
No date anywhere: no <time datetime>, no structured date, none in the text.
2026-09-16 · Ausgeliefertes HTML
Deuter — A date is visible to the reader ↗ — A date is visible to the reader: no
No date in any form.
2026-09-16 · Ausgeliefertes HTML
Deuter — A date is machine-readable ↗ — A date is machine-readable: no
No date in any form.
2026-09-16 · Ausgeliefertes HTML
Nike — A date is visible to the reader ↗ — A date is visible to the reader: yes
The reader sees "15. April 2026". Google sees nothing: no JSON-LD at all, no <time datetime>.
2026-09-16 · Ausgeliefertes HTML
Nike — A date is machine-readable ↗ — A date is machine-readable: yes
The reader sees "15. April 2026", and so does a machine: itemprop="dateModified" carries 2026-04-15T16:59:23Z. No date in JSON-LD, no <time datetime>.
2026-09-19 · Ausgeliefertes HTML
Garmin — A date is visible to the reader ↗ — A date is visible to the reader: yes
The reader sees "9. September 2026". No JSON-LD, no <time datetime>.
2026-09-16 · Ausgeliefertes HTML
Garmin — A date is machine-readable ↗ — A date is machine-readable: yes
The reader sees "9. September 2026", and so does a machine: meta article:published_time carries 2026-09-09. No date in JSON-LD, no <time datetime>.
2026-09-19 · Ausgeliefertes HTML
Puma — A date is visible to the reader ↗ — A date is visible to the reader: no
No date, visible or structured.
2026-09-16 · Ausgeliefertes HTML
Puma — A date is machine-readable ↗ — A date is machine-readable: no
No date, visible or structured.
2026-09-16 · Ausgeliefertes HTML
SChecked on the page: a dated measurement of a live page. It can change tomorrow; the date says when it held.
What you can do
Google’s own Helpful Content best practices, filtered to this dimension.
Content for people, not primarily for search engines; no mass or automated production without added value; no content that exists only to capture traffic.
Helpful Content: People-first / Avoid search-engine-first · Helpful content system · Spam detection systems (incl. SpamBrain)
The content satisfies what someone is actually searching for and leaves the feeling of being well served — clear search intent met.
Helpful Content: Search intent & good experience · Neural matching · RankBrain · BERT · Passage ranking system
Cleanly produced: free of errors, not carelessly made, without intrusive advertising, usable on mobile.
Helpful Content: Presentation & production · Page experience system
Clear who created the content and why; the responsible party/author is identifiable and trustworthy; for YMYL: accuracy and reliability as the core E-E-A-T component.
Helpful Content: Who/How/Why + Trust · Reliable information systems
For time-sensitive topics, content is current and maintained/updated as needed.
Helpful Content (implicit) + QRG Freshness · Freshness systems
No manipulative practices: no keyword stuffing, no link schemes, no misleading techniques.
Spam Policies + Search Essentials · Spam detection systems (SpamBrain) · Penguin (integrated into Core Ranking) · Exact match domain system
Which practice belongs to which dimension is our assessment [C]. Both ends, the practice and the named systems, are documented by Google [O]. All eight, with the systems behind them →
The systems
Primary (11)
Systems where Trustworthiness is the primary E-E-A-T dimension.
NavBoost, the system that evaluates clicks from Google search, watches what users do after a search. Do they click on a weather page and stay, or jump straight back to Google? The last, longest click of a session is the strongest signal, because it shows the need was genuinely met. Real user behavior thus becomes a re-ranking signal, without Google having to read the content directly.
Glue is the counterpart to NavBoost for special Google elements like carousels, 'People also ask' boxes, or news features. It measures whether users interact with these elements: tapping a carousel, expanding an answer in a PAA box, or arriving at a page via a news feature. Instant Glue is the real-time variant for fast-moving events.
Chrome data show how often a site is accessed in the Chrome browser overall, regardless of whether someone arrived via Google search or typed the URL directly. High direct visits show that users trust the site enough to navigate there without asking Google. The metric measures brand popularity and willingness to return.
Q* is a summary quality score for the entire website. It works like a 'balance' made up of the site's authority (siteAuthority), a low-quality flag (lowQuality), and the overall NSR value. The score changes slowly: internal Google documents describe it as 'largely static'. It is a sluggish foundation, not a day-to-day signal.
Not a live signal used daily in ranking — but the foundation on which all quality models are calibrated. The IS-Score (Information Satisfaction) summarizes how well a page meets a user's information need, rated by trained human raters. The QRG (Quality Rater Guidelines) are the public rulebook by which these raters judge.
Google deducts points for three concrete quality and navigation deficiencies. navDemotion penalizes poor page guidance, such as unclear navigation or misleading structure. anchorMismatch penalizes links whose link text does not match the destination, for example 'more info' instead of 'Germany rain radar'. serpDemotion comes from negative user behavior in search results: users click, bounce straight back and click on competitors.
Google deducts quality points in three ways. clutterScore measures layout overload: too many ads, pop-ups or distracting elements that disrupt the information flow. scamness measures fraud proximity: patterns that resemble disreputable sites. unauthoritativeScore measures missing credibility: no author details and no recognizable institution behind the page.
Panda lowers the ranking of an entire website when too much of it consists of thin or low-quality content. It applies to the whole site, not just to individual pages. If too large a share of a site is rated as 'thin content' (content-poor pages), all pages of the site rank worse. BabyPandaV2 is a newer, more finely calibrated variant of the same principle.
This system detects manipulative link patterns in anchor text, the clickable text of a hyperlink. A page that is linked exclusively with exact keyword texts ('Berlin weather', 'free weather forecast') looks unnatural, like deliberate SEO manipulation. Penguin is the historic Google update that addressed this problem. Its logic lives on in today's system.
SpamBrain is Google's AI-based spam detection system. Its most practically relevant aspect is 'scaled content abuse': the mass production of content following the same pattern every time, without genuine editorial added value. A typical example is thousands of location pages that only plug variables (city name, coordinates, measurement data) into a template, with no human editorial quality check.
Google detects a 'date lie' by comparing three different signals for a page's date. (1) The visible date in the article (bylineDate) is easy to fake. (2) The date from URL, markup or timestamp (syntacticDate) is also easy to manipulate. (3) The date inferred from the text content (semanticDate) can only be changed through genuine content updates. When the three don't agree, Google sees a 'date lie'.
Flanking (9)
Systems that support Trustworthiness indirectly.
Deep Dives (23)
NavBoost: How Google Uses Click Behavior to Re-Rank Search Results
From DOJ testimony to leaked API documentation — the most important ranking signal you can't directly manipulate
Panda: The Site-Wide Quality System Named After Its Inventor
How Google's war on content farms produced a patent trail that leads directly to the update's name — and what the API leak reveals about its current form
Q* and P*: The Two Pillars of Google's Ranking Architecture
The DOJ trial revealed that Google's ranking ultimately reduces to two top-level signals — Quality (Q*) and Popularity (P*). This is what the evidence shows about how they work.
hostAge-Sandbox: Why New Domains Start at a Disadvantage
The patent that reveals how Google evaluates domain legitimacy — and why patience is the only strategy for new sites
Anchor-Spam: How Google Detects Manipulative Link Patterns
From 'miserable failure' Google bombs to the patent that reveals how context analysis catches anchor text manipulation
FreshnessTwiddler: How Google Decides When Fresh Content Matters
The query-dependent freshness system — and the patent that reveals how update frequency, staleness, and link freshness become ranking signals
Date triangulation: How Google Determines When a Page Was Really Published
The patent that reveals six different date sources Google uses — and why faking a byline date doesn't work
IS-Calibration: The Human Rater System That Trains Google's Quality Models
The hardest evidence in the entire matrix — DOJ/sworn material confirms the rater system that calibrates every quality signal Google uses
SpamBrain: Google's AI-Powered Spam Detection System
How machine learning identifies scaled content abuse, link spam, and manipulated rankings — and why 'AI-generated content at scale' is the new content farm
chard: Google's Page-Level Content Quality Model
The leak field that rates individual page quality — calibrated by human raters, feeding into NSR, and why YMYL pages face stricter standards
contentEffort: How Google Estimates Editorial Effort
The AI model that estimates how much editorial work went into a page — and why 'effort' is becoming a ranking signal
Glue: How Google Measures Interaction with Special Search Features
The NavBoost counterpart for carousels, 'People also ask' boxes, and news features — measuring whether users interact with special elements, not just blue links
Chrome signals: How Browser Data Complements Search Data
The signals that measure how often a site is accessed directly in Chrome — independent of Google Search
IP-Prior: How Google Protects Click Data from Manipulation
The system that detects unnatural click patterns and protects NavBoost from gaming — the integrity layer of behavioral ranking
clutterScore, scamness, unauthoritativeScore: Three Quality Demotions
The three signals that penalize layout overload, fraud proximity, and untrustworthiness — and why they combine into a single Trust signal
QualityBoost-Demotions: The Penalty System for Quality Deficits
navDemotion, anchorMismatch, and serpDemotion — three specific penalties for navigational, linking, and SERP quality issues
RankBrain and DeepRank: How AI Changed Google's Understanding of Queries
From keyword matching to semantic understanding — the AI systems that interpret what users mean, not what they type
Tangram: The Assembly Layer That Builds Search Results
Tangram evaluates nothing itself — it takes ratings from NavBoost and Glue and decides what appears on the results page, where, and in what order
tofu, keto, Rhubarb: The Subchunk Refinement Signals
Codenames for refinement signals at the most granular level, the individual subsections of a page. They are the weakest evidence in the matrix.
Site Reputation Abuse: What Google Suspended in Europe — and What It Did Not
What Google suspended in the EEA is the manual action — the replacement removes exactly what parasite SEO pays for
Google Update History: What a Timeline of Named Updates Can and Cannot Show
Google names roughly ten updates a year and ships thousands of changes. The timeline is real, dated and official — and it is a filtered excerpt of Google's own communication decisions
What Good Content Is: contentEffort and Google's Soft Metrics
One leaked field, one rater definition, twelve official questions, and a checklist that marks which parts are evidence and which are our reading.
Links That Count: Five Link Tactics Against the Leak and the Patents
A popular link guide reads PageRank buckets, trusted anchors and context hashes out of Google's leak. The fields are real. The trusted sources are news sites, the bucket counts onsite anchors, and the cap is 200, not zero.