Case Study 5.2 — The costume of trust: when faked E-E-A-T meets a core update

A labeled composite, not a single real company. The pattern it depicts is real and well documented: across Google's broad core updates from 2018 onward, thin affiliate and health-adjacent sites that wore the trappings of authority — author boxes, "medically reviewed" lines, citations — without the substance were repeatedly reassessed and demoted. We build one representative site, "SupplementSage," from those common, documented patterns to teach the limits of E-E-A-T signals. No specific traffic or revenue figures are asserted as real; the site is illustrative.

Background

SupplementSage was a health-and-wellness affiliate site. Its business model was ordinary for its niche: publish articles about supplements and health concerns, rank them, and earn commissions when readers clicked through to buy. For a while, it worked. The site published quickly and broadly, covering hundreds of "best supplement for X" and "is Y safe" topics — squarely YMYL territory, where health and safety are on the line.

Crucially, SupplementSage had read the SEO advice about E-E-A-T. So it did what that advice, misunderstood, told it to do. Every article got an author box with a name and a friendly headshot. Many carried a "Medically reviewed by Dr. ___" line. Articles were sprinkled with citations — footnote-style links to studies. On paper, it looked like a site that had taken trust seriously.

The search/SEO issue

The problem was that the trappings were a costume over a body that had none.

The bylined "authors" were freelancers with no particular expertise in the subjects; some bios were padded or vague. The "medical reviewer" line, in the composite, described a review that was cursory or, in the worst version of this pattern, effectively nominal — a name lent, not a rigorous process applied. The citations were frequently decorative: linked to sources that did not actually support the specific claim, or to other affiliate content. And the site's real purpose — moving readers toward a commission — quietly shaped which supplements got recommended, a conflict of interest the site did not clearly disclose.

Every one of those is a trust failure, and — as §5.5 argued — trust is the center of the framework. The site had spent its effort on the appearance of Experience, Expertise, and Authoritativeness while the load-bearing member, Trustworthiness, was hollow.

What it shows

This is the mirror image of Case Study 5.1. There, Google retired a feature but kept the principle. Here, a site adopted the features of E-E-A-T while ignoring the principle — and discovered that the features are not the thing.

It also shows why Google's design leans on signals it is hard to fake. Anyone can add an author box; not everyone can be an author whose independent reputation checks out. Anyone can write "medically reviewed"; a genuine review process leaves a different footprint — real credentialed reviewers, real dates, real corrections, accuracy that holds up. Anyone can drop in citations; only diligent content cites sources that actually support its claims. The gap between costume and substance is exactly the gap Google's quality systems, trained against human raters doing reputation research, are built to detect.

Outcome

In the composite — matching the documented pattern of the period — a broad core update arrived, of the kind Google rolls out several times a year to reassess quality across the web. SupplementSage's rankings fell sharply and did not recover on their own. There was no manual action to appeal, no single "error" to fix in Search Console: this was an algorithmic reassessment, the site's overall quality re-judged and found wanting, the way §5.6 describes being "outperformed on many signals at once" rather than handed a scarlet letter.

The owners' first instinct was the usual one: add more of the trappings — more author boxes, more citations, another "reviewed by" line. It did not work, because it could not. You cannot out-costume a trust problem.

The lesson

E-E-A-T signals reveal real trust; they cannot manufacture it — and on YMYL topics, the gap between the two is where sites (and readers) get hurt. Everything in §5.4 came with a stated limit for exactly this reason. An author box can make genuine expertise verifiable; it cannot create expertise. A citation can let a reader check a claim; it cannot make a false claim true. A "medically reviewed" label can signal a real process; it cannot substitute for one.

The honest recovery path for a site like this is slow and unglamorous, and it is the opposite of adding more costume: publish less, and only what you are genuinely qualified to publish; commission real experts and run a real review process; disclose conflicts of interest; correct errors; and let independent reputation accrue. That is not an SEO tactic bolted onto a broken site. It is becoming the kind of site the framework was designed to reward — which, as this whole book argues, is also just good publishing.

There is a limit worth naming for balance, in the spirit of stating what a lesson cannot claim. A single core-update decline is not, by itself, proof that a specific site "faked its E-E-A-T"; core updates reassess many things at once, and Google is explicit that a drop does not always mean anything is "wrong." What the pattern across many such sites supports is the narrower, sturdier claim we are making: that the appearance of trust, unbacked by the real thing, is a fragile foundation — and the higher the stakes of the topic, the faster that fragility shows.

Discussion questions

1. SupplementSage did everything the shallow "E-E-A-T checklist" advice recommended. What was missing, and why did the checklist fail it?
2. Why did adding more trust signals after the decline fail to help? What would you have advised instead?
3. This case is a labeled composite. What is gained and what is lost by teaching with a composite rather than naming a real casualty? (Connect to the citation-honesty tiers from the style of this book.)
4. The lesson section deliberately limits its own claim about causation. Why is that caveat important, and how does it model the evidence-tier habit from Chapter 2?
5. Contrast SupplementSage with the "good publisher" version of a YMYL health site (the WellPath example from §5.7). List three concrete practices that separate the two.