The higher the stakes, the higher the bar

On August 1 Google rolled out what it described, in a brief and vague announcement, as a broad core update — no name, few specifics, the kind of low-key notice that undersold what followed. The community filled in the rest. As trackers logged which sites rose and fell, a pattern emerged: health, medical, wellness, and fitness sites had been hit disproportionately hard, and Barry Schwartz of Search Engine Roundtable gave the update the name it is still known by — Medic. The swings were large; some sites lost substantial shares of their traffic almost overnight, and weak health and financial pages were displaced by content from hospitals, institutions, and named, credentialed professionals. But the name flatters the wrong cause, because the update was not about medicine as a subject. It was a broad reassessment of content quality that simply landed hardest on one kind of content, and the clue to which kind lay in something Google had done a week earlier: it had updated its Quality Rater Guidelines — the manual its human evaluators use to judge how good search results are — sharpening its treatment of two linked ideas, E-A-T (expertise, authoritativeness, and trustworthiness) and YMYL (Your Money or Your Life). Read alongside the update, the guidelines tell a coherent story. Google was raising the quality bar most steeply for the pages where being wrong does the most damage, and rewarding the pages that could show they came from a credible, expert, genuinely trustworthy source. That is why this matters far past the health niche it was named for: it is a statement about how the bar itself works.

The concept that does the work here is YMYL — Your Money or Your Life — Google’s label for content that could significantly affect a person’s health, financial stability, safety, or well-being. What unites these topics is stakes: they are the ones where inaccurate, misleading, or poorly sourced information can cause real harm. A wrong answer about a medical symptom, a bad steer on a mortgage or an investment, unsafe instructions about a product or an emergency — these are not annoyances, they are risks to someone’s body, money, or safety. Google’s guidelines lay out several YMYL areas, among them health and medical topics, financial topics, safety, and other subjects bearing on major decisions or the welfare of groups of people. Two things are worth holding onto. First, it is the topic, not the industry, that decides: a general blog publishing health advice, an online shop selling supplements, an app explaining loans — all are handling YMYL content, whether or not they think of themselves as medical or financial outfits. Second, and most usefully, YMYL is a spectrum rather than a switch. A page about the best shows to stream this month carries almost no stakes; a page about how to treat chest pain at home carries extremely high ones; and a great many pages sit somewhere between. The same site routinely hosts pages at different points on that spectrum. So the question is never simply am I a YMYL site but rather where does this particular page fall, and the honest answer determines how high the bar will be.

Put the spectrum and the framework together and the principle becomes clear: the quality standard Google applies is not uniform — it scales with what a page puts at risk, and E-A-T is the lens through which the higher standard is applied. A point worth being precise about is that E-A-T is not a single ranking factor you can score or switch on. It is the framework Google’s human quality raters use to assess whether a page is the sort of result the algorithm ought to reward; the raters do not move your rankings directly, but their judgements help Google measure and tune whether its systems are surfacing trustworthy work. What the Medic update demonstrated is Google getting better at telling apart content that merely looks competent from content that actually issues from a credible, expert source — and weighting the latter far more heavily precisely where the stakes are high. The logic is almost moral, and almost obvious once stated: if a page can hurt someone when it is wrong, the engine should demand stronger evidence that it is right and that its author is qualified to say so. For a low-stakes page, looking helpful may be close enough; for a high-stakes one, looking helpful is not enough — the source has to be demonstrably trustworthy, because a confident, well-written, wrong answer about a drug interaction is more dangerous than no answer at all. That is why a hospital’s page outranks an anonymous wellness blog on the same query, even if both read well. Understanding that the bar rises with the stakes, and meeting it where it actually rises, is the earned-authority discipline the AC Group has worked by for {years} years.

Sort before you build. Take your important pages and place each on the stakes spectrum: clearly YMYL — health, money, safety, major decisions — clearly not, or somewhere between. The sorting matters because it tells you where to spend, and the goal is to match the investment to the stakes rather than gold-plate everything equally. For the high-stakes pages, the standard is demonstrable expertise and trust, and you meet it by being, and visibly showing that you are, a credible source. That means real authorship: put named people behind sensitive content and make their relevant qualifications visible, rather than publishing health or financial advice anonymously. It means sound sourcing: cite primary, authoritative references, and where the stakes justify it, have the content created or reviewed by genuinely qualified experts. It means the trust signals a careful reader and a careful algorithm both register — accurate and current information, a clear account of who you are and how to reach you, transparent editorial standards. And it means holding accuracy to a tighter tolerance than you would bother with for low-stakes content, because the cost of error is higher. For genuinely low-stakes pages, you can ease off proportionally; the discipline is proportion, not maximalism. One firm caution: there is no quick fix. Google has said plainly that recovering from a quality reassessment comes from genuinely improving over time and is recognized in later core updates, not from a hurried patch in the days after a drop. Building real, visible authority in proportion to what your topic puts at risk is the earned-authority discipline the AC Group has worked by for {years} years.