A core update isn’t a penalty
Structure and markup refreshed for current answer engines; the original analysis is preserved.
The May core update finished rolling out this week, and the instinct is to ask what you did wrong. But a core update is not a penalty for a specific mistake — it is a re-evaluation of relative quality, and Google is explicit that dropped pages have nothing to fix. The only response that works is becoming a more authoritative answer, especially in YMYL.
the short answer
A core update is not a penalty for a specific mistake. Google is explicit: dropped pages have nothing to fix, and a decline does not mean your site is bad. It is a re-evaluation of relative quality across the whole web — you can fall because everyone was reassessed. Proof: some sites recovered with no changes. The only response that works is becoming a more authoritative answer — and in YMYL, where the bar is highest, that is both protection and the point.
key takeaways
- The May core update finished rolling out this week. The instinct is to ask what you did wrong — but that is the wrong question.
- A core update is not a penalty for a specific mistake. Google is explicit: dropped pages have nothing to fix, and a decline does not mean your site is bad.
- It is a re-evaluation of relative quality across the whole web. You can fall because everyone was reassessed, not because you broke a rule.
- Proof it is not a penalty: some sites recovered with no changes at all, simply because the reassessment landed differently this time.
- The only response that works is becoming a more authoritative answer — and in YMYL fields, where the bar is highest, that is both protection and the point.
the misreading vs the reality
The left column is where the penalty framing leads: a hunt for the offence and a fix that does not exist. The right column is what actually happened — a re-rating of where you stand. The recoveries with no changes are the tell; a real penalty could not lift on its own.
Why the framing is the whole game
Almost every mistake people make after a core update comes from one word: penalty. If you believe you were penalised, you believe you did something specific and can undo it, so you go looking for the offence — the page, the tactic, the technical sin — and you start cutting, persuaded that somewhere in your own site is the lever that will undo the drop. Often you cut things that were perfectly fine, because there was no offence to find, and you destabilise a site that was merely re-rated, trading a recoverable dip in the rankings for the kind of self-inflicted, lasting harm no algorithm asked for. The penalty framing also breeds a particular impatience: if this is a punishment, surely there is a fix, and surely it should work now. When it does not, the temptation is to cut more, and the site that started the week merely re-rated ends it genuinely damaged by its owner’s own hands. The first and most valuable thing to do after a core update is therefore not technical at all. It is to reject the penalty story, accept that there may be nothing wrong with the pages at all, and stop yourself from thrashing while you still have something worth protecting.
Replace it with the accurate framing and the right behaviour follows. A core update re-rates the relative quality of everyone at once, so your movement is a comparison result, not a verdict on a misdeed — the same page, unchanged, can rank differently simply because the field around it was re-scored. That is exactly what the no-change recoveries demonstrate: standing can improve without any action on your part because the comparison was simply rerun. Internalising this turns a frightening, mysterious event into a familiar one — you were re-ranked among your peers, as you will be again at the next update — and it points you at the only thing that actually moves relative standing over time, which is being a genuinely better and more trusted answer than the alternatives. The framing is not a comfort; it is the difference between working on the real lever and yanking on a phantom one.
The argument, in three parts
A core update re-rates rather than punishes; your ranking is a judgement relative to everyone else; and the only durable response is building more authority. Open each part.
01 It re-rates you; it does not punish you
The word that does the damage is "penalty", because it frames a core update as something you triggered and can therefore undo. That is not how these updates work, and Google has said so plainly: a page that drops has nothing specific wrong to fix, and a decline does not mean the site is bad or broke a rule. A core update is a broad reassessment of quality and relevance across the entire web at once — the system re-rates everyone against everyone, and your position is the output of that comparison. So you can fall not because you did something, but because the comparison was rerun and you came out differently this time. Holding that distinction matters, because almost every wrong response to a core update flows from the penalty framing: the frantic search for the offence, the stripping of things that were fine, the fix for a problem that was never there — all of it effort spent fighting a story that was wrong from the first word.
02 Your ranking is a judgement relative to everyone else
The thing a core update measures is not an absolute score your page either passes or fails; it is where you stand among the alternatives for what you cover. That is why the clearest evidence in this update is the sites that recovered without changing anything. If a drop were a penalty attached to your pages, no one could come back without addressing the supposed fault — yet here, sites that fell in earlier updates rose again on no changes, because the reassessment simply placed them differently. Read that correctly and it reframes the whole problem. You are not trying to clear a fixed bar; you are trying to be more clearly the best answer than the others competing for the same ground, and that standing is re-judged every time the system runs. The lever is not a setting on your site. It is your position in a field.
03 The only durable response is more authority
If the measure is relative standing and there is no switch to flip, the response that actually works is the slow one: become a more authoritative, more trustworthy answer than you were, and let the next re-evaluation find you improved. Google’s own guidance leads here — assess your content honestly against quality questions, ask whether it is original, whether it shows real expertise, whether you would stake your name on trusting it, and then raise the substance rather than the surface where the honest answer is no. This is unglamorous and it does not pay out on your schedule, since recovery comes with a future update rather than a deploy. But it is the only thing that compounds, because it is building the actual asset the system is trying to measure. Every core update is, underneath, the same question asked again: are you genuinely the source a reader should trust here? The work is making the honest answer yes.
Why YMYL raises the stakes
If you work in a YMYL field — Your Money or Your Life, the topics that touch health, finance, safety, and wellbeing — the relative judgement a core update makes is stricter, because Google’s guidelines demand a higher standard of expertise and trust where being wrong can genuinely hurt someone. Low-authority content in these areas tends to be moved harder by core updates than the same quality of content in a low-stakes niche, which means the weakness you can get away with on a hobby blog is precisely the weakness most likely to cost you when the subject is someone’s health or savings. That is not Google being capricious; it is the reasonable position that a thin, anonymous page about a medical condition or a retirement decision should not outrank a genuinely expert one, and the system is tuned to enforce it more firmly the higher the stakes.
The implication is bracing but clarifying: in YMYL, earned authority is not one ranking factor among many but the central one, and there is no substitute for it. You cannot mark up your way to trustworthiness on a topic where trust is the whole question; you have to actually be a credible, qualified, well-corroborated source, and to make that credibility visible. The upside is that the same scrutiny that punishes weak authority rewards real authority more richly, because in a field where readers and raters are looking hard for reasons to trust, genuine expertise stands out. The next core update will ask the YMYL question again — can this source be trusted on something that matters — and the work is making sure your honest answer is yes, on a topic where a wrong answer is the kind that actually costs a reader something.
What to do with this
After this update, resist the urge to diagnose a crime. Look at which pages and queries moved, by all means, but read it as information about where you stand relative to others, not as evidence of a fault to repair. Then put your effort into the only thing that changes relative standing: making your content genuinely more expert, better-evidenced, and more trustworthy than the alternatives, using Google’s own quality questions as an honest mirror. If you are in a YMYL field, treat authority as the priority rather than a nicety, because that is where the bar is highest and the movement largest. And then wait, because recovery arrives with a future re-evaluation, not with a deploy, and the gap between the two is where most people lose their nerve and start cutting again.
Most of all, hold the long view that the penalty framing destroys. A core update is not a verdict you appeal; it is a periodic re-rating of how authoritative you are compared with everyone else covering your topics, and the way to do well across many of them is not to win any single one of them but to keep becoming, in the quiet stretches between updates, a better answer than you were the last time the question was asked. That is slow, it is unglamorous, and it is the only thing that compounds — which is why being the genuinely trusted, well-evidenced source is the groundwork the AC Group has built for 27 years, long before and long after any individual update.
How to read the movement without panicking
None of this means you ignore the data; it means you read it as information about your standing rather than evidence of a crime. Wait for a clean window first — Google points to the completion date as the earliest reliable comparison point, because mid-rollout numbers are noisy and a single bad day can send you chasing a ghost. Then look at patterns rather than incidents: which clusters of pages and queries moved, across which topics, on which devices and in which countries. A broad, even shift across a whole section says something different from a sharp drop on a handful of pages, and both are telling you where you sit relative to others, not which rule you broke. The goal of the read is orientation, not indictment — to understand the shape of your re-rating so you can invest where your authority is genuinely thin.
Resist two opposite errors as you do it. One is over-reading a single metric on a single day and rebuilding around noise; the other is refusing to look at all, on the grounds that nothing can be done. The truth sits between: the movement is real and worth understanding, but it is a comparison result that resolves slowly, so the right cadence is patient observation followed by substantive improvement, not a frantic loop of change and recheck. Compare the settled week after the update to the week before it began, note where you lost the most ground, and treat those areas as the honest signal of where a competitor is now seen as the better answer — then go and become the better answer there.
Core updates, plainly: quick answers
My rankings dropped after the core update. What did I do wrong?
Possibly nothing, and that is the part people find hardest to accept. Google has been explicit and consistent: pages that drop in a core update do not have a specific problem to fix, and a decline does not mean your site is bad or broke a rule. A core update is a broad re-evaluation of quality and relevance across the whole web, so your position can fall because the system reassessed everyone, not because you committed an error. Hunting for the one thing you broke is the wrong frame, and it usually leads to thrashing — stripping things that were fine, chasing a fix that does not exist. The honest first step is to accept that there may be no mistake, only a re-rating.
If there is nothing to fix, what am I supposed to do?
Become a genuinely better, more authoritative answer, and wait for the next re-evaluation. Because a core update rates you relative to everyone else, you do not climb back by patching a page; you climb back by being more clearly the most expert, trustworthy source for what you cover, so that when the system reassesses, it has more reason to prefer you. Google’s own advice points exactly here: review your content honestly against quality questions — is it original, does it show real expertise, would you trust it — and improve the substance. There is no quick lever because the thing being measured is not a setting; it is your standing, and standing is rebuilt slowly.
Why did some sites recover without changing anything?
Because the update re-rates relative quality, and relative quality can move in your favour without you lifting a finger. In this update, a number of sites that had dropped in earlier core updates came back despite making no significant changes, simply because the reassessment landed differently this time. That is the clearest proof that a core update is not a penalty tied to your specific pages: if it were, you could not recover without addressing the supposed offence. What it really shows is that your ranking is a judgement of where you stand among alternatives, and that judgement is revisited each update — which is why the durable work is raising your actual authority rather than reverse-engineering a single event.
Does it matter that we are in a YMYL field?
Yes, a great deal, and it raises the bar you have to clear. YMYL — Your Money or Your Life: health, finance, safety, the topics that can affect someone’s wellbeing — is held to a higher standard of expertise and trust in Google’s guidelines, and core updates tend to hit low-authority YMYL content harder. If you operate there, weak authority is not a small liability that surfaces occasionally; it is the thing most likely to move you in every core update. The flip side is that earned authority is worth more there too: in a field where trust is scrutinised, being genuinely, demonstrably expert is both the protection against the next update and the reason a reader should believe you at all, which are, when you look closely, really the same thing wearing two faces at once.
A note on sources and timing
This is written in mid-June 2022, days after the May core update finished rolling out. We have described it as Google framed it: a broad re-evaluation of quality across the web, after which dropped pages have nothing specific to fix and a decline does not mean a site is bad, with recovery coming through future updates. We have noted the no-change recoveries reported by practitioners and the higher bar Google’s guidelines set for YMYL content. We have not predicted future updates. The durable point holds regardless of the next one: a core update re-rates your relative authority rather than penalising a mistake, and being a genuinely trusted, well-evidenced source is the groundwork the AC Group has built for 27 years.