Google Algorithm Updates: What to Do When Rankings Move

If your organic traffic moved sharply in the last few days and you can't point to anything you changed, the cause is usually one of two things: Google shipped an algorithm update, or something else did it and it only looks like one. Telling the two apart is the first job, and it matters more than anything you do next.

Here is the point to internalize before anything else: a core algorithm update is not a penalty. Nothing on your site was flagged, there is nothing to "lift," and there is no reconsideration request to file for it. When a core update moves your rankings down, it almost always means Google re-evaluated the whole landscape of pages competing for your queries and currently judges a different set of pages more relevant or more useful for that search — not that you broke a rule and are being punished for it.

What a Core Update Actually Changes

A core update is a broad change to how Google's ranking systems weigh relevance, quality and usefulness across a huge number of queries at once. It is not a bug fix, not a manual review of your specific site, and not a targeted response to something you did. Google adjusts its systems, reruns them against the entire index, and the result is a new ordering of results for a wide range of searches.

Because the change happens at the level of the ranking systems rather than at the level of an individual page, it often moves a whole site at once, even sections of it that didn't change. That single fact confuses a lot of site owners: nothing was edited, nothing was flagged, yet dozens of pages move together. That's the expected shape of a core update, not a sign that something is broken.

It helps to think of it as a reweighting rather than a new rule. Google already measures a huge range of things about a page and the site it sits on — how thoroughly a topic is covered, how the site is generally regarded, how a page compares to others answering the same query, and more. A core update doesn't necessarily introduce a new measurement; it can simply change how much those existing measurements count relative to each other. A page that ranked well partly on the strength of its backlink profile might slide if depth of coverage is weighted more heavily afterward, even though nothing about the page or its links changed at all.

Core Updates vs. Targeted Updates

Not every ranking shift labeled an "update" is a core update, and the difference changes what a sensible response looks like.

• Spam-focused updates target manipulative techniques directly — cloaking, scraped or auto-generated content, link schemes. These are narrower and usually tied to identifiable violations rather than a general relevance shift.
• Review-focused signals evaluate content that reviews products or services, specifically whether it shows evidence of real, firsthand testing and independent judgment rather than reformatted spec sheets or affiliate copy with no hands-on evaluation behind it.
• Helpful-content-style quality signals, now folded into Google's broader ranking systems, weigh whether content was made primarily to serve a reader searching for information, versus made primarily to attract search traffic first and answer the reader second.
• A general core update is broader than all three: it's a recalibration of relevance and quality assessment across the board, not a response to one specific pattern.
• The distinction matters for diagnosis, not just vocabulary. If a spam or review-focused update is behind a drop, there's usually a specific, checkable policy or pattern to look at — Search Console's Manual Actions and Security Issues reports are the first stop, along with Google's published spam and content policies. If it's a broad core update, there's no specific policy to check against, because nothing was violated; the work is comparative rather than corrective.

Why a Core Update Isn't a Penalty

A penalty implies a violation, a punitive action taken against you specifically, and a defined path to appeal it. That mechanism does exist at Google — it's called a manual action, applied by a human reviewer who determined a specific page or site violates Google's spam policies. Manual actions show up explicitly, by name, in Search Console under the Security & Manual Actions report, and that report is the only place a formal reconsideration request applies.

A core update ranking change produces no such notice, because it isn't that kind of event. Nobody reviewed your site and decided to sanction it. "Recovering" from a core update doesn't mean fixing a violation — there wasn't one — it means becoming more relevant and more useful, in Google's ongoing assessment, than whatever is currently outranking you. That is a meaningfully different problem to solve, and conflating the two is the single most common mistake site owners make after a rollout.

How to Tell an Update From Something That Only Looks Like One

Before you attribute a ranking or traffic move to an algorithm update, rule out the more mundane explanations. Each of these produces a graph that looks a lot like "we got hit," but none of them are a core update, and none of them are fixed by the same response.

• Seasonality — the query volume itself moves every year on roughly this schedule. Check the same period a year earlier before you assume anything algorithmic happened.
• A technical regression — an accidental noindex tag shipped in a recent deploy, a broken sitemap, a robots.txt rule that now blocks a section, a spike in server errors, or a redirect chain introduced by a migration. These are self-inflicted and typically show up as a sharp, page-specific drop rather than a broad one.
• A tracking or analytics change — a new consent-mode default, a re-platformed analytics property, a swapped tag manager container. Traffic didn't actually change; what you're measuring did. Cross-check against Search Console clicks, which aren't affected by client-side tracking bugs, before concluding anything about rankings.
• SERP layout changes — Google added an AI-generated overview, a larger People Also Ask block, or a shopping carousel above the organic results for your terms. Your position in the ranking may not have moved at all; your visible click share did, because the page above you now takes up more room.
• A single lost link — one large referring site removed a link, deindexed a page, or added nofollow. That's real and worth investigating, but it's a link problem tied to one source, not evidence of a broad algorithmic reassessment.
• A site change that happened to land around the same time — a redesign, a URL migration, a new CMS, a consolidated set of pages. Any of these can produce a ranking shift on their own, and if the timing overlaps with a public rollout it's easy to blame the wrong one. Check your own deployment history with the same rigor you'd apply to checking Google's.

Confirming an Update Happened — and That It Actually Hit You

Google publishes a running record of confirmed update rollouts, including start and end dates for the rollout window. Line your traffic or ranking change up against that record before assuming causation — a drop that starts three days before a confirmed rollout began almost certainly has a different cause.

Industry chatter — a spike in SEO forums, communities and trackers reporting volatility around the same dates — is useful corroborating evidence, but it's not proof for your specific site; treat it as a signal to check your own data more carefully, not as the conclusion itself.

Inside Search Console, open the Performance report and compare the period before and after the suspected date range, then segment by query, page, country and device rather than looking at the aggregate line alone. A broad, even decline across many query types and pages looks like a genuine core update. A decline concentrated in one content type, one page template, or one set of pages tied to a single technical change points somewhere else, even if it happened to coincide with a rollout window.

Also give the rollout time to finish before drawing conclusions. Updates typically roll out gradually over days to a couple of weeks, and rankings can wobble noticeably mid-rollout before settling into whatever the update's final effect actually is.

What to Do — and What Not to Do

Once you're reasonably confident a core update is the cause, the useful work is comparative and unhurried: look at what's now outranking you for the specific queries that matter, and be honest about whether those pages demonstrate more depth, more direct experience, or a clearer answer to the query than yours does. Audit your own content the same way — not by word count, but by whether a knowledgeable reader would find it genuinely useful, and whether you'd be comfortable putting your name on it as the most complete answer available. Look specifically at the queries where you dropped the most, rather than the site as a whole; a broad core update still tends to hit some topics harder than others, and that pattern tells you where the gap actually is. Ordinary technical maintenance — fixing crawl errors, cleaning up thin or duplicate pages you already knew were weak, improving page experience — remains worth doing regardless of any update; it just isn't a targeted fix for this specific event.

The instinct to do something immediately is understandable and almost always counterproductive. A handful of reactions consistently make things worse rather than better:

• Don't panic-rewrite content within days of a rollout. Google evaluated the version it already crawled; rewriting before you understand what actually changed risks introducing new problems on top of a change you haven't diagnosed.
• Don't mass de-index or delete pages as a blanket reaction. Bulk "pruning" on the theory that some thin pages are dragging the whole site down can remove pages that were never the issue and cost you internal linking and long-tail traffic you did have, without evidence it addresses the actual cause.
• Don't reach for the disavow tool reflexively. Disavowing links is a remedy for toxic-link penalties and manual actions tied to unnatural links, not for a broad relevance reassessment. Used without evidence, it can strip out links that were genuinely helping you.
• Don't chase a recovery in the first two weeks. The rollout itself can take that long, and ranking movement during an active rollout isn't a stable reading — treat anything you see before it finishes as provisional.

Honest Recovery Expectations

The realistic timeline is longer than most people want to hear: for a genuine core update, meaningful recovery generally has to wait for the next core update, because that's when Google's ranking systems are recalculated wholesale again. There's no continuous mechanism where fixing something today moves your ranking back tomorrow — the improvements you make between updates position you for the next recalculation, they don't trigger an immediate one.

It's also worth being honest that recovery isn't guaranteed at all. Sometimes an update reflects a durable shift — the intent behind a query has genuinely moved toward a different kind of result, or a category of content that used to rank well no longer represents what's most useful for that search. In that case there's no old position waiting to be restored, because the target itself changed. The fair way to frame the work is "become more competitive for this query as it's now understood," not "undo a penalty and return to where I was." Those are different goals, and only one of them is actually available to you.