Surviving Google Core Updates: What to Do When Your Traffic Tanks
A core update hit and your charts fell off a cliff. Before you panic-delete half your site or buy a "recovery service," read this. Recovery is real, but it doesn’t work how most people think.
It usually starts with a Slack message: "Is something wrong with analytics?" Nothing is wrong with analytics. A Google core update rolled out, and organic traffic just stepped down 30, 40, sometimes 60 percent within a week. No manual action. No warning. No specific page to blame, because everything fell a little and some things fell a lot.
The usual reaction is shock, then furious Googling, then the temptation to do something drastic tonight. So let's replace panic with a process.
First, confirm it was actually the update
Serious diagnosis starts with boring verification. Check the dates: Google announces core updates on its Search Status Dashboard, and rollouts take up to a couple of weeks. Does your decline align with the rollout window, or did it start earlier (suggesting a technical issue, a season, or a lost feature like a top story slot)?
Then segment before concluding anything:
- By page type in Search Console: did blog content drop while product pages held? That's information.
- By query type: informational queries hit harder than branded ones is the classic core-update signature. If branded search fell too, look for technical problems instead.
- Against competitors: check a handful of rival sites in a visibility tool. If your whole niche reshuffled, you're in an ecosystem event, not a personal one. Sometimes you even find you lost less than the sites you envy.
And rule out coincidences: a migration, a redesign, robots.txt changes, a CMP update blocking rendering. Core updates get blamed for a lot of self-inflicted wounds that happened the same week.
What a core-update drop actually means
Google's own framing is worth taking literally: core updates don't target sites, they reassess the whole web. Content that fell isn't flagged as bad; other content is now judged to deserve those positions more, under a refreshed definition of quality and relevance.
Which is why there's no penalty to remove, no reconsideration request to file, and no single fix. The system re-scored relevance and trust, and your site landed lower under the new scoring. Recovery means genuinely scoring better, then waiting for a future update to re-evaluate you at scale. Google has been fairly consistent about that last part: meaningful recoveries tend to register around subsequent core updates, months later, not days after you fix things.
That timeline is the hardest pill. Anyone selling you a 30-day core update recovery is selling you weather control.
The honest audit
With diagnosis done, audit the site the way a skeptical stranger would. Recent updates have been unusually consistent about what they reward and punish, so aim your questions there:
Is a meaningful share of your content written for rankings rather than readers? The tells: articles targeting every keyword variation with interchangeable 1,500-word answers, topics far outside your lane chased for traffic, intros that restate the question for three paragraphs, roundups of products nobody at the company touched. Sites built heavily on this pattern, including AI-scaled versions of it, were exactly the profile hit hardest through 2023 and 2024.
Would anyone recognize you as a source? Real author names with real credentials, first-hand evidence in reviews (your own photos, your own measurements), an about page that proves humans, and expertise concentrated on topics you plausibly know. This whole cluster, call it E-E-A-T if you like, isn't a checklist item; it's the aggregate impression your site leaves.
Does your experience quietly annoy people? Ad density that buries the answer, aggressive interstitials, autoplay, content split across pages for pageviews. Google says helpfulness; users say friction. Same thing.
Is there dead weight? Hundreds of thin, outdated, zero-traffic pages dilute the site-level quality picture. The pruning playbook (consolidate, improve, or noindex) applies here just as it does to indexing problems.
The recovery plan, sequenced
- Weeks 1-2: Diagnose and segment as above. Freeze drastic moves. Do not delete half the site in a weekend rage.
- Month 1: Fix the worst experience issues (ads, popups, speed) and rewrite or consolidate your ten most important declining pages, adding genuine firsthand value that competitors lack. Compare hits, er, compare your page honestly against the pages now outranking it and note what they do better. Usually something is visibly better: specificity, evidence, freshness, focus.
- Months 2-4: Work through the content inventory tier by tier. Strengthen author and trust signals sitewide. Keep publishing, but only within topics where you have real standing; breadth without authority is what got many sites here.
- Ongoing: Track rankings weekly, not hourly. Note the next core update's dates and watch your trajectory around it.
What not to do
Don't chase the update with tricks: swapping dates to fake freshness, mass "optimizing" with the same AI that wrote the problem, buying links to "boost authority." Don't panic-migrate domains; the assessment follows the content. And be wary of survivor-bias case studies: for every "we recovered with this one change" thread, hundreds of sites made the same change and stayed flat.
Questions from the trenches
"Traffic dropped but rankings look stable. What gives?" Check where you're ranking, not just position. AI Overviews, featured snippets and bigger ad blocks push classic blue links further down, so position four today can receive half the clicks position four earned two years ago. Compare click-through rate by query in Search Console over time. Sometimes the update didn't move you; it moved the page layout around you. The response is different too: win the snippet, strengthen brand search, diversify beyond queries that AI now answers directly.
"Should I disavow links after a core update?" Almost certainly not. Core updates aren't link penalties, and Google has repeatedly said most sites shouldn't touch the disavow tool. Disavowing after a core update is a ritual sacrifice: it feels decisive and does nothing. Spend those hours improving your ten most important pages instead.
"Part of my site recovered and part didn't. Is that normal?" Completely, and it's a gift: your own site is now a controlled experiment. Compare the recovered sections against the flat ones honestly: depth, evidence, author credibility, user intent match. Whatever separates them is your roadmap, written in your own data.
Sites that genuinely recover tend to share one trait: they stop asking "what does Google want" and start asking "why would a reader choose us." The answers overlapped almost perfectly, which is, in the end, the entire point of core updates. Rebuild toward that, give it the months it honestly takes, and the next reassessment has something real to find.
Free tools for this topic
Check your own site in 30 seconds
Run a free SEO audit and see exactly where your pages stand on the issues covered in this article.
Run Free SEO AuditKeep reading
Mobile-First Indexing Problems That Are Costing You Rankings
Googlebot is a phone now. Full stop. Whatever your mobile visitors can’t see, Google can’t rank, and plenty of "responsive" sites still hide half their value on small screens.
“Discovered – Currently Not Indexed”: Why Google Ignores Your Pages
Google knows your page exists. It just doesn’t think it’s worth crawling yet. That stings, but it’s fixable once you understand what the status is really telling you.