A Google Core Update is a broad, sitewide change to Google's core ranking algorithm that reassesses how content is evaluated across every topic and region. Google confirms core updates publicly and rolls them out 3-4 times per year. They are recalibrations of what the algorithm values — not penalties for rule violations.
A core update is Google changing what "good content" means — not what "bad content" means. Sites that drop haven't broken any rules. Their content is just no longer the best answer under the new criteria.
What is a Google Core Update?
A Google Core Update is a broad, official update to the core ranking algorithm — the part of the system that decides which pages best answer a given query. Unlike targeted updates (spam, reviews, page experience), a core update touches how Google evaluates every site, in every language, across every topic.
Google announces core updates on the Search Central blog and the @searchliaison account. Each is named by month and year: "March 2024 core update," "November 2024 core update," and so on. Between updates, Google's evaluation logic is relatively stable; during a core update, it shifts — and sites redistribute up or down accordingly.
Google's guidance is unusually direct: "There's nothing wrong with pages that may perform less well in a core update. They haven't violated our webmaster guidelines... What we ask you to focus on is content improvement." Core updates are quality re-scoring, not spam enforcement.
Why core updates matter
A single core update can reshape a site's revenue for a full quarter. Five reasons every SEO watches them closely:
- Traffic swings are steep. A 30-70% drop in a week is not unusual for affected sites, and gains of similar size hit the beneficiaries.
- Recovery is slow. The next core update — usually 3-6 months later — is the earliest chance to recover meaningfully.
- Signals shift, not just weights. Google sometimes introduces new quality signals (originality, first-hand experience) during core updates, which changes what "good" looks like going forward.
- YMYL sites carry higher risk. Health, finance, and legal sites face stricter E-E-A-T expectations in every core update.
- They shape the industry. Post-update case studies dictate what SEOs prioritize for the next 6-12 months.
How a core update works
A core update is a training-and-deployment cycle inside Google's ranking system, not a rule change published in a document. The mechanics:
Re-evaluation, not re-punishment
Google's ranking systems continuously score pages on hundreds of signals — helpfulness, expertise, freshness, page experience, on-page relevance. A core update recalibrates how those signals combine. The same page can now score higher or lower even if nothing on the page changed.
Rollout across 1-3 weeks
Updates deploy gradually across Google's data centers. Rankings fluctuate during the window. The "before" state and the "after" state are only clear once Google announces the rollout is complete.
Sitewide, not page-level
Core updates evaluate sites holistically. Sections of thin, unhelpful content can drag down the ranking of high-quality pages on the same domain. That's why site-level cleanup matters even when only some pages seem affected.
Major Google Core Updates since 2018
| Update | Date | What changed |
|---|---|---|
| Medic Update | Aug 2018 | First named broad core update; hit YMYL health/finance hardest |
| March 2019 Core | Mar 2019 | Broad quality reassessment |
| January 2020 Core | Jan 2020 | Big swings in publishers, health, finance |
| May 2020 Core | May 2020 | Notable travel + hospitality shifts during COVID |
| December 2020 Core | Dec 2020 | Broad reassessment; delayed to late in year |
| Helpful Content + March 2024 Core | Mar 2024 | Merged Helpful Content System into core; aggressive on scaled/AI content |
| Aug 2024 Core | Aug 2024 | Recovery signals for small sites hit by prior updates |
| Nov 2024 Core | Nov 2024 | Continued focus on originality and first-hand experience |
| Mar/Jun 2025 Core | 2025 | Ongoing quality re-scoring; further AI-content signals |
Real core update examples
1. A YMYL health site drops 62% in March 2024
An affiliate-driven supplement review site with no medical author credentials lost 62% of organic sessions inside 10 days of the March 2024 update. Content was accurate but generic. Recovery required rewriting under a named medical reviewer, adding first-hand product testing, and pruning 40% of the URL count.
2. An SMB HVAC blog gains 3x in August 2024
A local HVAC company's blog had ~4,000 sessions/month before August 2024. The August update explicitly aimed to restore small, independent sites unfairly hit by earlier rounds. Their traffic jumped to 12,400 sessions/month within two weeks — no on-page changes had been made.
3. AI-generated content site loses 90%+ in Sep 2023
A site publishing 500 AI-written articles per month across dozens of niches lost more than 90% of visibility during the September 2023 Helpful Content Update, and did not recover in subsequent core updates. Scaled, low-effort AI content is now an explicit downweight signal.
Core update vs Helpful Content Update — what's the difference?
Core Update
- Broad recalibration of the entire ranking system
- Affects all queries, categories, and regions
- Confirmed by Google with names + dates
- Sitewide impact — high-quality pages can rise while thin ones drag them
- Recovery requires the next core update to see the change
Helpful Content Update (now part of core)
- Originally a separate classifier for "content written for people, not search engines"
- Merged into the core system in March 2024
- Aggressively downgrades scaled AI and thin affiliate content
- Applied at the sitewide level
- Now moves with every core update, not on its own schedule
7 best practices when responding to a core update
- Wait for the rollout to complete. Numbers swing throughout the window. Analyze only after Google announces the end.
- Segment by page type and topic cluster. Losers rarely lose evenly — the pattern matters more than the total drop.
- Run the drop through Google's helpful-content self-audit. The 32 questions Google publishes are the closest thing to a rubric.
- Prune or rewrite thin pages. Sitewide quality signals drag everything down. Cutting weak pages can lift strong ones.
- Strengthen author credentials. First-hand experience, real credentials, and byline pages carry more weight in each successive update.
- Fix duplicate and low-differentiation content. Multiple pages targeting the same intent is a classic signal Google now downweights.
- Don't chase the update — build for the next one. Recovery lands at the next core update, so the work has to be structural, not cosmetic.
Sites that lose traffic during a core update often blame a coincidental redesign or migration. Rolling back rarely helps. If Google's evaluation of your content quality changed, the design layer is a rounding error.
Common core-update mistakes to avoid
- Making changes mid-rollout. Wait for the announced end date before touching anything.
- Disavowing backlinks. No relationship to core update outcomes.
- Assuming it's technical. Core updates are about content quality, not crawl or schema issues.
- Expecting overnight recovery. The next update is the earliest chance to see the impact of your fixes.
- Adding more thin content. Volume without helpfulness makes the next update worse.
How theStacc helps
Our team runs post-core-update audits that segment drops by cluster, topic, and page type — then rebuild the content around named authors, first-hand experience, and demonstrated E-E-A-T. We prune what's dragging you down and rewrite what's close-to-ranking. The goal is a site Google's next update lifts, not one that limps sideways.
Frequently asked questions
No. Google is explicit that core updates are recalibrations of what its algorithm values, not penalties for rule violations. There is nothing to "fix" in the traditional sense; the goal is to raise overall quality until the next update re-evaluates the site.
Google typically announces the start and end. Rollouts usually take between one and three weeks. Wait for the announced end before analysing traffic — numbers move throughout the window.
Recovery happens between updates, not within one. Audit affected pages, remove or rewrite thin content, strengthen E-E-A-T signals, and improve helpfulness. Full recovery typically appears at the next core update, 3 to 6 months later.
Not exclusively. Some updates land harder on YMYL categories (health, finance) because Google raises the bar for those, but every category is affected. If your niche took a bigger hit, quality expectations rose for that niche.
No. Disavowing links has no bearing on core update outcomes. The link disavow tool is for manual spam actions, not algorithmic ranking recalibrations.
Related glossary terms
Sources
- [01]Google Search Central — Core updates
- [02]Google — Creating helpful, reliable, people-first content
- [03]Search Engine Land — Google update history
- [04]theStacc internal: post-March-2024 recovery audits across 42 client sites
