Google rolled out its first core update of 2026 on 27 March, and early data shows more than half of monitored websites are already seeing ranking shifts. Some sites have lost 20 to 35 percent of their organic traffic. Others have gained.
The difference between the winners and the losers? Not whether they used AI. Whether they used it well.
This is the third Google algorithm change in four days. A spam update completed in under 20 hours on 24-25 March, the fastest in Google's dashboard history. The core update landed 27 March. And the February 2026 Discover update is still settling. Three overlapping updates targeting the same problem from different angles: low-value content produced at scale.
What Google Actually Said
Google's announcement was deliberately bland. The Search Status Dashboard said:
"Released the March 2026 core update. The rollout may take up to 2 weeks to complete."
Google Search Status Dashboard, 27 March 2026
On LinkedIn, Google added a single clarifying line: "This is a regular update designed to better surface relevant, satisfying content for searchers from all types of sites."
No new guidance. No blog post. No detailed explanation. Google's official position is that nothing special has changed and that creators should keep making "helpful, reliable, people-first content."
But the ranking data tells a different story.
Three Updates in Four Days
The timing here matters. Google doesn't usually stack updates like this.
24-25 March: Spam update. Completed in under 20 hours, making it the shortest confirmed spam update ever. Search Engine Journal confirmed it targeted existing spam policies, particularly scaled content abuse, without introducing new categories. Glenn Gabe's reaction: "Wait, what? The March 2026 Spam Update has completed rolling out. Damn, that was fast."
27 March: Core update. The broad ranking recalibration now rolling out over two weeks. This one reassesses content quality across every site, in every language, in every region.
Still active: February Discover update. The Discover-specific core update from early February boosted local content and penalised clickbait. Its effects are still being measured.
Three updates running in parallel means some sites will see compounding effects. A page that lost Discover visibility in February, got flagged by the spam update on Monday, and then gets reassessed by the core update this week could face a steep drop. Or a site that was doing things right could see gains stack.
What the Update Actually Targets
Google won't say it explicitly, but the pattern is clear from the data. Technical analysis from TechBytes identified a new layer in Google's ranking systems: a Gemini 4.0 Semantic Filter designed to separate high-value content from what the industry has started calling "agentic slop."
The filter calculates something called "Information Gain," the mathematical difference between what a new page adds and what the top 100 existing results already cover. Content that just restates what everyone else has said gets zero information gain. Content with original data, first-hand testing, or real expert insight scores higher.
This isn't about AI versus human. It's about original versus derivative. A page written entirely by a person that just paraphrases Wikipedia will score the same zero as an AI-generated summary. And an AI-assisted article built on real client data, original testing, and verifiable expertise can score well.
Barry Schwartz, founder of Search Engine Roundtable, put it directly in his Search Engine Land coverage: the update follows the same "helpful content" playbook, but the detection systems are sharper.
The Line Between AI Content That Ranks and AI Content That Got Wiped
This is the question most UK business owners will ask this weekend. "I use ChatGPT to help write blog posts. Am I in trouble?"
Probably not, if you're doing it right. Here's where the line sits.
What gets penalised:
- Hundreds of pages generated with AI and published with no human review
- Blog posts that restate the same advice found on every other website
- Service pages with identical structures and interchangeable industry names
- FAQ sections that answer questions nobody is asking, padded for keyword count
- Content farms publishing 50+ articles per week across dozens of categories
What survives (and can gain):
- AI-assisted writing that's built on first-hand experience and edited by someone who knows the subject
- Case studies with real client outcomes, real numbers, real project details
- Analysis backed by proprietary data or original research
- Content where the author can demonstrably prove their expertise
- Articles that add something to the conversation that doesn't exist elsewhere
We saw the same pattern when we crawled 1,460 top websites for AI readiness. The sites doing well were the ones that invested in original implementation, not the ones that auto-generated files from SEO plugin defaults. Quality of execution, not presence alone, is what separates visibility from invisibility.
A Five-Minute Content Audit You Can Run Right Now
Open your website's blog or news section. Pick any three posts and ask these questions about each one:
- Could a competitor publish this exact article under their name? If the answer is yes, the content has zero information gain. It needs your data, your experience, your specific take.
- Does it contain anything that requires real-world experience to know? A plumber writing about common boiler faults they've actually fixed is different from a plumber republishing boiler troubleshooting guides from manufacturer websites.
- Can you prove the author wrote it? Not just an author bio. Linked case studies, client testimonials, professional qualifications, a track record on the topic. Google calls this E-E-A-T: Experience, Expertise, Authoritativeness, Trustworthiness.
- When was it last updated? Stale content with outdated advice (pricing from 2023, regulations that have changed) signals abandonment.
- Would you trust it if you found it on a competitor's site? Be honest. If it reads like filler, Google's systems are increasingly likely to agree.
If your content passes those five checks, this update is working in your favour. If it doesn't, you've got a two-week window while the rollout completes to start fixing the weakest pages first.
Why AI Visibility Matters More After This Update
Here's something the SEO recovery guides aren't covering. This core update doesn't just affect traditional search rankings. It affects how Google's AI systems assess your site's authority for AI Overviews, Discover recommendations, and the broader signals that determine whether AI platforms like ChatGPT and Claude recommend your business.
Sites that demonstrate real expertise don't just rank better in search. They become the sources that AI systems trust and cite. And businesses that have invested in AI visibility through structured identity files, verified authorship, and original content are exactly the businesses this update rewards.
When we asked Google Gemini what it learns from AI discovery files that it can't learn from a website, it identified five categories of information that HTML alone can't communicate. Brand naming rules. Verified Q&A pairs with source URLs. Permission signals that tell AI systems "yes, recommend us." None of that is AI slop. All of it strengthens the trust signals this update is looking for.
The businesses most exposed right now are the ones with thin websites, no original content, and no AI discovery files. They're invisible to AI systems and now they're losing ground in traditional search too. That's a double hit.
If you haven't checked your AI visibility yet, the free AI Visibility Checker on our sister site shows you exactly what AI systems see when they visit your website.
What to Watch Over the Next Two Weeks
The rollout takes up to two weeks. During that period, rankings will fluctuate as Google's systems reassess content across the web. A few things to keep in mind.
Don't panic on day one. Early fluctuations are normal. Google has said explicitly that a drop during rollout doesn't necessarily mean your site has a problem. Wait until the rollout completes before drawing conclusions.
Monitor Search Console daily. Watch both Search and Discover performance separately. With three overlapping updates, you need to understand which channel is affected and by how much.
Track your competitors. If they're dropping and you're stable, do nothing. If you're both dropping, the issue might be industry-wide quality recalibration. If they're rising while you're falling, look at what their content has that yours doesn't.
Don't delete content in a panic. Removing pages mid-rollout can cause more volatility. Instead, identify your weakest content and plan improvements. Prioritise the pages that drive the most traffic or the most business.
Google's December 2024 core update showed that design-first sites with original content recovered fastest. The same principle applies here. Invest in quality, not volume.
Frequently Asked Questions
When did the March 2026 core update start rolling out?
Google began the rollout at 2:00 AM PT on 27 March 2026 and expects it to take up to two weeks to complete. The update was announced on the Google Search Status Dashboard.
Does this update ban AI-generated content?
No. Google has been consistent on this: AI content isn't penalised for being AI-generated. It's penalised for being low quality, derivative, or produced at scale without original insight. AI-assisted content built on real expertise and edited by a knowledgeable human can rank well.
How do I know if my site is affected?
Check Google Search Console daily during the rollout. Look at both Search results and Discover performance separately. Compare traffic to the two weeks before 27 March. Wait until the rollout finishes (around 10 April) before drawing firm conclusions.
What is "AI slop" and why does Google care?
"AI slop" is the industry term for mass-produced AI content published without human review, original research, or real expertise. It adds nothing to the conversation. Google cares because it dilutes search quality: users get dozens of pages saying the same thing in slightly different words, none of them useful.
Why did Google release three updates in one week?
The spam update (24-25 March) targets policy violations. The core update (27 March) reassesses content quality broadly. The Discover update (February, still settling) adjusts the content feed. They serve different purposes but all address the same root problem: low-value content flooding search results.
Should I delete my AI-written blog posts?
Don't delete content during an active rollout. Instead, audit your posts using the five-question test above. Prioritise improving your weakest pages by adding original data, real examples, and expert commentary. Thin pages that can't be improved may be worth removing after the rollout completes.
What is the "Information Gain" metric?
Information Gain measures how much new, useful content a page adds compared to what already exists in the top search results. A page that restates existing advice scores zero. A page with original research, first-hand case studies, or proprietary data scores higher. Google's systems increasingly use this to separate valuable content from rehashed filler.
How does this affect UK local businesses specifically?
Local businesses with real expertise often do well in core updates. A Kettering plumber writing about real jobs they've completed has more information gain than a national directory republishing generic plumbing advice. Original, experience-based content from local experts is exactly what this update rewards.
Is Your Content Ready for This Update?
Google's March 2026 core update rewards original expertise and penalises derivative content. If you're unsure where your website stands, our free AI Visibility Checker shows exactly what AI systems see when they assess your site.
Check Your AI VisibilitySources
- Google Search Status Dashboard - Core Updates
- Google March 2026 core update rolling out now - Search Engine Land
- Google Begins Rolling Out March 2026 Core Update - Search Engine Journal
- Google's March Spam Update Felt Muted But May Signal Bigger Changes - Search Engine Journal
- Google March 2026 Core Update: Technical Analysis - TechBytes