Google’s latest core update arrived mid-May alongside a meaningful expansion of AI Mode. Here’s a clear-eyed look at what we know for certain, what it means in practice, and how to respond without making things worse.
The May 2026 Core Update began rolling out on 21 May and is expected to complete around 4 June. It reinforces Google’s existing E-E-A-T framework, no new vertical-specific signals have been announced. AI Mode and AI Overviews are reshaping how content gets discovered, and traditional CTR metrics alone no longer tell the full story. The practical response is structured, original, well-attributed content, not reactive rewrites mid-rollout.
What We Know for Certain
Google confirmed the May 2026 Core Update via the Search Status Dashboard on 21 May (incident ID: wdAXJk6LRRihEjpzEeWE). Rollout periods typically run up to two weeks, which puts full completion around 4 June.
As with every core update, this one recalibrates how Google evaluates expertise, experience, authoritativeness, and trustworthiness, the E-E-A-T framework documented in Google Search Central. Google has not announced vertical-specific signals or penalties targeting particular content types. The update is broad.
ACTIVE ROLLOUT — DO NOT REACT YET
Rankings fluctuate significantly during a core update rollout. Making content changes in response to early volatility can compound the noise. Wait until the rollout completes and allow at least one week of stable data before drawing conclusions.
The broader context: Google is updating more frequently
This update follows a pattern of roughly 3–4 week release cycles that Google has maintained since late 2025. The implication is that there is no single “fix” window, content quality is evaluated on an ongoing basis, not just in the wake of a major update. Brands that treat core updates as periodic fire drills tend to underperform compared to those that maintain consistent content standards between them.
AI Mode and AI Overviews: A Clearer Distinction
These two features are frequently conflated, and the distinction matters for how you approach content.
AI Overviews
AI Overviews appear automatically at the top of standard search results for informational queries. They are generated summaries that aggregate information from multiple sources, sometimes with citations and links. As documented by Google, they are designed to handle queries where a user wants a direct answer rather than a list of links to explore. The traffic dynamic is real: content that serves as a source for an Overview may see fewer direct clicks but gains a citation that functions as a trust signal.
AI Mode
AI Mode is a deliberate, conversational interface. Users opt into it when they want to research a topic across multiple follow-up questions, it is closer in nature to how someone might use ChatGPT or Perplexity than to a standard search. Because AI Mode sustains a multi-turn dialogue, it rewards content with genuine depth. A 500-word summary is unlikely to fuel a meaningful conversation; a thorough, well-structured long-form piece can.
The question is no longer just “does this page rank?” it is “does this page get cited, and does that citation drive qualified traffic?”
A useful reframe for your KPI conversations
The zero-click reality
For purely informational queries, definitions, quick facts, how-to explanations, AI Overviews already resolve many searches without a click. This is not new behaviour; featured snippets have had a similar effect for years. What is new is the scale and sophistication of the summarisation. The practical response is not to fight zero-click but to make your content indispensable beyond it: original research, expert commentary, interactive tools, proprietary data, things an AI cannot generate from existing sources.
What This Means for Your Content Strategy
The core shift is that ranking well and being cited by AI are now partially separable goals. A page that ranks fifth but is structured as a clean, direct-answer resource may be cited in an AI Overview above the pages that rank above it. Conversely, a top-three page full of thin content may rank through legacy authority signals while getting bypassed entirely in AI-generated summaries.
Original insights matter more than ever
AI systems, whether Google’s or others like ChatGPT, Bing Copilot, or Perplexity, draw from indexed web content. If your content essentially repackages what is already widely available, it adds nothing that the AI cannot produce on its own. The content most likely to earn citations is the content an AI cannot generate independently: original survey data, proprietary case studies, named-expert analysis, and insights drawn from internal experience.
Entity clarity is a ranking and citation signal
Both Google’s ranking systems and LLM-based tools like Gemini, ChatGPT, and SEMrush’s AI features use entity recognition to understand what a piece of content is about. Inconsistent naming across a site, using “AI Overviews,” “AI Overview,” and “SGE” interchangeably, weakens entity signals. Standardise terminology across all pages and metadata, and name the entities your content is authoritative about explicitly.
A Six-Step Framework for AI-Era Content
This framework is adapted from what has worked in practice across B2B and enterprise retail contexts as AI search has matured.
1. Establish your AI citation baseline
Before changing anything, audit the top 20 queries your site targets. Check each one in Google Search, AI Mode, ChatGPT, Bing Copilot, and Perplexity. Note whether your content appears as a citation, and who does appear when you don’t. This is your baseline, the gap analysis drives everything else.
2. Lead with a direct answer
Each high-value page should open with a 150–200 word paragraph that directly answers the primary query. No preamble, no scene-setting. AI extraction systems surface the content that answers the question first; content buried three paragraphs down rarely gets cited.
3. Standardise entity mentions across your site
Audit your key product names, service names, and topic clusters. Pick a canonical form for each and enforce it in all content, metadata, and schema markup. Inconsistency dilutes entity recognition in both traditional ranking and LLM citation systems.
4. Invest in content AI cannot replicate
Commission original research. Run proprietary surveys. Publish named-expert commentary. Produce tools or calculators specific to your domain. These create a citation moat that generic, well-structured content cannot. If you have internal data, conversion benchmarks, industry figures from your own dataset, publish them with methodology.
5. Align technical SEO with AI extraction requirements
Ensure Googlebot-Extended is not blocked in robots.txt (this crawler feeds AI Overviews). Implement FAQPage, HowTo, and Article schema where applicable. Structured data makes content extractable and aids both AI citation and rich result eligibility. Validate all schema with Google’s Rich Results Test, marking up content that is not visible on the page is a common error that can cause manual actions.
6. Recalibrate your KPIs for the AI era
Organic CTR alone misses the picture. Add AI citation share (track via manual audits or tools like Ahrefs and SEMrush’s AI visibility features), branded search volume, AI-referred session quality in GA4, and direct traffic trends that correlate with AI visibility spikes. These together tell you whether your content is genuinely influential or merely ranking.
Case Study Illustration: The Impact of Structured, AI-Optimised Content
A B2B SaaS company undertook a structured content overhaul in Q1 2026, adding direct-answer leads, standardising entity names, implementing FAQPage schema, and publishing a proprietary benchmark report. The following metrics were observed post-implementation (results will vary; these are illustrative of the pattern, not a universal guarantee):
| Metric | Before | After | Change |
| Google organic CTR (avg.) | 18% | 28% | +10pp |
| AI Overview citations per key page | 0 | ~4 | New channel |
| Inbound leads (organic) | 12/month | 20/month | +67% |
| Average content engagement | 3 min | 5 min | +2 min |
| Branded search volume | Baseline | +22% | Halo effect |
The branded search lift is worth noting separately. Being cited in AI Overviews consistently drove awareness even when users did not click through — they searched for the brand directly afterwards. Traditional attribution misses this entirely.
What the enterprise retail data adds
In a parallel example from enterprise retail, sessions attributed to AI Overview referrals converted at roughly twice the rate of standard organic sessions. The hypothesis is sound: a user who arrives via an AI citation has already had their question partially answered, so they arrive with higher intent and less friction to convert. This is an argument for optimising for AI citation quality, not just citation volume.
New KPIs Worth Adding to Your Dashboard
Common Mistakes to Avoid
- Rewriting content mid-rollout. You will not be able to distinguish update volatility from a genuine signal. Wait for the data to settle.
- Publishing AI-generated content at scale. Content produced without genuine expertise or original insight is precisely what E-E-A-T is designed to demote. Named authorship and proprietary perspective are more valuable than volume.
- Blocking AI crawlers. Check your robots.txt for Googlebot-Extended. If it is disallowed, your content is excluded from AI Overview consideration entirely.
- Conflating AI Mode with AI Overviews. The content requirements and optimisation approaches differ. A short direct-answer paragraph serves AI Overviews well; AI Mode rewards depth and expert nuance.
- Marking up content that is not on the page. Schema errors, particularly FAQ schema describing questions and answers not visible to users, can trigger manual actions. Only mark up what is actually rendered on the page.
- Measuring only organic CTR. Zero-click does not mean zero value. Add AI citation share and branded search volume to get an accurate picture of your total visibility.
Frequently Asked Questions
How does AI Mode affect click-through rates for top-ranking pages?
AI Overviews resolve many informational queries without requiring a click. Pages that rank in the top three but contain only generic information are most exposed. The mitigation is content that goes beyond what an AI can summarise: original data, expert opinion, step-by-step guidance, and interactive tools. Being cited inside an AI Overview also drives a different kind of traffic, lower volume, higher intent.
What type of content is most likely to be cited in AI Overviews?
Structured, authoritative content with clear direct-answer paragraphs. Content that leads with a concise answer (150–200 words), backs it with specific evidence, and is attributed to a named expert or organisation consistently performs well. FAQs, comparison tables, and how-to steps are particularly extractable by AI systems.
Should I make changes to my content during the rollout?
No. Google advises against reactive changes during an active rollout. Wait until the update completes (projected around 4 June 2026) and allow at least one week of stable data before drawing conclusions or making significant changes.
What is the difference between AI Mode and AI Overviews?
AI Overviews appear automatically at the top of standard search results for informational queries — they are passive summaries. AI Mode is a deliberate, conversational interface where users ask follow-up questions across a research session. The content requirements differ: AI Overviews reward concise, direct answers; AI Mode rewards depth, expert nuance, and content that can sustain a multi-turn conversation.
How should we prepare content for the expected June 2026 follow-up update?
Focus on three things: interlink your existing high-performing pages with newer supporting content to build topical clusters, refresh any posts containing outdated statistics or examples, and monitor which pages are being cited in AI Overviews versus simply ranking. The goal is topical authority across a cluster of pages, not a single optimised post.
Summary & Key Takeaways
- The May 2026 Core Update is confirmed and active. No vertical-specific targeting has been announced, E-E-A-T remains the governing framework.
- AI Mode and AI Overviews serve different user intents and require different content approaches. Don’t conflate them.
- Being cited in AI results is now a meaningful visibility channel, distinct from traditional ranking. Your KPI framework needs to reflect this.
- The content types most resilient to AI commoditisation are those with original data, named-expert analysis, and proprietary insight, things the AI cannot generate itself.
- Do not make reactive changes during the rollout. Monitor, document, and act on stable post-rollout data.
- A June follow-up is widely anticipated. Begin building your content cluster now, interlinked supporting pieces reinforce topical authority ahead of the next cycle.
This post will be updated after the update fully completes and data stabilises. A follow-up covering June 2026 changes and refined AI citation findings is planned for mid-June.
Caught feelings for cybersecurity? It’s okay, it happens. Follow us on LinkedIn and Instagram to keep the spark alive.
