The State of
Search 2026

Google's 2026 Core Updates, Decoded

Three major algorithm events in the first five months of 2026, each tightening the screws on thin, AI-generated, and expert-lite content.

Google released three significant updates in 2026, each building on the same thesis: verifiable expertise beats stated authority. The era of ranking well by following on-page tactics alone is over.

The March 2026 Core Update (launched March 27, completed April 8) re-weighted 'Information Gain' as a ranking signal, measuring how much genuinely new knowledge a piece adds relative to what already ranks. It drove far higher volatility than December 2025, with 79.5% of URLs in the top 3 changing positions and 24.1% of pages ranking in the top 10 falling out of the top 100 entirely. Sites built on original research and demonstrated expertise recovered or improved; intermediary content farms took the hardest hits.

The March 2026 Spam Update, released in quick succession, targeted AI-generated content lacking human oversight. AI content reviewed by subject-matter experts was largely unaffected; unsubstantiated claims without evidence were penalised regardless of production method.

Google I/O 2026 (May 19) reframed the Search product entirely: AI Mode crossed 1 billion monthly users, Gemini 3.5 Flash became the default model across AI surfaces, and Search queries reached an all-time high quarter-over-quarter.

The May 2026 Core Update (May 21 to June 2) completed its rollout on June 2, 2026. Google's guidance is unchanged: make helpful, original content for people. There is no quick recovery from a core update; the biggest changes tend to follow the next update. Audit and improve content quality now.

The Zero-Click Era is Here

58.5% of US Google searches end without a single click to an external website (SparkToro/Datos, 2024). When an AI Overview appears, clickstream estimates put the zero-click rate as high as ~83%, and around 90%+ in Google AI Mode, though these figures vary by source and methodology.

The zero-click phenomenon began with featured snippets around 2014, but AI Overviews have accelerated the trend dramatically. SparkToro and Datos clickstream data shows 58.5% of US Google searches end without an external click, and on mobile that rises to 77%.

The impact is most severe at the top of the results. Ahrefs' December 2025 analysis of 300,000 keywords found that position 1 CTR drops 58% when an AI Overview is present. Positions 2-3 drop 50-47%, positions 4-10 drop roughly 38-29%, and position 10 still loses 19.4%. Only branded queries are an exception: CTR actually increases on brand-name searches when an AI Overview appears, because the overview acts as a trust signal rather than an answer.

HubSpot publicly attributed a 70-80% organic traffic loss between late 2024 and Q2 2025 to AI Overviews absorbing informational queries. Its CEO noted that top-of-funnel content now feeds Google's answers without generating clicks back to the source. This pattern now affects the majority of informational-heavy sites in every vertical.

The counterintuitive finding: AI-referred visitors tend to convert better than traditional organic. Ahrefs found their own AI search traffic represented 0.5% of total visits but drove 12.1% of signups, a 23x advantage in their own data (one company's traffic, not a universal benchmark). Other vendors report similar directional lifts. The implication: fewer clicks, but higher intent per click.

The AI Search Landscape in 2026

Google still owns 90%+ of traditional search. But in the AI chatbot layer, the race is wide open, and fragmented in ways that fundamentally change your visibility strategy.

The most important data point of 2026: 90% of brands have zero mentions in AI search results, according to Q1 2026 research tracking 177 brands across healthcare, SaaS, and financial services. If you're not actively optimising for AI visibility, you almost certainly have none.

Google AI Overviews now reach 2.5 billion monthly users across 200+ countries in 40+ languages, and appear on a large and fast-growing share of queries, especially informational ones. Trigger-rate estimates vary widely by study and query mix, from the low teens to roughly half, but the trend is unmistakably up. The Gemini app separately crossed 900 million monthly active users in May 2026, up from 750 million in Q4 2025, fuelled by Android integration.

In the independent AI chatbot layer, the competitive picture has shifted radically. Eight months ago, ChatGPT held 89% of measurable B2B AI referral traffic. As of March-April 2026 (Goodie Wave 2 report), it holds 62.6%. Claude surged from 1.4% to 18.5% in the same period, the biggest share shift of any platform. Gemini reached 10.6% and Perplexity 7.3%. The 'Big 1' is now the 'Big 4'.

Each platform has different retrieval logic, which means optimising only for ChatGPT now misses roughly a third of the AI traffic landscape compared to 12 months ago. Platform-aware content strategy is no longer optional.

Modern SEO: What Still Works in 2026

Technical SEO and content fundamentals have not died. They have become the foundation for every other layer of search visibility, including AI citation.

E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) is no longer a checklist. It's the filter Google applies at every layer of search including AI Overviews. Google's AI Search guidelines confirm: content must be original, useful, and written for people. AI-generated content is permitted with human oversight and authenticity signals.

Original research is now the single highest-leverage content investment. It earns external links (traditional SEO signal) and AI citations simultaneously. A quarterly owned-data study (survey, aggregated client data, or industry analysis) creates content that intermediary sites simply cannot replicate.

AI Overview reference articles are cited from positions up to 40, not just the top 10. Pages ranking 11-20 have a real shot at AI citations, widening the playing field beyond traditional first-page thinking.

One critical technical fix often overlooked: AI crawlers. Ahrefs found around 6% of sites block GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, or Google-Extended in robots.txt. If these bots cannot crawl your site, you do not exist in AI search. Audit your robots.txt today.

GEO: Generative Engine Optimization

GEO is the practice of optimising content so AI-powered platforms select, summarise, and cite it in generated answers. It is distinct from SEO, and in 2026, it is non-optional.

Generative Engine Optimization emerged as a formal discipline in 2024 and reached mainstream adoption urgency in 2026. The core difference: SEO earns clicks; GEO earns citations. The goal shifts from ranking a URL to becoming a trusted source that AI systems synthesise from.

Unlike deterministic search (keyword to ranked list), generative AI uses vector-based probabilistic models to synthesise answers from multiple sources. It does not 'rank' your page. It decides whether to trust and cite it. Trust signals for generative engines include: clear machine-readable structure, factual claims with sourced evidence, entity recognition across multiple platforms, content that directly answers specific questions, and narrow focus pages rather than broad catch-alls.

GEO is not replacing SEO. The most effective 2026 strategies use SEO for discoverability and GEO for AI visibility simultaneously. Strong SEO signals (topical authority, backlinks from credible domains, fast crawlable pages) are often prerequisites for being cited by AI systems. A page that ranks well is far more likely to be cited than one Google has never indexed properly.

The 5 GEO content principles that research consistently identifies: (1) Answer-first structure: lead with the direct answer, then elaborate. (2) Factual density: specific numbers, named entities, sourced claims. (3) Narrow page focus: dedicated pages for specific sub-queries rather than broad pillars. (4) Entity consistency: your brand and experts must appear uniformly across your own site, social, press, and third-party platforms. (5) Information gain: add unique insight beyond what the AI already synthesises; if your content merely repeats the summary, visitors who do arrive leave disappointed.

AEO: Answer Engine Optimization

AEO focuses on LLM visibility specifically, optimising for probabilistic, generative models rather than deterministic search engines. It is the technical layer that sits under GEO.

Answer Engine Optimization builds on GEO by adding a specific focus: getting your brand explicitly recommended in LLM-generated responses across ChatGPT, Perplexity, Gemini, and Claude. It is not about rankings. It is about prompt-level visibility.

Traditional SEO relies on deterministic systems: match keywords, rank by link equity. AEO operates on probabilistic models: you provide clear, factual, structured data so generative AI engines can synthesise answers and cite you as an authoritative source. The shift in mental model matters. You are not competing for a position. You are building a pattern of trustworthiness that AI training data and retrieval systems recognise.

Core AEO tactics for 2026: (1) Unblock AI crawlers in robots.txt (prerequisite for everything else). (2) Implement schema for AI interpretation: FAQ, HowTo, Article, Organisation, Person. Google's own AI documentation lists structured data as a best practice for AI visibility. (3) Natural language query optimisation: structure content to directly answer conversational prompts, not keyword-stuffed queries. (4) Citation engineering: place your content and brand mentions within domains that AI models frequently cite; source-quality backlinking matters more than raw domain authority. (5) Prompt-aligned formats: Q&A, TL;DRs, and structured H2/H3 subheadings match how AI engines extract and reassemble answers. (6) Query fan-out coverage: address the sub-queries, comparisons, and reformulations that AI generates when answering a broad question about your topic.

The AEO opportunity is significant: while paid CTR on AI Overview queries dropped 68%, AI search visitors convert at a dramatically higher rate than traditional organic. Fewer clicks, far higher intent per click. Owning the AI answer for commercial queries is becoming more valuable than holding the #1 organic result for informational ones.

llms.txt: The Honest 2026 Verdict

The hype was significant. The empirical data is clear. Here is what you actually need to know about llms.txt before investing time in it.

llms.txt is a proposed standard (similar to robots.txt) that would guide AI crawlers to your most important content in a clean markdown format. The idea sounds logical. The reality as of mid-2026 is very different.

A comprehensive analysis by Limy.ai monitored over 500 million AI bot visits across a 90-day window. Of those, only 408 visits targeted llms.txt directly, a statistically negligible figure. GPTBot, ClaudeBot, PerplexityBot, OAI-SearchBot, and Google-Extended all skip the file and crawl HTML directly. A SE Ranking study of 300,000 domains found only 10.13% adoption after 18 months of industry conversation.

Google's Gary Illyes confirmed in July 2025 that Google does not support llms.txt and is not planning to. John Mueller compared it to the discredited keywords meta tag. No major LLM provider (OpenAI, Anthropic, Google, Meta, or Mistral) has publicly committed to using llms.txt in production search surfaces.

The verdict: do not spend meaningful resources on llms.txt optimisation in 2026. The one legitimate use case is IDE agents (Cursor, Continue, Cline) and some MCP integrations, which do read it. If your audience uses coding tools, a low-effort implementation is reasonable. For everyone else, the return-on-investment does not justify the effort. What actually improves AI citability: clean crawlable site architecture, fast page speed, comprehensive structured data markup, clear machine-readable content structure (headers, lists, tables), and unblocking AI bots in robots.txt.

Your 2026 Search Strategy Action Plan

Thirty concrete actions, organised by priority. Start with the foundations: everything else builds on them.

The actions below are organised into three priority tiers. Foundations should happen this week. Content strategy changes this month. Multi-platform AI visibility work this quarter.