How AI Overviews in Google and Yandex Capture Traffic in 2026
Generative modules like AI Overviews answer user questions before they reach the links. SEO strategy needs reinvention: what matters is not rankings, but the brand's share of voice in generative results. As a side effect, UTM tags for traffic tracking have become unreliable.
AI-processed from Habr AI; edited by Hamidun News
Rising website rankings in Google search results no longer guarantee traffic. Generative results (AI Overviews in Google and YandexNeuro) intercept user queries on the zero screen — users get ready-made answers before reaching regular links.
How Query Fan-Out Works
Previously, search engines simply ranked pages by relevance. Now it works differently. When a user enters a query, the algorithm applies query fan-out — breaking down one complex query into several simpler subqueries to compile a complete answer. The language model synthesizes information from different sources into a single text and displays it directly in search results.
The problem is that models begin to train on each other's outputs. One generative model trains on the output of another, which trains on a third. The result: language model collapse. Content becomes increasingly homogeneous, quality declines because the system trains on its own errors. If you publish a rewrite of something already on the internet, the algorithm will notice and won't rank it.
Share of Voice: The New Visibility Metric
Traditional metrics are dying. Ranking position? No longer critical. In its place comes Share of Voice — how many times your site is mentioned in generative answers relative to competitors.
Three paradigms simultaneously:
- SEO — optimization for traditional results (link lists)
- AEO (Answer Engine Optimization) — optimization for generative answers (appearing in AI Overviews)
- GEO (Generative Engine Optimization) — optimizing content for training generative models
But how do you measure Share of Voice if old tracking tools (UTM parameters, GA, Yandex.Metrica) provide broken data due to redirects through AI Overviews? A new methodology for counting transitions was needed when direct attribution is impossible.
What Must Change in SEO Strategy
Forget about the race for rankings. Climbing from 50th to 10th place may bring zero additional traffic if AI Overviews have already answered the question. Generative results capture clicks in the chain: question → ready-made answer (AI) → if details are needed → link. Instead, you need to bet on originality. Search engines penalize rewrites and thin content. To appear in a generative answer, you need to write research, a case study, an unconventional take on a problem — something that hasn't been in the public domain yet.
And finally, metrics. Instead of tracking rankings by Semrush, analyze brand mentions in AI Overviews results, measure Share of Voice, compare yourself with competitors in the generative space.
What This Means
The era when a Google ranking guaranteed traffic is ending. The age of Share of Voice and original content is arriving. For website owners, this is a radical strategic overhaul — from chasing rankings to creating content worthy of appearing in AI model answers.
- Oct 28, 2026A 2015 CNN Beat GPT-6 Astra and Claude Fable 5.1 at Skin Cancer Diagnosis
- Oct 28, 2026Google taught an LLM to “sleep”: SLEEP consolidates knowledge into weights, up to 10 million tokens
- Oct 28, 2026Keystone removes the model name from code: open-source LLM gateway adds 0.64 ms
- Oct 28, 2026Wired Tested 6 AI Agents in a Day: Dots Leads, Gemini Spark Fails at Lunch
Key Questions Answered
How Query Fan-Out Works?
Previously, search engines simply ranked pages by relevance. Now it works differently. When a user enters a query, the algorithm applies query fan-out — breaking down one complex query into several simpler subqueries to compile a complete answer. The language model synthesizes information from different sources into a single text and displays it directly in search results. The problem is that models begin to train on each other's outputs. One generative model trains on the output of another, which trains on a third. The result: language model collapse. Content becomes incre
What are the key details on Share of Voice: The New Visibility Metric?
Traditional metrics are dying. Ranking position? No longer critical. In its place comes Share of Voice — how many times your site is mentioned in generative answers relative to competitors. Three paradigms simultaneously: SEO — optimization for traditional results (link lists) AEO (Answer Engine Optimization) — optimization for generative answers (appearing in AI Overviews) GEO (Generative Engine Optimization) — optimizing content for training generative models But how do you measure Share of Voice if old tracking tools (UTM parameters, GA, Yandex.Metrica) provide broken da
What Must Change in SEO Strategy?
Forget about the race for rankings. Climbing from 50th to 10th place may bring zero additional traffic if AI Overviews have already answered the question. Generative results capture clicks in the chain: question → ready-made answer (AI) → if details are needed → link. Instead, you need to bet on originality. Search engines penalize rewrites and thin content. To appear in a generative answer, you need to write research, a case study, an unconventional take on a problem — something that hasn't been in the public domain yet. And finally, metrics. Instead of tracking rankings b
Ship AI Faster with Founders, Engineers & Weekly Architecture Reviews
Inside Vibe Club (hamidun.com/club) — weekly live masterminds with Zhemal Khamidun, Tech Radar, full LMS access, and 370+ production Claude Code skills.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.