What Are AI Overviews?

How Do AI Overviews Affect Organic Search Traffic?

AI Overviews significantly reduce the share of searches that result in a click to any website. A Pew Research Center study of 68,879 US Google searches found that users clicked a traditional result in only 8% of searches that showed an AI summary, versus 15% for searches without one (Pew Research Center, 2025). Clicks to links embedded inside the overview itself occurred in just 1% of cases.

The scale of this shift is large. Google CEO Sundar Pichai stated that AI Overviews surpassed 2 billion monthly users in more than 200 countries and 40 languages, up from 1.5 billion the prior quarter (TechCrunch, Q2 2025 Alphabet earnings, 2025). At that reach, even small changes in per-search click behavior translate to billions of sessions where organic traffic is absorbed by the summary.

The keyword-level effect is sharper still. An Ahrefs analysis of 300,000 keywords using Google Search Console data found that AI Overview presence correlates with a 58% lower average click-through rate for the top-ranking page (Ahrefs, 2026). SEO teams now treat AI Overview presence as a primary variable when projecting organic traffic for any keyword target.

How Are AI Overviews Generated?

Google builds AI Overviews using its Gemini models in a retrieve-then-generate pattern. Gemini reads candidate passages from top-ranked indexed documents and writes a natural-language summary, citing source pages as in-line links. The user receives a direct answer without needing to evaluate or compare individual results.

The queries that trigger AI Overviews skew informational: how-to guides, definitions, comparisons, and research questions. Navigational queries (looking for a specific domain) and transactional queries (buying intent) produce overviews far less often. This means that informational content faces the highest traffic displacement, since it is precisely the category Google's model can most easily answer in place of the original source.

Because AI Overviews are injected via JavaScript after the initial page load, standard HTTP requests that read raw HTML will not capture them. A fully rendered page snapshot is required to see what a real user sees.

SEO and content strategy. Teams monitor which target keywords now surface AI Overviews, since those queries warrant different traffic forecasts. They also study which source pages Google cites inside the summaries, to understand what content attributes earn a citation slot.

Competitive intelligence. Research teams track how competing brands and products are described in AI Overviews for high-value informational queries. The framing inside a summary can shape perception before a user ever visits any website.

LLM evaluation and training data. AI teams collect AI Overview text to study retrieval behavior, citation accuracy, and where the model paraphrases versus drifts from the cited source material.

Web data pipelines. Engineers building SERP monitoring pipelines need to capture AI Overview content reliably. Massive's Search endpoint (/search) accepts an awaiting=ai parameter that holds the request open until the AI Overview has fully rendered before returning the page, removing the race condition that affects static HTML snapshots.

Target cited-source status, not just ranking position. Appearing as a cited source inside an AI Overview drives brand awareness even when the user does not click through. Pages with clear factual claims, strong authorship signals, and well-structured content (headers, lists, tables) are more likely to be selected as sources.

Add relevant structured data. FAQPage, HowTo, and Article schema give Google's retrieval layer explicit signals about the content type and the relationship between questions and answers. These signals do not guarantee citation, but they reduce ambiguity during retrieval.

Adjust traffic forecasts for keyword segments. A keyword ranking first with an AI Overview present should not carry the same traffic forecast as the same position without one. The 58% average CTR reduction from Ahrefs (Ahrefs, 2026) is a useful baseline when adjusting projections for AI Overview-affected queries.

Use a rendering-aware capture method for monitoring. Because AI Overviews require JavaScript execution, plain HTTP requests miss them entirely. A headless browser or a rendering service with awaiting=ai support is necessary for accurate SERP monitoring at scale.

AI Overviews are now a permanent feature of Google Search. With 2 billion monthly users (TechCrunch, 2025) and a 58% CTR reduction for top-ranked pages when present (Ahrefs, 2026), they require concrete changes to how SEO practitioners set traffic expectations and how data engineers capture search results. Earning citations, improving structured data, and using rendering-aware tooling are the practical responses available to anyone whose business depends on organic search visibility.