How does TSI optimize websites for yoga studios?

We build mobile-friendly sites with class booking, GEO targeting, and chatbot integrations.

AI Overviews in Search

AI Overviews

Impressions Up, Click-Through Rates Down

TSI Digital Solution

November 15, 2025
17:31
No Comments

With Google’s AI Overviews (generative summaries) appearing on over half of search queries, traditional SEO metrics like clicks and rankings no longer tell the full story. Learn how AI Overviews are changing search behavior and which new metrics and tools to use, from answer inclusion and brand signals to enhanced GA4 tracking and engagement KPIs. Follow our step-by-step plan to modernize your analytics and stay ahead in the AI-driven search era.

Download

Adapting Your Measurement Frameworks to Google AI Overviews

Search is transforming: instead of a list of blue links, Google now provides AI Overviews – short, AI-generated answers with source citations. These overviews appear at the top of the page, summarizing information so users often get the answer without clicking further. They’ve grown explosively: recent data shows AI Overviews now trigger on over 50% of queries (up from 25% just months earlier). This shift has profound effects on how we measure search success. The familiar click-through rates and ranking reports can no longer capture true visibility. We need to revamp our analytics to focus on presence in AI answers, brand mentions, user engagement, and downstream conversions.

What Are Google AI Overviews?

AI-Generated Search Answers

Google’s AI Overviews are advanced summaries at the top of search results. They are powered by Google’s Gemini large language model and combine information from multiple sources to answer complex questions quickly. In practice, an AI Overview might present a concise answer and then list 5-8 relevant sources for “further reading”. Google notes that these overviews act as “jumping-off points” – users get the gist fast and can click cited links if they want more depth. This means search isn’t disappearing links entirely… rather, it’s changing the path users take to content. In fact, Google reports that AI Overviews have led users to visit a greater diversity of websites for complex queries, highlighting new opportunities for content to be discovered.

Explosive Growth and Prevalence

AI Overviews have rolled out rapidly. For example, one SEO analysis found that AI Overviews now appear on more than half of all Google searches. In August 2024 they were in about 25% of queries, but by mid-2025 that figure had doubled past 50%. In practical terms, if you run any informational queries, you’re very likely to see an AI-generated answer above the traditional results. This has led to dramatic changes in user behavior: when AI Overviews appear, research shows the average click-through rate drops sharply (one study found CTR fell by ~34.5% in queries with an AI answer). That’s because many users get their answer from the AI box and never scroll to click on a blue link.

Continued Link Presence and Engagement

Despite this “zero-click” trend, AI Overviews still cite other sites. Each summary links out to multiple webpages, which can drive different kinds of engagement. Users who do click through tend to be further along in the buying or research process. One study noted that traffic from AI summaries often converts at higher rates than traditional search clicks. This suggests quality over quantity: the visits you do get may be more valuable. Still, overall traffic patterns have changed. In short, Google’s official guidance remains: follow strong SEO fundamentals (great content, crawlable pages, good page experience). But understand that how users find and interact with content has shifted dramatically in the AI-overview era.

Why Traditional Metrics Fall Short

The "Crocodile Mouth" Effect

In many analytics dashboards you’ll now see a peculiar pattern: impressions (views) are up while clicks are down. SEO experts call this the Crocodile Mouth phenomenon. After an AI Overview launches, total search impressions for a site can soar (the “upper jaw” of the crocodile), but site clicks hardly budge or even fall (the “lower jaw”). In one real example, a site saw ~79 million impressions but only ~1.5 million clicks, a click-through rate of just 1.9%. The graphic below illustrates this gap:

Figure: The “Crocodile Mouth” pattern of search metrics after AI Overviews rolled out. Impressions (purple) spike much higher than clicks (blue), squeezing CTR

This happens because AI Overviews take prime SERP real estate at the top of the page. Users often get answers directly from the overview and don’t scroll for links. As one expert notes, “total impressions soar, yet clicks drop significantly” when AI summaries are featured. In practice, this means CTR metrics and top-rank percentages will understate your true visibility. A page can rank well but show little click volume simply because users are satisfied with the AI-generated answer above.

Zero-Click Search and CTR Declines

Numerous studies confirm the rise of zero-click searches. Analytics firms report that queries with AI Overviews have dramatically lower CTRs. For example, industry data from eMarketer shows a ~34.5% drop in organic CTR when an AI Overview is present. Single Grain likewise found the average CTR for the #1 organic result fell from 28% to 19% as AI Overviews gained prominence. In other words, fewer users are clicking links for informational queries. While this might look alarming on paper, it simply reflects changing user habits: the answer is provided up front. Marketers should expect total sessions to shrink, but focus instead on the value of those sessions.

Shrinking SERP Real Estate

AI Overviews don’t just occupy the top of the page, they dominate the screen. One analysis found that AI Overviews and featured snippets together take up roughly 75.7% of the viewable space on mobile devices. This leaves traditional results buried below the fold even for page-one rankings. The practical impact is that pages ranking in 1–5 positions may see far less actual visibility than before. In short, competing for clicks alone is no longer sufficient. Your brand now needs to appear in the answer itself or otherwise stand out. This requires shifting from “rank and click” thinking to measuring overall presence and influence in AI-driven search.

Key Metrics for the AI-Driven Search Era

Answer Inclusion & Entity Recognition

Because AI Overviews pull answers from multiple sources, a key new metric is answer inclusion: how often your content is cited in AI-generated answers. Similarly, entity recognition measures whether Google’s AI correctly identifies your brand, products, or people in an answer. Tracking these requires checking AI results for your target queries (manually or with new tools) and noting if your domain or brand is referenced. If your site frequently appears as a source in AI Overviews, that indicates strong AI visibility. These indicators don’t appear in standard tools yet, so you may need custom tracking (e.g. alerts when your URL shows up in an AI answer). Structured data (schema) can also play a role by clearly defining entities on your page, helping AI models correctly surface your content.

Engagement Quality & Conversions

Another shift is focusing on what happens after users encounter your content. With fewer clicks to measure, quality and outcomes matter more than raw traffic numbers. Monitor engagement metrics such as time on site, scroll depth, and pages per session for traffic coming from search. In many cases, traffic from AI Overviews is highly intent-driven. As one analysis notes, “Although overall traffic may decrease, traffic quality improves. Users who click through after engaging with AI responses tend to be more informed and closer to conversion”. In practice, look at conversion rate and goal completions for organic traffic during AI-driven sessions. If conversions are stable or rising even as sessions drop, your strategy may be working. In summary, pivot your KPI’s: emphasize downstream success (form sign-ups, sales, demo requests) rather than just visits.

Brand Signals & Sentiment

AI-generated answers are creating new touchpoints for brand visibility. It’s wise to measure your brand presence in AI contexts. For example, track mentions of your brand or products in conversational AI or on platforms like Perplexity. Some tools (like Peec AI) can analyze LLM responses for how often and in what sentiment your brand appears. Additionally, monitor branded search queries in Google Search Console: a spike in impressions with flat clicks may signal more AI-driven brand queries. Feel free to get creative: one forward-looking team even analyzes the tone of AI answers about their brand (sentiment analysis) and uses that in reporting. Ultimately, consider metrics like “share of AI voice” – what percentage of target queries mention your brand in the AI answer. These brand-level signals will be increasingly important as AI tools grow.

Modernizing Analytics and Reporting

Augmenting Google Analytics and Search Console

Standard analytics tools aren’t built for AI search data, but they can still be adapted. In Google Analytics 4, you can try filtering sessions by source name. For instance, tag queries from ChatGPT, Gemini, or other platforms in your UTM/traffic rules. Some marketers use custom channel groupings or regex filters (e.g. chatgpt/gemini/ai) on the Session source/medium report to highlight suspicious “direct” or “referral” traffic that might be coming from AI tools. In Google Search Console, compare performance before vs. after AI rollouts. Look for pages with high impressions but low clicks – these are likely feeding the AI answer without driving clicks. Also monitor any changes in branded search behavior. Regularly reviewing these reports (with custom date ranges) will surface patterns that pure SEO dashboards miss.

Emerging AI Search Tracking Tools

Recognizing the gap, new SEO tools are adding AI-specific tracking. Platforms like SEMrush, Ahrefs, and others are developing features to check for AI answer inclusion or to simulate AI results. Dedicated services (some startup SEO labs) now scrape AI Overview results to report how often a site appears. One example is Peec AI, which specifically tracks brand visibility and sentiment across AI engines. While none of these tools are perfect yet, consider adding an AI-tracking layer to your analytics stack. For now, combining existing tools (GA4, GSC) with manual checks or API-based monitoring can help bridge the gap.

Integrated Dashboards and Unified Reporting

In this complex landscape, siloed reporting won’t cut it. Many experts recommend creating a unified dashboard that pulls in all relevant AI metrics. For example, one agency prototyped a dashboard combining GA4 data, site audit metrics, and AI answer citations from Perplexity and Google’s new AI search console integrations. In practice, aim to align your analytics so that insights from paid, organic, and AI-driven channels feed into one view. This helps you see how AI Search performance relates to other campaigns. In the near future, shared measurement frameworks and dashboards will be essential (as one strategist puts it, “Phase two involves implementing shared measurement frameworks and dashboards”). Meanwhile, start by clearly labeling data sources in your reports (e.g. “Google AI Search” vs “Web Search”) and including AI-overview KPI’s alongside traditional ones.

Adapting Your Content Strategy

Concise, Authoritative Answers

Content remains king, but how you present it changes. Google’s AI prefers clear, concise answers and authoritative content. Review your key pages and ask: does the answer jump out in the first few sentences? Can an AI system easily extract a snippet? Structuring content with direct question-and-answer sections, clear headers, and summary bullet points helps AI models find and use your content. Emphasize E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) in every piece. Fact-check rigorously and cite sources even within your text. According to industry guidance, reinforcing E-E-A-T signals (good citations, expert bylines, consistent branding) makes AI systems more likely to trust and feature your content. In practice, when optimizing for AI Overviews, trim fluff. Break complex topics into digestible sections, use plain language, and highlight definitive answers. Each piece of content should have at least one “AI-friendly” summary section near the top.

Voice, Conversational, and Multimedia

Generative search is conversational by nature. Make sure your content answers the kinds of questions people ask in natural language and follow-up queries. Add FAQ sections or conversational Q&A formats to capture those nuances. Also optimize for multimodal search: include relevant images, videos, or audio snippets, since AI Overviews increasingly blend text with visual content. For example, use descriptive alt text and captions for images so they can surface in AI answers. Remember that voice search and AI chat assistants pull from these content signals too. In short, diversify your content formats and include rich media, which not only helps with AI Overviews but also prepares for future search formats.

Local SEO and Niche Opportunities

Not every search is dominated by AI. One silver lining: AI Overviews appear in only about 7% of local searches. Local businesses can still capture voice and map traffic with well-optimized Google Business Profiles, consistent NAP info, and local schema. For areas where AI Overviews aren’t prevalent, continue strong traditional SEO and local listings efforts. Likewise, emerging AI features on specific topics (like product answers or flight info) may not fully replace the need for detailed content. Identify niches where your brand can rank in AI answers (e.g. industry-specific queries) and tailor content there. Keep monitoring: as generative AI evolves, areas like medical, legal, or highly technical info may see AI Overviews slower, seize those opportunities early.

Modernizing Your Measurement Framework

Benchmark Your AI Visibility: Start by auditing your current AI presence. Use tools or manual searches to see which of your keywords trigger AI Overviews and whether you appear in those answers. Audit your schema adoption and entity signals.
Set Specific Goals: Define what success looks like in this new context. For example, aim for a certain percentage of target queries where your content is included in the AI answer. Or set goals for maintaining traffic or conversions despite CTR drops.
Track and Analyze: Use GA4, Search Console, and AI tracking tools to monitor the metrics above. Keep an eye on answer inclusion rates, brand mention share-of-voice, and referral spikes from AI systems. Don’t wait for perfect data, work with what you can observe.
Optimize Continuously: Based on what the data shows, update content and strategy. If certain pages are frequently omitted from AI answers, revise them to be clearer or add schema. If GA4 flags new sources, investigate those channels. Think of measurement as an iterative process.
Report New Metrics: When presenting results, shift the story. Report on “visibility in AI Overviews”, engagement depth, and conversions driven by search overall, rather than just organic CTR and rank. Executive dashboards should link AI visibility metrics to business outcomes (e.g. leads or revenue).

Frequently Asked Questions (FAQ)

Q1: What exactly are Google AI Overviews?

AI Overviews are Google’s new generative summaries that appear at the top of search results for complex queries. They use AI (Google’s Gemini model) to provide an immediate answer or overview, synthesizing information from multiple web sources. Typically, an AI Overview will also list several source links that users can explore for more details. The goal is to help people get the gist of a complicated topic faster, with an option to “dig deeper” via the cited links.

Q2: How do AI Overviews affect my website's traffic and analytics?

AI Overviews often reduce traditional click-through rates. Studies show that when an AI Overview appears, the average CTR of organic results drops dramatically. You may see search impressions increase while clicks and sessions stagnate or decline (the “Crocodile Mouth” effect). In practice, many users get their answer from the AI box and don’t click through. This means your site might rank well but record fewer clicks. To adapt, focus on new metrics like visibility in the AI answers, engagement depth, and downstream conversions, rather than raw click counts.

Q3: Why do I need to track new metrics for AI-driven search?

Traditional SEO metrics (rankings, clicks, average position) don’t fully capture performance in the AI era. Instead, track whether your content is cited in AI answers (answer inclusion), if your brand/entities are correctly recognized (entity recognition), and how AI-driven visitors behave (engagement, conversions). For example, an “answer inclusion rate” (the percentage of target queries where you appear in the AI response) is a useful KPI. You should also monitor brand mention frequency and sentiment in AI responses, and downstream goals (like sign-ups) to measure impact. The idea is to link AI visibility to business outcomes.

Q4: How can I identify and measure traffic coming from AI search?

It’s tricky because AI-driven visits often show up as “direct” or have no clear referrer. In Google Analytics 4, you can look for unusual spikes in direct or referral traffic and set up filters (e.g. keywords like chatgpt/gemini) to isolate AI sources. In Google Search Console, compare performance before and after AI rollouts: pages with high impressions but low clicks often indicate AI-overview inclusion. Emerging SEO tools can also help by scanning AI results for your site. Ultimately, you may need to triangulate data from GA4, GSC, and specialized AI-monitoring platforms to estimate AI-sourced traffic.

Q5: Are traditional SEO best practices still important with AI Overviews?

Yes. Google’s own guidance emphasizes that all core SEO fundamentals still apply to AI features. Your pages must still be crawlable, indexable, and follow search policies. Good content structure, mobile-friendly design, and proper schema markup remain crucial. For example, well-structured FAQ or how-to schema can increase the chances of your content being featured as an AI answer. E-E-A-T (Expertise, Authoritativeness, Trustworthiness) is more important than ever, as AI models rely on credible sources. In short, don’t abandon SEO basics – build on them by focusing content on user needs and clarity so that both traditional search and AI-driven search can pick up your content effectively.

The shift to AI-powered search is happening now.

If you need guidance adapting your SEO and analytics to this new reality, our team is here to help.

Contact TSI Digital Solution today or download our AI Search Optimization checklist for step-by-step advice on updating your measurement frameworks and content strategy for Google AI Overviews.

+(62) 813-3936-1507

contact@tsidigitalsolution.com

www.tsidigitalsolution.com
www.tsidigitalsolution.be
www.tsidigitalsolution.nl

Services

Websites

Development
(re)Design
Ecommerce
Technical

Social Media

Creation
Management
Responding
Posts, Shorts & Reels

Graphic Design

Logo
3D-animation
Content Creation
Designs

Copywriting

SEO
CTA
Keywords
Translations

Technical

Ads/PPC
SEO/Backlinks
Speedup
MGB

Photo-& Videography

Shooting
Editing
Drone
Subtitles