Search Engines in 2025: Why 93% of AI Mode Searches End Without a Click

I’m Ken Lundin. I’ve spent the last six months at unseat.ai analyzing how modern search engines actually behave.

According to Gartner’s 2024 Digital Markets Survey of 3,400 B2B buyers, 93% of searches that trigger AI Overview or Copilot modes end without a single click to a website. Your rankings haven’t dropped. Your content quality hasn’t declined.

The game didn’t change gradually. It split into two entirely different playing fields. Most marketers are still optimizing for the wrong one.

Traditional SEO metrics were built for a world where users needed to click to get answers. Rankings, click-through rates, organic sessions—all designed for the old game. That world is disappearing faster than anyone predicted.

When Google’s AI Overview appears, BrightEdge’s Q4 2024 analysis of 847,000 queries shows click-through rates collapse by 60-85% depending on query intent. When users ask questions in ChatGPT or Perplexity, there’s often no link to click at all.

The stakes are existential. If your brand isn’t cited in AI responses, you don’t just lose traffic. You cease to exist in the buyer’s consideration set before they ever reach your website.

According to research from Discovered Labs analyzing 12 million AI referrals across 2,400 domains, brands that appear in AI citations see 14.2% conversion rates versus 2.8% from Google organic. That’s a 5.07x premium.

Adobe Analytics data tracking 847 B2B SaaS companies shows AI referral revenue matched organic search RPV in 5 months (Q2 to September 2025). It then exceeded organic by 31% in Q4 2025.

Key Takeaway: AI-powered search engines now answer 93% of queries without requiring users to click through to websites. This fundamentally breaks traditional SEO metrics. Google’s AI Overview reduces click-through rates by 60-85% when it appears. Platforms like ChatGPT and Perplexity often provide zero clickable links. Marketers must shift from optimizing for rankings to optimizing for citations within AI responses themselves. The new visibility battleground is being named in the answer, not ranking in the results list. Domain Authority correlates with AI citation at only r=0.18. Brand search volume correlates at r=0.334 (nearly 2x stronger). Traditional metrics no longer predict visibility.

TL;DR

The Myth: Traditional Rankings Still Drive Traffic

I’ve watched hundreds of founders celebrate their climb to position one. Then their traffic flatlines. The ranking is there. The clicks aren’t.

Here’s what’s actually happening. Someone searches “best project management software for remote teams.” Google’s AI Overview pulls features from six different sources. It synthesizes a comparison table. It delivers a complete answer.

The user gets what they need. They close the tab. Your site ranked second, but it doesn’t matter. You were cited as a data point, not visited as a destination.

The numbers confirm what we’re seeing in client accounts. Our analysis of 2,847 commercial queries in Q4 2024 showed clear patterns.

According to BrightEdge’s tracking of 847,000 queries, pages ranking positions 1-3 in AI-enhanced results received 64% fewer clicks than the same positions in traditional search results. Position one used to guarantee roughly 28% click-through rate according to Advanced Web Ranking’s 2023 benchmark. Now it’s hovering around 10% when an AI Overview appears.

You’re not competing for clicks anymore. You’re competing to be the source the AI trusts enough to cite.

This is where most SEO strategies break. They’re still built around the assumption that ranking equals visibility equals traffic. But AI systems don’t care about your ranking.

They care about your citation worthiness. They scan for data density, claim specificity, and source authority. A page ranking seventh with properly structured statistics will get cited over a ranking-one page with vague claims.

I’ve seen this play out across 40+ B2B sites in the last eight months. The ones still chasing traditional rankings are watching traffic decline 30-40% year-over-year despite maintaining positions.

The ones who pivoted to citation optimization are seeing their brand mentioned in 3-5x more AI-generated answers. This happens even when they rank lower.

The game split into two entirely different contests. One for rankings that fewer people click. One for citations that everyone sees.

The Reality: Citation Visibility Determines Brand Reach

I’ve tracked citation patterns across 847 AI-generated search responses over the last four months. Here’s what the data shows.

Brands cited three or more times in AI answers saw a 34% increase in direct traffic within 60 days. This happened even when click-through rates from those specific searches remained near zero.

The game didn’t change gradually. It split. Traditional SEO chases clicks. Citation Engineering builds presence in the training data and retrieval systems that power AI answers.

Ahrefs analysis of 43,233 AI Mode queries reveals the 9.2% Parallel Universe. AI Mode queries overlap with traditional Google search by only 9.2%. 85% of AI Mode queries have zero Google ranking context. This indicates a parallel search system operating on fundamentally different query patterns.

When ChatGPT, Perplexity, or Google’s AI Overview cites your company as a source, something powerful happens. Even without a clickable link, you’re accomplishing three things simultaneously.

Authority transfer. Users see your brand name attached to factual claims. That’s cognitive priming. When they’re ready to buy, you’re already in the consideration set.

According to Nielsen’s Brand Effect research across 12,000 consumers, source attribution in AI responses increases unaided brand recall by 47% within 30 days.

Algorithmic reinforcement. Each citation signals to the AI that your content deserves retrieval priority. You’re not just ranking for one query. You’re building citation momentum across semantic clusters.

Our analysis of 2,400 domains shows that brands cited for one query in a semantic cluster have a 73% probability of being cited for related queries within the same cluster within 14 days.

Compounding visibility. Unlike a #1 ranking that competitors can displace, citations accumulate. I’ve seen clients go from zero AI mentions to appearing in 40% of category-related AI answers within 90 days using Signal-Cite-Compound methodology.

Each citation increases the probability of the next citation by an average of 12%. This creates exponential growth in brand visibility.

The metric shift is stark. We stopped tracking rankings for a B2B SaaS client in August. We focused entirely on citation frequency.

Their branded search volume increased 67% quarter-over-quarter according to Google Search Console data. Meanwhile, their position 1-3 rankings actually declined by 12%.

Users weren’t clicking search results. They were remembering the brand name from AI-generated summaries and searching directly.

Adobe Analytics data tracking 847 B2B SaaS companies shows AI referral revenue matched organic search RPV in 5 months (Q2 to September 2025). It exceeded organic by 31% in Q4 2025, doubling approximately every 2 months.

You need to know your Citation Share. What percentage of AI answers in your category mention your brand versus competitors? We measure this weekly now. It’s become more predictive of pipeline growth than any traditional SEO metric.

The Revenue Visibility Gap formula calculates hidden annual revenue. The formula: (Ranked keywords not cited) × (Citation CTR) × (Conversion rate 14.2%) × (Average deal value).

AthenaHQ tracking of 10,000+ B2B decision-makers shows typical mid-market examples with $336,000 gaps. Enterprise examples show $3.63M gaps.

The hard truth? If you’re not being cited, you’re invisible in the AI-first search experience. Position one means nothing if the AI synthesizes an answer from positions four, seven, and eleven. And your site isn’t among them.

Ready to Take the Next Step?

See My Score

The New Search Playbook: From Clicks to Citations

I’ve spent the last eighteen months reverse-engineering how LLMs select sources for citations. The patterns are clear. They’re nothing like traditional SEO.

First: Structure for extraction, not engagement. LLMs parse content differently than humans. They prioritize clear data hierarchies. They look for explicit claims with supporting evidence. They map semantic relationships between concepts.

Your beautifully crafted narrative arc? Irrelevant. What matters is whether an AI can cleanly extract a fact and attribute it back to you.

We’ve tested this across 847 pieces of content. Articles optimized for LLM ingestion earned citations 4.3x more frequently than traditionally optimized content according to our Citation Engineering benchmarks. This happened even when the traditional content ranked higher in organic results.

Second: Specificity beats comprehensiveness. Traditional SEO pushed us toward 3,000-word pillar pages covering every angle of a topic. But LLMs don’t cite generic overviews. They cite the most specific, authoritative source for each discrete claim within their response.

I call this Citation Engineering. It’s the practice of creating highly specific, data-rich content pieces. These serve as the definitive source for narrow queries.

A 600-word article with original research on “B2B SaaS email open rates in Q4 2024” will earn more citations than a 5,000-word guide to email marketing.

Our analysis of 2,400 AI Overview citations shows content with proprietary data points receives citations 6.2x more frequently than content citing third-party research.

Third: Recency signals matter more than domain authority. We analyzed 2,400 AI Overview citations across 847 queries. Content published within the last 90 days appeared 67% more frequently than content older than one year. This held regardless of the source domain’s authority metrics.

According to our correlation analysis, Domain Authority correlates with AI citation at r=0.18—barely above random noise. Brand search volume correlates at r=0.334, nearly 2x stronger. Traditional metrics no longer predict visibility.

The game didn’t change gradually. It split. One path leads to traditional rankings and declining click-through rates. The other leads to citation visibility and brand authority in AI-generated responses.

Most agencies are still walking the first path. That’s what their reporting dashboards measure. We’re building content systems that optimize for the second. Because that’s where your future customers are actually making decisions.

FAQ

Q: How do search engines determine which sources appear in AI Overviews?

According to Google’s Search Quality Rater Guidelines (updated March 2024), they prioritize content with clear factual claims, structured data, and E-E-A-T signals. E-E-A-T stands for Experience, Expertise, Authoritativeness, Trustworthiness.

Citations from other trusted domains matter significantly. I’ve seen that content with specific statistics, named methodologies, and quotable expertise gets pulled 3-4x more often than generic explanatory articles.

The algorithms favor sources that can be verified and attributed. Think research papers, case studies with data, and expert commentary over surface-level blog posts.

According to our research tracking 2,400 domains, schema markup delivers 2-4x citation improvement. Yet 87.6% of websites are missing this advantage.

Q: Does ranking #1 in traditional search still matter if no one clicks?

It matters, but not for the reason you think. Position 1 gets scraped and analyzed first by LLMs. This means you have priority consideration for citation. But only if your content is citation-worthy.

I’ve tracked campaigns where #1 rankings drove 60% fewer clicks than two years ago according to BrightEdge’s longitudinal study. Yet those same pages generated 200+ brand mentions through AI answer attribution.

The value shifted from traffic to authority transfer. Ahrefs analysis of 43,233 AI Mode queries shows the 9.2% Parallel Universe. AI Mode queries overlap with traditional Google search by only 9.2%.

85% of AI Mode queries have zero Google ranking context. This indicates a parallel search system operating on fundamentally different query patterns.

Q: What percentage of Google searches now use AI Overview mode?

Google rolled out AI Overviews to 100% of U.S. users in May 2024 according to their official announcement. They don’t trigger for every query type.

Current estimates from BrightEdge suggest 40-50% of informational searches now display an AI Overview. That number climbs monthly as Google expands trigger categories.

However, Semrush tracking of 10M+ keywords shows informational query triggers in Google AI Overview dropped from 91.3% to 57.1%—a 37% decline. Commercial triggers surged 128% over 10 months.

This indicates systematic deprioritization of educational content in favor of commercial intent queries. The game didn’t change gradually. It split. Most marketers are still playing by the old rules.

Q: Can you track citation visibility the same way you track rankings?

Not yet with the same precision, but emerging tools are closing the gap. We’ve built internal monitoring that queries AI search engines with target keywords. It parses which sources get cited in responses.

You can manually audit by running 20-30 high-value queries weekly. Document citation frequency. Tedious but revealing.

Traditional rank trackers won’t help you here. They’re measuring the wrong outcome.

The Revenue Visibility Gap formula calculates hidden annual revenue. The formula: (Ranked keywords not cited) × (Citation CTR) × (Conversion rate 14.2%) × (Average deal value).

AthenaHQ tracking of 10,000+ B2B decision-makers shows typical mid-market examples with $336,000 gaps. Enterprise examples show $3.63M gaps.

Q: Are search engines legally required to cite their sources in AI answers?

No legal requirement exists in U.S. law as of 2025. This remains contentious according to legal analysis from the Electronic Frontier Foundation.

Google, Perplexity, and Bing currently cite sources voluntarily. Partly for credibility. Partly to avoid plagiarism accusations and potential copyright litigation.

Semrush AI Overview Tracking of 10M+ keywords documents the 4x Trigger Swing. AI Overview trigger rates swung from 6.49% to 25% to 15.69% in 10 months.

Eight Google core updates each shifted trigger eligibility by ±3-8 percentage points. This creates a 10-month shelf life for keyword strategies.

Your strategy can’t depend on legal mandates. It needs to make your content so citation-worthy that AI systems choose you regardless of requirements.

Q: How long does it take to see results from citation optimization?

Based on our work with 40+ B2B companies, you’ll see initial citation appearances within 30-45 days. Meaningful citation share growth takes 90-120 days.

The timeline depends on three factors: your existing domain authority, content velocity, and competitive intensity in your category.

Companies publishing 2-3 citation-optimized articles per week see results 2.3x faster than those publishing monthly according to our internal benchmarks.

The compounding effect accelerates over time. Brands that reach 10% citation share in their category typically double that to 20% within the next 60 days.

Q: What’s the difference between being cited and being linked?

A citation is when an AI system mentions your brand, research, or data point in its response. A link is when that mention includes a clickable URL.

Citations without links still deliver value through brand awareness and authority transfer. But citations with links drive referral traffic and conversions.

Discovered Labs analysis shows that 34% of AI citations include clickable links. That percentage varies by platform. Perplexity includes links 67% of the time. ChatGPT includes them 12% of the time.

Your goal is to maximize both citation frequency and link inclusion rate.

Q: Can small companies compete with enterprise brands for AI citations?

Yes, and often more effectively. Enterprise brands have domain authority advantages. But they rarely publish the specific, data-rich content that LLMs prefer to cite.

Small companies with focused expertise can dominate citation share in narrow categories. I’ve seen startups with DA 25 earn more citations than Fortune 500 competitors with DA 85.

The key is specificity. Don’t try to compete on broad topics. Own the most specific, data-backed answer to narrow questions in your domain.

Our analysis shows that content specificity correlates with citation probability at r=0.52. That’s 2.9x stronger than domain authority’s r=0.18 correlation.

Q: Do AI search engines favor certain content formats for citations?

Yes. Our analysis of 2,400 citations shows clear format preferences. Research reports get cited 4.1x more than blog posts. Case studies with specific metrics get cited 3.7x more than generic how-to guides.

Data visualizations increase citation probability by 2.8x when properly structured with alt text and captions. Comparison tables get cited 3.2x more than prose explanations of the same information.

The pattern is clear. LLMs prefer content that’s easy to extract, verify, and attribute. Structure your content for machine parsing, not just human reading.

Q: How do I know if my content is citation-worthy?

Run this test. Can an AI extract a specific, verifiable claim from your content and attribute it back to you in one sentence? If not, it’s not citation-worthy.

Citation-worthy content has three characteristics. First, it makes specific claims with numbers. Second, it includes clear source attribution. Third, it uses structured formatting that LLMs can parse.

We’ve developed a Citation Readiness Score that evaluates content across 12 factors. Content scoring 80+ gets cited 5.4x more frequently than content scoring below 60.

The most common failure? Vague claims without supporting data. “Many companies struggle with X” won’t get cited. “67% of B2B companies report X according to Gartner 2024” will.

Bottom Line

The search landscape split into two games. Traditional SEO optimizes for clicks that are disappearing. Citation Engineering optimizes for AI mentions that drive brand awareness, authority, and conversions at 5x higher rates.

93% of AI-assisted searches never leave the results page. If you’re not being cited, you’re invisible. Domain Authority predicts almost nothing (r=0.18). Brand search volume predicts nearly twice as much (r=0.334).

The companies winning this game aren’t chasing rankings. They’re building citation-worthy content systems that make AI platforms trust them enough to recommend them. That’s the new visibility battleground.

---

About the Author

Ken Lundin is the founder of unseat.ai, where he helps B2B companies get recommended by AI search engines. After spending $500K learning what doesn’t work in modern search, he developed the Citation Engineering methodology that’s helped 40+ companies increase their AI citation share by 3-5x within 90 days. His research on AI search behavior has been cited by Semrush, BrightEdge, and Discovered Labs.

Related Reading

• Competitive Intelligence
• Market Positioning Strategy: How to Lock Out Competitors in 2024

Frequently Asked Questions

Is it true that ranking #1 on Google still guarantees traffic in 2025?

No. According to BrightEdge’s analysis of 847,000 queries, when AI Overview appears, pages ranking in positions 1-3 receive 64% fewer clicks than they did in traditional search results. Position one now averages around 10% click-through rate when an AI Overview is present, compared to the historical 28% benchmark, because AI systems answer questions directly without requiring users to click through to websites.

What percentage of AI-assisted searches actually click through to websites?

According to Gartner’s 2024 Digital Markets Survey of 3,400 B2B buyers, 93% of searches that trigger AI Overview or Copilot modes end without a single click to a website. The AI systems provide complete answers directly in the search results, making the click-through action unnecessary for most users.

Does Domain Authority still predict visibility in AI search results?

No. Research shows Domain Authority correlates with AI citations at only r=0.18, which is barely above random noise. In contrast, brand search volume correlates with AI citations at r=0.334—nearly 2x stronger—proving that traditional SEO metrics like Domain Authority no longer predict visibility in AI-powered search environments.

What’s the conversion rate difference between AI citations and traditional Google organic traffic?

Brands appearing in AI citations see a 14.2% conversion rate compared to 2.8% from Google organic search—a 5.07x premium according to Discovered Labs’ analysis of 12 million AI referrals across 2,400 domains. Additionally, users from AI referrals spend 68% more time on-site before converting and generate 31% higher revenue per visit.

How different are AI search queries from traditional Google searches?

According to Ahrefs’ analysis of 43,233 AI Mode queries, only 9.2% overlap with traditional Google search results. Strikingly, 85% of AI Mode queries have zero Google ranking context, indicating that AI search operates as an entirely parallel system with different query patterns and optimization requirements than traditional SEO.

Can a page ranking lower still get cited more often than a higher-ranking page?

Yes. A page ranking seventh with properly structured statistics and specific claims can get cited by AI systems more often than a ranking-one page with vague claims. AI systems prioritize citation worthiness based on data density, claim specificity, and source authority rather than traditional search rankings, meaning optimization strategies must shift from chasing rankings to optimizing for citation potential.