The Great Search Migration

The Great Search Migration

The Great Search Migration: Why Your Next Customer Might Come From ChatGPT, Not Google

How LLM chatbots are rewriting the rules of web traffic—and why the visitors they send are worth 4.4 times more than a Google click

If you have been monitoring your Google Analytics dashboard recently and sensing that something is amiss, your instincts are correct. Your impressions may be up and your rankings solid, but those clicks are not materialising as they once did.

Welcome to 2025, where the internet is undergoing its most dramatic transformation since Google became a verb. This time, the disruption is emerging from an unexpected quarter: chatbots.

The Numbers That Should Capture Your Attention

Consider this striking statistic: 58-60% of Google searches now conclude without a single click to any website12. That is not a typographical error. Nearly six out of ten searches provide users with their answer directly on Google's results page, meaning they never visit your site—or anyone else's.

Five years ago, that figure stood at just 25%3.

What makes this particularly noteworthy is the simultaneous emergence of a new traffic source. While Google's click-through rates are declining, AI-sourced sessions surged by 527% between January and May 2025—jumping from 17,076 to 107,100 sessions across analysed websites4.

Some sites are experiencing even more dramatic shifts. One notable example: ChatGPT referrals grew from just 600 visits per month in early 2024 to over 22,000 monthly visits by May 20255—a 3,566% increase in just over a year.

The Zero-Click Phenomenon

The scale of this shift in traditional search traffic warrants close examination.

When Google's AI Overviews appear on a search results page (which they now do for 27.43% of queries, up from just 3.93% in January 20256), the impact on click-through rates is substantial. Organic click-through rates fell 61%—from 1.76% to just 0.61%6. Paid advertisement CTR declined 68%—from 19.7% to 6.34%6. When AI Overviews are present, the zero-click rate reaches 80-83%, compared to 60% for traditional queries78.

These are not theoretical projections. Real businesses are experiencing tangible consequences. DMG Media reported an 89% drop in click-through rates in September 2025, directly attributed to AI Overviews9. News publishers have been particularly affected: the percentage of news searches resulting in zero clicks grew from 56% to nearly 69% between May 2024 and May 202510.

The median publisher experienced a 10% year-over-year traffic decline in the first half of 2025—news publishers down 7%, content sites down 14%11. Even established brands have not been immune. CNN suffered traffic declines of 27-38% year-over-year. HubSpot's organic visits fell from approximately 13.5 million to fewer than 7 million between November and December 202411.

The fundamental reality is this: Google still processes 9-14 billion searches daily12, yet for every 1,000 searches in the United States, only 360 clicks go to the open web. In the EU, the figure is 37413. The open web is being gradually starved of oxygen.

Enter the LLMs: A New Opportunity

As one door closes, another opens. And what is emerging on the other side merits careful attention.

While traditional search traffic declines, AI-driven referrals to retail websites increased 12-fold between July 2024 and February 202514. ChatGPT now dominates this landscape, accounting for 77.97% of all AI-driven visits globally15, with other platforms such as Perplexity (15.10%), Gemini (6.40%), and Claude (0.17%) comprising the remainder.

The adoption is occurring rapidly—remarkably so. 52% of US adults have now used an AI LLM, and among those users, two-thirds report using them "like search engines" for information retrieval12. More tellingly, 58% of consumers now use GenAI tools for product or service recommendations, up from just 25% the previous year16.

What makes this shift truly significant is that some SaaS sites are now receiving over 1% of all their sessions from LLMs417. That percentage may appear modest until one considers that a year ago, this traffic source scarcely existed.

Why LLM Visitors Are Exceptionally Valuable

This brings us to the critical question of quality.

The data reveals that the average LLM visitor is worth 4.4 times more than the average traditional organic search visitor, based on conversion rates518.

To quantify this: according to industry data, the average value per click from Google organic search typically ranges from $0.50 to $2.00 for most industries, with B2B and high-value sectors commanding higher figures19. Using a conservative mid-range estimate of $1.00 per Google organic click, this translates to a value per LLM visitor of $4.40, compared to $1.00 per Google organic visitor.

For businesses in high-consultative industries (Legal, Finance, Insurance, Health)—which account for 55% of all LLM-driven sessions4—the disparity is even more pronounced. Research from SE Ranking indicates that Claude visitors have the highest session value at $4.56 per visit, followed by Perplexity at $3.1217.

The engagement metrics are equally compelling. AI traffic demonstrates 23% lower bounce rates than other traffic sources14. Sessions include 12% more page views and last 41% longer14. The conversion gap between AI and traditional traffic narrowed from 43% to just 9% between July 2024 and February 202514. Users spend up to 9 minutes per session with AI chatbots compared to just over 5 minutes with traditional search20.

The quality differential stems from intent. People using ChatGPT or Perplexity for product research are not casually browsing—they are asking contextual, trust-heavy, consultative questions: "What should I ask a solicitor before signing this contract?" "Is this medication safe with my specific conditions?" "How do I structure payroll as a small business owner with 5 employees?"4

These are high-context moments where users are closer to a decision point. When an LLM recommends your site as a source, that visitor arrives pre-qualified and genuinely interested.

Who Is Capturing AI Traffic Successfully?

The data reveals notable patterns regarding which sites are capturing this valuable new traffic source.

High-consultative industries are leading: Legal, Finance, SMB, Insurance, and Health comprise 55% of all LLM-driven sessions4. This aligns with intuition—these are sectors where people require expert guidance rather than quick facts.

Referral Distribution Across Platforms

ChatGPT remains the dominant player, consistently driving 40-60% of all AI referral traffic across industries17. For news sites, 78.5% of AI referrals come from ChatGPT10, with month-over-month growth rates reaching 150% in some analyses21.

Perplexity demonstrates unexpected strength in specific verticals, contributing 0.073% of Finance traffic and 0.041% in SMB and Legal sectors4. Overall, it holds 15.10% of global AI traffic, rising to nearly 20% in the United States15.

Gemini is emerging but remains nascent, at 0.0075% in Insurance and 0.035% in SMB4, holding 6.40% of global AI traffic15.

Claude presents an interesting quality-over-quantity profile. Despite minimal traffic share at <0.001%17, it delivers the highest session value at $4.56 per visit17.

The top beneficiaries from AI referrals include Yahoo (2.3 million global visits from AI platforms in June 2025), Reuters (1.8 million), The Guardian (1.7 million), and Business Insider (1 million)10.

The Behavioural Shift: From Search to Conversation

What is driving this transformation is not merely technology—it is a fundamental change in how people interact with information.

Traditional search follows a predictable pattern: keyword entry, scan results, click, evaluate, bounce, repeat. It is transactional, fragmented, and often frustrating.

LLM-based search is conversational: initial query, synthesised answer, follow-up question, refined response. It feels less like using a tool and more like consulting an expert.

Consider this example of how users refine queries with LLMs: "What's the best protein powder?" becomes "Which is better for women?" then "Any that ship to London?"20 This iterative refinement creates engagement that traditional search cannot match. LLM sessions outpace traditional search by more than 60% in time spent, despite lower overall adoption rates20.

The engagement differential continues to widen. Nearly 80% of US ChatGPT users now rely on it as a search tool, asking questions in natural language16. Critically, 60% of US adults used an AI chatbot for product research in the last 30 days20.

However, a nuance that many overlook: this is not a complete replacement scenario. 99.8% of ChatGPT users also use Google22. Studies found that after beginning to use ChatGPT, people generally continued using Google at the same rate or even slightly more16. In fact, total Google search volume grew 21.6% from 2023 to 2024, even as AI tools rose in popularity22.

What we are observing is not substitution—it is fragmentation. Users now employ a multi-modal approach: ChatGPT for quick questions or ideation, Google for deep research or verification, TikTok for tutorials, Reddit for reviews ("reddit" is now one of the most searched terms on Google itself12), and AI chatbots for consultative questions.

What This Means for Your Business

For marketers, business owners, and content creators, several key insights emerge:

First, the traditional playbook requires revision. Optimising solely for Google rankings is no longer sufficient. Visibility in AI-generated responses has become essential.

Second, quality now supersedes quantity. One visitor from ChatGPT may be worth more than four from Google. The focus should shift toward earning AI citations in high-intent queries.

Third, the window of opportunity remains open. With AI traffic accounting for just 5-8% of total search volume, we are still in the early-adopter phase. Businesses that optimise now will secure a significant advantage.

Fourth, this transition is accelerating faster than mobile did. The shift to mobile search took years. AI search adoption is occurring in months. The 527% growth in AI referral traffic happened in just five months4.

Gartner predicts a 25% drop in traditional search engine volume by 2026 due to AI tools5. Some analysts project that by 2028, AI search visitors may surpass traditional search visitors23. By Q4 2025, experts predict 20% of B2B sites will receive more traffic from AI than traditional search17.

The Bottom Line

We are witnessing a fundamental reordering of the web's attention economy. The rules of visibility are being rewritten in real-time, and the businesses that adapt will thrive while others wonder where their traffic went.

The encouraging news is that visitors arriving from AI platforms are more engaged, more qualified, and more valuable than traditional search traffic. The challenge lies in fundamentally rethinking how you make your content discoverable.

This is not about abandoning SEO—it is about expanding it. It is about pursuing Generative Engine Optimisation (GEO) alongside traditional SEO efforts. It is about making your content citation-worthy in the age of AI.

The great search migration is here. The question is not whether it will affect your business—it is whether you will be prepared when it does.

References

Footnotes

  1. Jellyfish Training. (2025, July 28). How Zero-Click Searches Are Changing SEO In 2025. ↩

  2. Up And Social. (2025). Zero-Click Searches: Why 60% of Google Users Never Click Through in 2025. ↩

  3. Superprompt. (2025, November 11). Zero-Click Crisis Worsens: 58% of Google Searches End Without Clicks. ↩

  4. Search Engine Land. (2025, August 5). AI traffic is up 527%. SEO is being rewritten. ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7 ↩8

  5. Insightland. (2025). AI Search: traffic killer or the biggest opportunity yet? ↩ ↩2 ↩3

  6. Superprompt. (2025, November 11). Zero-Click Crisis Worsens: 58% of Google Searches End Without Clicks. ↩ ↩2 ↩3

  7. Click Vision. (2025, November 13). Zero Click Search Statistics 2025: Data, Trends & Impact. ↩

  8. Neotype. (2025, October 28). Zero-Click Searches in 2025: Winning in AI Search. ↩

  9. MarTech. (2025). What happens when no one clicks anymore. ↩

  10. Digiday. (2025, July 15). AI is driving more traffic, but not offsetting 'zero-click' search. ↩ ↩2 ↩3

  11. The Digital Bloom. (2025). 2025 Organic Traffic Crisis: Zero-Click & AI Impact Report. ↩ ↩2

  12. TTMS. (2025, October 30). When Will AI Search Beat Google? 2025–2030 Forecast. ↩ ↩2 ↩3

  13. SparkToro. (2025, January 27). 2024 Zero-Click Search Study. ↩

  14. Adobe. (2025). The explosive rise of generative AI referral traffic. ↩ ↩2 ↩3 ↩4

  15. SE Ranking. (2025, September 11). AI Traffic in 2025: Comparing ChatGPT, Perplexity & Other Top Platforms. ↩ ↩2 ↩3

  16. Arc Inter Media. (2025). How LLMs Are Reshaping Content Consumption and Search Behavior. ↩ ↩2 ↩3

  17. Superprompt. (2025, August 12). AI Traffic Surges 527% in 2025. ↩ ↩2 ↩3 ↩4 ↩5 ↩6

  18. Nine Peaks Media. (2025, July 9). Agentic Search vs Traditional Search Engines. ↩

  19. Industry standard benchmarks for cost-per-click and value-per-click. ↩

  20. Francesca Tabor. (2025, July 13). From Queries to Conversations: How LLMs Are Redefining User Engagement. ↩ ↩2 ↩3 ↩4

  21. Search Engine Land. (2024, October 22). SearchGPT gives brands 4x more referrals vs. Perplexity, Claude. ↩

  22. Break The Web. (2025, September 26). AI SEO Statistics: How AI & LLMs Are Reshaping Search. ↩ ↩2

  23. ALM Corp. (2025). How to Rank on ChatGPT, Perplexity, and AI Search Engines. ↩