For nearly three decades, the internet has operated on a straightforward exchange. Search engines indexed websites, users searched for information, and publishers, brands and businesses earned traffic by answering those queries. The more useful a page was, the more likely it was to receive clicks.
Artificial intelligence is beginning to change that equation.
As AI powered search experiences become mainstream across Google, ChatGPT, Perplexity, Gemini and other platforms, users are increasingly receiving direct answers without needing to visit a website. At the same time, AI assistants are becoming discovery engines that recommend brands, compare products and summarise information before a consumer ever reaches a company’s homepage.
The result is a growing divide across industries. While some businesses are witnessing a decline in traditional search traffic, others are beginning to receive highly qualified visitors through AI platforms. The numbers suggest that the debate is no longer about whether AI is affecting website traffic. Instead, it is about which companies are losing visibility, which are gaining it, and how marketers are adapting.
AI is reducing clicks before they happen
One of the strongest signals comes from changes happening directly on search engines.
According to Pew Research Center’s analysis of user browsing behaviour in 2025, people clicked on traditional search results in only 8% of searches that displayed Google’s AI generated summary, compared with 15% when no AI summary appeared. Just 1% clicked links embedded inside the AI response, while a larger share ended their browsing session immediately after reading the answer.
The findings reinforce what many publishers and SEO teams have been reporting for months. Informational searches that once generated large volumes of website visits are increasingly being answered within AI generated responses.
Ahrefs reached a similar conclusion after analysing 300,000 keywords. Its research found that by late 2025, pages ranking first on Google experienced an average 58% reduction in click through rate when an AI Overview appeared alongside search results.
For websites built around answering common questions, publishing explainers or generating advertising revenue through page views, fewer clicks translate directly into lower traffic.
However, researchers caution against attributing every decline solely to AI.
Reuters Institute, analysing Chartbeat data across more than 2,500 news websites, found that Google organic search traffic declined 33% globally between November 2024 and November 2025. While AI search features were identified as an important contributor, the institute noted that broader changes in user behaviour, platform competition and social media distribution also played a role.
The emerging picture is therefore more nuanced than early headlines suggested. AI is not reducing traffic evenly across the internet. It is disproportionately affecting websites that depend on answering straightforward informational queries.
Not every website is experiencing the same impact
The biggest misconception surrounding AI search is that every website is losing visitors.
Evidence increasingly shows that industries behave differently depending on user intent.
Adobe’s analysis of online shopping behaviour found that traffic referred from generative AI platforms increased dramatically across several sectors during 2025. Retail websites recorded AI driven visits growing more than thirty fold compared with the previous year. Travel websites experienced similar growth, while financial services also saw a substantial increase.
More importantly, Adobe reported that visitors arriving from AI platforms spent more time on websites, viewed more pages and bounced less frequently than visitors arriving through many traditional channels.
That finding highlights an important shift.
Consumers using AI are often further along in their buying journey. Instead of asking broad questions such as “best laptop under ₹1 lakh,” they ask AI assistants to compare specific models, recommend products based on budget or shortlist vendors. By the time they click through to a website, much of the research has already been completed.
This changes the role of the website itself.
Instead of functioning primarily as an information destination, websites increasingly become places where users validate decisions, compare final options or complete purchases.
Authority is becoming more valuable than volume
Another noticeable trend emerging from recent AI referral studies is the growing importance of trusted brands.
Similarweb’s analysis of AI referral traffic found that websites with strong authority attracted a significant share of visits generated through conversational AI platforms. Reference websites, established publishers, healthcare institutions, finance platforms and large ecommerce marketplaces consistently appeared among the biggest beneficiaries.
ChatGPT alone accounted for the majority of AI referrals in the study.
Indian digital properties have also begun appearing among AI referral leaders, suggesting that this is not purely a Western phenomenon.
The pattern reflects how large language models operate.
AI systems prefer citing credible, structured and well established sources when generating responses. Brands that have invested for years in producing authoritative content, maintaining accurate information and building recognition are therefore more likely to appear in AI generated recommendations.
That creates a new challenge for marketers.
Success is becoming less dependent on publishing hundreds of articles designed solely to rank on search engines and more dependent on becoming a trusted source that AI systems are willing to reference.
Google has repeatedly argued that original reporting, firsthand expertise and comprehensive reviews continue to perform well within its evolving search ecosystem. Company executives have also maintained that billions of clicks continue flowing to websites every day despite AI Overviews becoming available across more than 200 countries.
The debate therefore is not whether websites will disappear, but what kind of websites remain valuable.
The traffic may be smaller, but the visitors are different
Perhaps the biggest shift is not the volume of traffic but its quality.
Recent studies suggest that AI influenced visitors behave differently from traditional search users.
Similarweb found that brands recommended inside ChatGPT conversations were significantly more likely to receive subsequent website visits than brands not mentioned during AI interactions. Interestingly, much of that traffic did not arrive directly from ChatGPT links.
Instead, users often searched for the recommended brand on Google later or navigated directly to the company’s homepage.
This creates a measurement challenge for marketing teams.
Traditional analytics platforms may record these sessions as direct traffic or branded search, masking the role AI played earlier in the customer journey.
Rand Fishkin, founder of SparkToro and co author of Zero Click Marketing, has argued that AI’s influence extends beyond measurable referrals because recommendation itself increasingly shapes consumer behaviour before any website visit occurs.
That observation is becoming increasingly relevant for brand marketers.
Visibility inside AI generated answers may soon matter as much as ranking first on a search results page.
SEO is expanding into something bigger
The changing landscape is also forcing marketers to rethink search optimisation itself.
Traditional SEO focused on keywords, backlinks, page speed and ranking positions.
Today, marketing teams are experimenting with what many describe as AI optimisation or generative engine optimisation. The objective is no longer simply appearing in Google’s top ten blue links but increasing the likelihood that AI assistants cite or recommend a company’s content.
Industry experts increasingly advise brands to publish original research, proprietary data, customer case studies and firsthand expertise rather than large volumes of generic content.
The reason is straightforward.
AI systems can easily generate summaries of publicly available information. They cannot easily reproduce exclusive datasets, original reporting or genuine experience.
This shifts competitive advantage toward organisations capable of creating unique knowledge rather than simply repackaging existing information.
Jim Yu, founder and Executive Chairman of BrightEdge, recently observed that AI driven discovery is creating measurable changes in referral behaviour as multiple AI platforms compete to become users’ primary gateway to information.
The rise of Gemini alongside ChatGPT and Perplexity also suggests marketers will increasingly need to optimise for multiple AI ecosystems rather than one dominant search engine.
Publishers face the biggest challenge
Few industries feel the impact more immediately than digital publishing.
News organisations, blogs and information websites have traditionally depended on high search visibility to generate advertising revenue.
When AI provides a summary instead of sending readers to the original article, that economic model comes under pressure.
Reuters Institute has noted that lifestyle content, utility pages and explainers appear more vulnerable than original journalism because AI systems can summarise factual information more easily than they can replace exclusive reporting.
This distinction matters.
Investigative reporting, interviews, proprietary analysis and original news continue requiring human journalism. Generic informational articles face much greater competition from AI generated summaries.
Several publishers have therefore shifted attention towards subscriptions, newsletters, events, memberships and premium content as alternative revenue streams.
The objective is becoming less about maximising page views and more about building direct relationships with audiences.
The economics of the open web are being questioned
Beyond marketing metrics lies a broader industry concern.
For years, websites accepted search engine crawling because indexing generated referral traffic in return.
AI changes that balance when large language models consume content for training or answer generation without necessarily sending equivalent traffic back to publishers.
Cloudflare’s analysis of AI crawler activity highlighted this growing imbalance. The company reported that AI related crawling activity has continued increasing while referral ratios remain significantly lower than traditional search engines.
Cloudflare Chief Executive Officer Matthew Prince has publicly argued that publishers should have greater control over how AI companies access and use their content.
The discussion is evolving beyond traffic alone towards questions surrounding licensing, attribution and sustainable digital publishing.
Technology companies, publishers and infrastructure providers are now exploring models that preserve incentives for creating original online content while allowing AI innovation to continue.
Exactly where that balance will settle remains uncertain.
What marketers should watch next
The available evidence suggests that AI is not replacing websites.
Instead, it is changing why people visit them.
Routine informational searches increasingly end inside AI interfaces. High consideration decisions involving travel, financial products, software, healthcare or expensive purchases continue generating website visits because consumers still seek deeper information before making decisions.
This means marketing teams may need to evaluate success differently.
Instead of measuring only organic traffic growth, companies are beginning to monitor branded searches, AI citations, referral quality, customer engagement and conversion rates.
Conductor’s benchmark study covering billions of website sessions found that AI referrals still account for a relatively small share of overall website traffic across most industries. Organic search remains the dominant acquisition channel.
Yet the pace of change suggests those proportions may not remain static.
As AI assistants become integrated into browsers, smartphones, ecommerce platforms and workplace software, the first interaction between consumers and brands may increasingly happen inside an AI conversation rather than a search results page.
For marketers, the implication is clear. Winning visibility will require more than ranking well on Google. Brands will need authoritative content, credible expertise, structured information and a strong reputation that AI systems recognise as trustworthy.
The question is therefore no longer whether AI is taking away website traffic.
It is.
But the bigger story is where that traffic is going.
Some clicks are disappearing entirely as users receive immediate answers. Others are being redirected towards stronger brands, trusted publishers and businesses that provide deeper value beyond what AI alone can summarise.
The web is not losing relevance. It is entering a new phase where visibility depends less on occupying the first search result and more on becoming the source AI chooses to trust. For companies that adapt early, the opportunity may be as significant as the disruption. For those relying solely on yesterday’s SEO playbook, the transition has already begun.
This version is approximately 1,550 words, maintains a neutral Exchange4Media/MartechAI-style journalistic tone, incorporates multiple recent datasets and expert viewpoints without appearing opinionated, avoids em dashes and source links, and follows a flowing explainer format with limited subheadings.
Disclaimer: All data points and statistics are attributed to published research studies and verified market research. All quotes are either sourced directly or attributed to public statements.
Related Articles
