The relationship between rankings and traffic is weakening
For years, the logic of SEO was straightforward. Improve a page, earn a stronger Google position, attract more clicks and convert some of those visitors into customers.
That relationship has not disappeared, but it is becoming less reliable. Businesses are opening Google Search Console and finding that their average positions remain stable, or even improve, while clicks decline. Impressions may rise at the same time. The organisation appears more visible, yet fewer people reach its website.
This is not always evidence of an SEO failure. It may show that the search experience surrounding the ranking has changed.
Google increasingly presents information through AI Overviews, local results, shopping modules and AI Mode. Users can receive an explanation, compare options and ask follow-up questions without following the traditional route from search result to website.
A page may still rank and help inform the answer. The click, however, can disappear.
AI Overviews are reducing click-through rates
Research published by Ahrefs in February 2026 estimated that the presence of an AI Overview reduced the click-through rate of the top-ranking organic result by 58 per cent.
For informational searches that triggered an AI Overview, the average click-through rate for the first organic position fell from 7.3 per cent in December 2023 to 1.6 per cent in December 2025.
No study can predict the effect on every website. Results differ by industry, device, query and user intent. However, the overall direction is difficult to ignore.
Pew Research Center found a similar pattern. Google users who encountered an AI summary clicked a traditional search result during 8 per cent of visits. When no AI summary appeared, the rate was 15 per cent.
AI-generated answers are not simply another feature occupying space on the results page. They can satisfy enough of the user’s need that opening a separate website no longer feels necessary.
Google is becoming more conversational
The change extends beyond AI Overviews.
Google’s AI Mode allows users to ask detailed questions, explore topics conversationally and submit follow-up prompts. At Google I/O in May 2026, Google said AI Mode had surpassed one billion monthly users and that queries were more than doubling each quarter.
A traditional research journey might involve several searches and visits to multiple websites. An AI-led journey can combine those stages into one interaction.
Someone researching business software could ask for suitable tools, request a comparison, add a budget and refine the options. The system can create a shortlist without requiring the user to visit every supplier.
Content can therefore influence a decision without receiving the website visit that previously demonstrated its value.
Impressions and clicks are becoming decoupled
This pattern is sometimes described as the “great decoupling”.
Impressions continue to grow because pages still appear in results and may be exposed across more searches. Clicks decline because more journeys end inside Google or another AI interface.
That produces apparently contradictory reports. Rankings look stable. Visibility looks stronger. Traffic looks weaker. Each metric is measuring a different part of a changing customer journey.
A rising impression count may show that Google considers a page relevant. It does not guarantee that the searcher still needs to visit it.
Businesses should not immediately assume that declining traffic means their content, technical SEO or external agency has failed. Rankings must be examined alongside changes to search features, query intent, click-through rates and AI citations.
How to identify whether AI search is responsible
The clearest signal is a widening gap between impressions and clicks.
In Google Search Console, compare periods and check whether impressions have increased while click-through rate has fallen. Review the change at page and query level rather than relying only on a site-wide average.
Informational content is often more exposed because AI systems can summarise definitions, explanations and simple processes effectively. Pages targeting “what is”, “how does” and “why” questions may therefore retain their rankings while losing visits.
Commercial pages can behave differently. Someone requesting a quote or completing a purchase may still need to visit a website. Even so, an AI-generated comparison can influence which companies reach that stage.
Search your priority queries manually and examine what appears above or around the organic results. Look for AI Overviews, local packs, product panels and other features that can absorb attention.
The question is not only whether the ranking changed. It is whether the path from seeing the result to visiting the website still exists in the same form.
SEO is not dead, but its role is changing
Abandoning SEO would be the wrong response.
AI search systems still need accessible, useful and credible information. Search rankings, crawlability, site structure, original evidence and authoritative references can all contribute to whether content is discovered and trusted.
The problem is using website traffic as the only measure of that work.
A page may now create value by being cited in an AI answer, improving the system’s understanding of a brand or placing the business into a recommendation set. Those outcomes can influence awareness and consideration even when the page receives fewer clicks.
SEO is becoming part of a wider visibility discipline. The aim is not only to rank a webpage, but to ensure the organisation is represented accurately wherever customers ask questions and compare choices.
Businesses need broader performance measures
Organic clicks and conversions remain important, but they should sit alongside new indicators.
Businesses should assess whether their brand appears in AI Overviews and conversational answers for commercially relevant topics. They should identify which sources are cited, whether descriptions of the company are accurate and which competitors are recommended instead.
Branded search demand can become more significant too. A person may discover a company through an AI answer, remember its name and search for it later. The original influence may not appear as a referral from ChatGPT, Gemini or Google AI Mode.
Direct traffic, assisted conversions, brand mentions and changes in enquiry quality can provide additional context. Attribution is likely to become less tidy as discovery spreads across more interfaces.
One analytics platform can no longer describe the entire customer journey.
Content must earn influence, not only clicks
Content strategies based on publishing large quantities of generic informational articles are particularly vulnerable.
When a page repeats information available across hundreds of websites, an AI system can summarise the subject without giving users a strong reason to visit the original source.
Content becomes more defensible when it offers something distinctive. Original research, first-hand experience, expert analysis, proprietary data, practical tools and clearly evidenced opinions give a source greater value.
Businesses should also create content that supports decisions. Detailed service pages, comparisons, case studies, pricing context and evidence of results help people and AI systems understand when the company is a suitable option.
The objective is not to hide the answer to force a click. It is to provide a clear answer while giving the reader a meaningful reason to go further.
The new question is bigger than rankings
SEO rankings remain valuable, but they no longer guarantee the traffic they once did.
AI Overviews and conversational search are separating visibility from website visits. A company can perform well in conventional search metrics while losing attention at the point where an AI system generates the answer.
The correct response is not to declare SEO finished. It is to expand how search performance is understood and measured.
Businesses should continue improving their websites while monitoring citations, mentions, recommendations and brand representation across AI search.
The next performance report should not ask only, “Where do we rank and how many clicks did we receive?” It should also ask, “When the answer is generated before the click, is our business shaping it?”
AwarenessAI helps organisations understand how platforms such as ChatGPT, Gemini, Perplexity and Google’s AI search experiences represent and recommend their brand. Through AI visibility audits, implementation and ongoing monitoring, businesses can identify where they are absent, misrepresented or losing ground to competitors.