The End of the Click
What happens when customers no longer need to visit your website to make a decision?
EdgeUp RESEARCH
What Happens When Customers No Longer Need to Visit Your Website to Make a Decision?
Executive Summary
For more than two decades, digital marketing has been built around a fairly predictable customer journey.
Someone has a question.
They search.
They receive a page of results.
They click.
They visit websites, compare their options, and eventually make a decision.
That journey is changing.
Long before artificial intelligence became part of mainstream search, a growing percentage of Google searches were already ending without a click to another website. In 2024, research from SparkToro and Datos found that 58.5% of U.S. Google searches resulted in zero clicks.
Artificial intelligence is pushing that behavior further.
Pew Research Center found that when Google's AI-generated summaries appeared in search results, users clicked traditional search results only 8% of the time, compared with 15% when an AI summary did not appear. Users clicked sources cited inside the AI summary just 1% of the time.
Separate research from Ahrefs found that by December 2025, the presence of a Google AI Overview correlated with a 58% lower click-through rate for the number-one organic search result.
Taken together, these findings suggest something more significant than a decline in website traffic.
The role of search itself is changing.
Increasingly, the technology between the customer and the business is not merely retrieving information. It is interpreting information, comparing alternatives, answering questions and, in some circumstances, helping customers decide.
That creates a new challenge for businesses.
For years, digital marketing asked:
Can customers find us?
The emerging question is:
When customers ask AI what they should choose, will it recommend us?
The Click Became the Scoreboard
The commercial internet trained businesses to think in clicks.
Search engine optimization improved the likelihood that someone would find a website.
Paid advertising purchased opportunities to generate visits.
Social media posts drove people toward landing pages.
Analytics platforms measured what happened after visitors arrived.
The basic logic was straightforward:
Visibility → Click → Website → Evaluation → Decision
Businesses became remarkably sophisticated at measuring that journey.
They could see where traffic came from, which pages visitors viewed, how long someone remained on a site, what advertisement produced the visit, and whether the visitor ultimately made a purchase, reservation, call or inquiry.
The click became an enormously useful signal.
It showed intent.
And, importantly, it created an observable customer.
A person who clicked could be measured.
A person who never arrived could not.
That distinction matters much more in an AI-driven search environment.
Zero-Click Search Did Not Begin With AI
It would be easy to blame artificial intelligence for the decline of clicks.
It would also be wrong.
Google has answered many searches directly for years, including weather, sports scores, maps, business hours, definitions, calculations, featured snippets and knowledge panels.
All reduced the need for users to leave Google's results page.
A 2024 study published by SparkToro using Datos clickstream data examined Google search behavior in the United States and European Union.
The study found that 58.5% of U.S. Google searches resulted in zero clicks.
About 37% of searches ended the browsing session entirely, while approximately 22% resulted in the user changing or continuing the search rather than clicking an external result.
That figure is important.
But it must be interpreted correctly.
It does not mean that AI answered 58.5% of searches.
Much of the zero-click behavior measured in the study had nothing to do with generative AI.
What it does establish is something more fundamental:
Even before AI became deeply integrated into search, the assumption that a search would lead to a website visit was already breaking down.
Artificial intelligence arrived in an environment where users were increasingly accustomed to getting information without clicking.
Then AI Entered the Search Results
In 2024, Google began rolling out AI Overviews, using generative AI to produce synthesized answers directly within traditional search results.
Instead of simply displaying links, Google could now gather information and present an answer above them.
That created an obvious question:
Would people still click?
Pew Research Center investigated that question using actual browsing behavior.
Researchers analyzed the March 2025 web activity of 900 U.S. adults who agreed to have their browsing activity measured.
The difference between searches with and without AI-generated summaries was striking.
When an AI summary appeared, users clicked a traditional search result during 8% of visits.
When no AI summary appeared, users clicked a search result during 15% of visits.
Almost twice as often.
Even more revealing, users clicked a source cited within an AI-generated summary during only 1% of visits.
The presence of AI also appeared to affect whether people continued searching at all.
Users ended their browsing session following 26% of searches containing an AI summary, compared with 16% of searches without one.
Pew did not attempt to prove that every one of those users received a complete answer from AI.
Nor should we.
But the behavioral difference is difficult to ignore.
When AI-generated information appeared, users were significantly less likely to continue to an outside website.
Even the Number-One Result Is Losing Clicks
Perhaps the most consequential evidence comes from what is happening to the top organic result.
For years, reaching the first position in Google search results was one of the most valuable achievements in digital marketing.
The reason was obvious.
Position one captured attention.
Attention generated clicks.
Ahrefs studied whether AI Overviews were changing that relationship.
In its updated analysis published in February 2026, Ahrefs examined 300,000 keywords, half of which generated AI Overviews and half of which were informational searches without AI Overviews.
The researchers compared Google Search Console click-through data from before the rollout of AI Overviews with December 2025 performance.
Their conclusion:
The presence of an AI Overview correlated with a 58% lower average click-through rate for the top-ranking organic page.
The effect was not limited to position one.
Ahrefs reported lower click-through rates across the first page of organic results, including approximately:
- 50.8% lower for position two
- 46.4% lower for position three
- 38.8% lower for position four
- 32.6% lower for position five
This does not mean rankings stopped mattering.
A number-one result is still far more valuable than being invisible.
But the relationship between ranking and traffic has changed.
A business can win the search result and still lose much of the click behavior that historically came with it.
The Question Is Changing
Something else is happening at the same time.
People are learning to ask search engines different kinds of questions.
Traditional search behavior often relied on keywords:
wine bar Long Beach
plumber Pasadena
dentist near me
Those searches primarily asked the search engine to retrieve options.
AI makes much more natural questions possible.
A customer can ask:
“Which wine bar in Long Beach has a large selection by the glass but also serves craft beer?”
Or:
“Who is a reliable plumber in Pasadena who handles emergencies and has experience working on older homes?”
Or:
“Which dentist near me would be best for someone who gets really anxious about dental work?”
Notice what changed.
These are not simply searches.
They contain criteria.
The user is no longer asking only:
Who exists?
They are asking:
Who fits what I need?
Pew's research offers some evidence of this shift.
Only 8% of Google searches containing one or two words generated an AI summary in its March 2025 dataset.
Among searches containing ten words or more, 53% generated an AI summary.
Queries beginning with question words such as “who,” “what,” “when” or “why” generated AI summaries 60% of the time.
AI becomes increasingly relevant precisely when questions become more complicated.
And complicated questions are often where recommendation begins.
Search Is Becoming Part of the Decision
This is the larger issue businesses need to understand.
Traditional search engines primarily helped customers locate information.
AI can increasingly participate in evaluating that information.
Consider the difference.
Traditional Search
Customer question
↓
Search results
↓
Customer opens several websites
↓
Customer compares businesses
↓
Customer chooses
Now consider a conversational AI experience.
AI-Assisted Discovery
Customer question
↓
AI gathers and interprets information
↓
AI compares relevant options
↓
AI provides an answer or recommendation
↓
Customer chooses
The customer may still visit a website.
They may call the business.
They may check reviews.
They may ask another question.
But some of the research and comparison that previously happened across websites can now happen before a customer ever reaches them.
The intermediary is no longer just showing the customer where information lives.
It is increasingly helping the customer interpret it.
This Is No Longer an Edge Case
This shift would be less significant if AI-powered search remained something used only by early adopters.
Google's own numbers suggest otherwise.
In May 2026, Google reported that AI Mode had surpassed one billion monthly users worldwide, only one year after its introduction.
The company also said AI Mode queries were more than doubling every quarter.
Google's incentives are worth remembering here. These are Google's own usage claims, not independent audience measurements.
Still, the scale matters.
AI-assisted search is no longer merely an experiment hidden inside a technology lab.
It is becoming part of how mainstream consumers interact with one of the world's dominant information systems.
The Invisible Lost Customer
This creates a problem that traditional analytics are poorly equipped to measure.
Imagine a customer asks:
“What is the best wine bar in Long Beach if I want a big selection by the glass?”
An AI system evaluates several businesses.
It recommends one.
The customer accepts the recommendation and goes there.
The winning business may never receive a website click.
But it still won the customer.
Now consider the businesses that were not recommended.
They may see absolutely nothing: no declining conversion rate tied to that individual customer, abandoned shopping cart, website bounce, lost lead in the CRM or indication that the business was ever considered.
The customer disappeared before traditional analytics could see them.
EdgeUp refers to this as the invisible lost customer.
The business participated in the decision environment without participating in the measurable customer journey.
This may become one of the most important measurement problems created by AI-assisted discovery.
A Website Can Influence a Decision Without Receiving the Visit
There is another important distinction here.
Fewer clicks do not necessarily mean websites matter less.
That conclusion would be too simplistic.
AI systems still need information.
Business websites can remain important sources for understanding what a company does, where it operates, what it sells, how it describes itself and what differentiates it.
Other sources may matter too, including business profiles, review platforms, directories, local publications, social networks, news coverage, community sites and other third-party sources.
The change is not necessarily that the website becomes irrelevant.
The change is that the website may influence a customer's decision without the customer personally visiting it.
That creates an entirely different way of thinking about digital content.
Historically, businesses wrote webpages primarily for humans who arrived.
Increasingly, those pages may also become evidence used by AI systems helping humans decide.
What the Research Does Not Yet Tell Us
There is a temptation with any major technological shift to run several steps ahead of the evidence.
We should resist it.
Current research gives us good reason to conclude that AI-generated search experiences are associated with fewer clicks.
It shows that zero-click behavior is widespread.
It demonstrates that users engage differently when AI-generated summaries appear.
It shows that top organic search rankings are receiving fewer clicks when AI Overviews are present.
And it demonstrates rapid consumer adoption of AI-powered search.
What existing research does not yet tell us nearly as well is:
How does AI decide which local business to recommend?
Does traditional Google ranking predict the recommendation?
How important are reviews?
Does consistency across multiple sources matter?
Does the language businesses use to describe themselves affect outcomes?
Do ChatGPT, Gemini and other systems recommend the same businesses?
How often does changing one criterion in a customer's question change the recommendation?
What causes a recommendation to move from one competitor to another over time?
Those questions require a different kind of research.
And they matter because declining clicks are only the beginning of the story.
The bigger question is what happens instead of the click.
What Businesses Should Learn From This
Businesses should not abandon SEO.
They should not stop caring about website traffic.
They should not suddenly rewrite every webpage for artificial intelligence or chase whichever new acronym happens to be fashionable this month.
That would be an overreaction.
Instead, businesses should recognize that the customer journey now contains another possible path.
A customer can increasingly gather information, compare options and develop a preference before ever arriving at a business's website.
That means traditional digital metrics tell only part of the story.
Traffic still matters.
Rankings still matter.
Conversions still matter.
But there is another question worth measuring:
When AI participates in the customer's decision, where does our business stand?
What We Learned
The available evidence points to several conclusions.
Zero-click behavior is not new. More than half of U.S. Google searches were already ending without a click in 2024.
AI appears to accelerate the decline in clicks. Pew found traditional-result clicking fell from 15% to 8% when AI summaries appeared.
Ranking first no longer guarantees the traffic it once did. Ahrefs found a 58% lower average click-through rate for the number-one organic result when an AI Overview was present in its December 2025 analysis.
More complicated questions increasingly trigger AI-generated responses. The longer and more question-like the search, the more frequently AI appeared in Pew's research.
And perhaps most importantly:
A customer decision can now occur without producing the digital signals businesses historically relied upon to observe it.
Conclusion
For most of the internet's commercial history, businesses could reasonably assume that online discovery would eventually produce a visit.
The website sat in the middle of the journey.
That assumption is becoming less reliable.
Artificial intelligence is increasingly capable of absorbing information that once required several clicks, interpreting it for the customer and presenting a synthesized response.
For simple questions, the result may simply be an answer.
For commercial questions, something more important can happen.
The answer can become a recommendation.
The click is not dead.
Neither is search.
Neither is SEO.
But the click is no longer the only moment that matters.
Businesses now need to think about what happens before a customer ever reaches them.
Because the next customer may never visit your website before choosing you.
And the customer who chooses your competitor may never visit yours either.
For businesses, that creates a new question:
If the customer never clicks, how do you know whether you were chosen?
That is the research problem ahead.
About EdgeUp Research
EdgeUp studies how artificial intelligence recommends businesses for specific customer questions.
Our research focuses on recommendation outcomes: which businesses are selected, which competitors are selected instead, what information appears to support those decisions, and how those results change over time.
The objective is simple.
Don't just be found. Be recommended.
Sources
SparkToro / Datos. “2024 Zero-Click Search Study.” Published July 2024. Behavioral analysis of Google search activity in the United States and European Union.
Pew Research Center. “Google users are less likely to click on links when an AI summary appears in the results.” July 22, 2025. Analysis of March 2025 browsing behavior from 900 U.S. adults.
Ahrefs. “Update: AI Overviews Reduce Clicks by 58%.” February 4, 2026. Analysis of 300,000 keywords and aggregated Google Search Console click-through data.
Google. “A New Era for AI Search.” May 19, 2026. Google-reported AI Mode adoption and search usage data.
