For a long time, searching the internet followed a familiar ritual.
You had a question. You typed a few words into a search box. You received a list of links, opened a handful of pages, compared what you found, and eventually arrived at an answer.
It wasn’t always efficient or accurate. The roles were clear: search engines found the information, and we made sense of it. That relationship is changing.
Today, you can ask an AI assistant to research a subject, compare competing products, recommend a solution, or explain something you don’t understand. You might receive a complete answer without opening a single website.
Tomorrow, you might not even need to ask every question yourself. An intelligent agent could potentially anticipate certain needs, discover relevant options, evaluate alternatives, and execute decisions with your authorization.
What happens when technology moves from finding information to interpreting it, making recommendations, and acting on our behalf? That’s the question at the heart of After the Search.
Search has never stood still
It’s tempting to tag the current transformation as the death of SEO or search. But that view overlooks something important. Search has been reinventing itself for decades.
Early internet directories organized websites into categories.
Search engines made it possible to retrieve information from an expanding web.
Google’s rise popularized a model built around relevance, authority, and a familiar page of ten blue links.
Over time, those links were joined by images, videos, maps, shopping results, knowledge panels, featured snippets, and increasingly sophisticated answers.
Meanwhile, discovery expanded beyond search engines.
People started finding information through social media, online communities, video platforms, marketplaces, and recommendation algorithms. Sometimes they searched deliberately. Other times, information found them.
Each transition changed how people discovered information and how businesses competed for attention. Yet the underlying need remained remarkably consistent.
People wanted answers. They wanted to make decisions and discover things they didn’t already know existed. The interfaces changed, the need didn’t.
Artificial intelligence represents another chapter in that evolution, and its implications could be considerably broader. This time, the transformation is happening across the entire process of discovery.
The search result is no longer necessarily the destination
Search engines operated as gateways for much of the internet’s commercial history.
They helped people navigate toward information published elsewhere. For businesses and publishers, that created a recognizable exchange: make information discoverable, attract visitors, and create opportunities to build relationships with those visitors.
That exchange was never perfect. Search engines increasingly answered questions directly, and other platforms developed their own ways of keeping users within their ecosystems.
But generative AI introduces another complication.
An AI-generated answer can draw upon information published across the internet while reducing the need to visit the websites that supplied it.
And we’re starting to see evidence of how that changes behavior.
Research published by Pew Research Center in July 2025 examined Google searches conducted by 900 American adults during March of that year. It found that users clicked traditional search results during 15% of visits without an AI summary, compared with 8% when an AI summary appeared. Only 1% of visits involving an AI summary resulted in a click on a cited source.
The findings are important, but shouldn’t be mistaken for evidence that AI summaries alone caused the difference. People encounter different types of results for different queries, and not every search has the same underlying intent.
But the broader question is difficult to ignore.
What happens to the economic relationship between search engines, publishers, and businesses when information can create value without generating a website visit?
For marketers, this creates a measurement problem.
For publishers, it raises questions about the incentives to produce original information.
And for search platforms, it introduces difficult questions about attribution, transparency, and the long-term sustainability of the information ecosystem.
The traditional relationship between visibility and traffic is becoming harder to interpret.
Yet much of our industry continues to approach this transformation through the familiar language of rankings, citations, and optimization. Those metrics have their place. But they describe only parts of a much larger system.
Discovery is becoming increasingly fragmented
The rise of AI assistants is interesting when you see it alongside another development: people were already discovering information across an expanding collection of platforms.
Think about how someone discovers a new restaurant, researches a software product, follows breaking news, or learns a new skill.
The journey might start with Google. It might start with YouTube, Reddit, Instagram, or a recommendation from someone they trust.
Increasingly, it might begin with a conversation with an AI assistant.
And those journeys aren’t mutually exclusive.
Someone might discover a product through social media, research it through an AI assistant, validate the recommendation through online communities, and eventually purchase it through a marketplace.
This fragmentation is particularly visible in news consumption.
The Reuters Institute’s 2026 Digital News Report found that weekly use of AI chatbots for news had increased from 7% in 2025 to 10% in 2026 across its surveyed markets. Yet only 1% identified AI as their main news source.
That contrast is revealing.
Newer discovery interfaces can gain adoption without immediately replacing established ones. They may serve different needs, coexist within the same journey, or gradually change the role of existing platforms.
News consumption is only one example, and we shouldn’t apply its findings to every category of online discovery. Nevertheless, it illustrates why treating the future as a competition between traditional search and AI search may be too simplistic.
The more interesting question is how these systems interact.
What happens when people discover information through one platform, evaluate it through another, and act through a third?
And how should businesses understand their visibility when they can no longer observe the entire journey?
Perhaps the future of search isn’t a single destination.
Perhaps it’s an increasingly interconnected collection of discovery experiences.
When discovery becomes disconnected from traffic
For businesses, the challenge isn’t limited to understanding where people discover information.
It’s understanding what that discovery is worth.
For years, search marketing has relied on a reasonably familiar measurement model. Rankings created visibility. Visibility generated clicks. Clicks brought people to websites, where their behavior could be measured and connected, however imperfectly, to commercial outcomes.
AI-powered discovery complicates every stage of that model.
Consider two findings from research published by Ahrefs in February 2026.
An analysis of approximately 76,000 websites found that Google sent 190 times more traffic to websites than ChatGPT. While Google accounted for nearly 40% of website traffic in the dataset, ChatGPT contributed just 0.21%.
Another Ahrefs study, examining 300,000 keywords alongside Google Search Console data, estimated that Google’s AI Overviews were associated with a 58% reduction in click-through rates for top-ranking pages.
Together, these findings reveal an interesting contradiction.
AI is changing how people discover and consume information, yet traditional search continues to generate substantially more measurable website traffic. At the same time, AI-generated answers within search engines appear to be weakening the relationship between rankings and clicks.
This raises a question worth investigating: if discovery increasingly happens without a website visit, are we measuring its value correctly?
The answer may lie partly in what we’re measuring.
An AI-generated recommendation might introduce someone to a brand without producing an immediate visit. That person could later search for the brand directly, consult reviews, ask colleagues for opinions, or return days later through another channel.
Traditional analytics might capture the eventual visit without identifying the earlier interaction that influenced it.
It means we have an attribution problem that referral traffic alone cannot resolve. And perhaps a more fundamental problem with how we define visibility.
We need more experimentation and fewer assumptions
Whenever technology disrupts an established industry, the market for explanations grows almost as quickly as the technology itself.
AI search is no exception.
New optimization frameworks, methodologies, and acronyms are emerging constantly. Familiar SEO practices are being repackaged for generative engines, alongside entirely new theories about how brands should influence AI-generated answers.
Some of those ideas are useful. Others are plausible but insufficiently tested.
And many recommendations are difficult to verify because the systems we’re trying to understand are constantly changing.
Consider something as straightforward as earning a citation.
A citation is an observable outcome. But observing it doesn’t automatically explain why a particular source was retrieved, selected, or referenced. Nor does it establish that the citation influenced someone’s decision.
The same problem applies to many of the numbers discussed in this article.
Research demonstrating an association between AI-generated answers and lower click-through rates is valuable. But it doesn’t automatically establish causation, explain every user’s behavior, or quantify the resulting economic impact.
That’s why methodology matters.
We need to distinguish between what we can observe, what we can reasonably infer, and what remains an untested hypothesis.
We need experiments that can be reproduced, research that acknowledges its limitations, and frameworks that evolve when new evidence contradicts established assumptions.
And perhaps most importantly, we need to connect our understanding of these technologies with the actual problems businesses and consumers are trying to solve.
Otherwise, we risk building an entirely new vocabulary around the same old optimization checklists.
Introducing After the Search
That’s the thinking behind After the Search.
I work in search and content marketing, and much of my professional experience has involved understanding how people find information, how brands compete for visibility, and how content contributes to business outcomes.
The rise of AI-powered discovery has made those questions more interesting. But it has also made me increasingly skeptical of treating every industry development as another collection of optimization techniques.
There’s a much bigger conversation to be had.
After the Search is my attempt to explore that conversation through original perspectives, practical field notes, structured experiments, and research.
I want to investigate how search and discovery systems are evolving, examine the assumptions shaping our industry’s response, and understand the implications for businesses, marketers, publishers, and the broader internet.
That includes questioning ideas I currently believe to be true.
Some experiments will validate existing practices. Others may challenge them. Certain predictions will prove accurate, while others will become reminders of how difficult technological forecasting can be.
The objective isn’t to claim that we already understand the future of discovery. It’s to develop a better understanding of it through evidence, experimentation, and conversations with people who are confronting these changes firsthand.
Over time, I hope After the Search becomes more than a publication.
I want it to develop into a place where practitioners, researchers, marketers, and technology leaders can exchange ideas, challenge assumptions, and collectively understand where discovery is heading.
But that begins with asking better questions, doing the work, and sharing what we learn.
And this time, understanding what happens next may be more important than understanding what came before.
Welcome to After the Search.