Editorial
The Gist
- Search is collapsing into conversation. LLMs increasingly answer questions directly, compressing discovery into a single interface and sidelining clicks as the primary interaction model.
- Visibility now matters more than traffic. Brands compete inside model-generated answers, not just on their own properties, forcing a rethink of content distribution, authority and credibility.
- CX now begins before the site visit. The first impression often happens inside an AI response, reshaping funnels, metrics and the role of experience design.
Two years ago, it would have been hard to imagine that the way we searched and interacted was about to change drastically. We had models that were frozen in time through 2020 or earlier. One year ago, it felt like LLMs hit a plateau. And yet here we are.
A new search world is being written as we experience it. The old analogy of "we will build it as we fly" has never been better illustrated.
For over two decades, our central journey to interact with the Web was to go to a search engine, type a query, scan through results, click on the links that caught our attention and then repeat. Tons of open tabs for each new result. Some of the linked pages were useful, some weren't. Then we repeated the process.
Now the landscape is slowly changing. By 2028, half of the searches on the Web will involve an LLM, but, more importantly, we will not be doing as many searches because LLMs will give us what we are looking for without having to go elsewhere. Everything stays in this conversation, in this tab, in this window. Clicking is no longer necessary.
Although we are building the plane as we fly, some elements have emerged that appear to shape this new reality.
Table of Contents
- Frequently Asked Questions About AI-Driven Search and Discovery
- Clicks Are Not as Relevant
- Traffic Is No Longer the Prize
- We Have 2 Audiences: LLMs and Humans
- Models Are Doing Their Own Research
- Look Beyond the Horizon
- Come Up With Different Flavors of Content
- Rethink Your Metrics
- Test Across Models
- A New Moment for CX
- Prepare Your Stack
Frequently Asked Questions About AI-Driven Search and Discovery
As large language models reshape how people find and evaluate information, these are the questions CX, marketing and digital leaders are asking most.
Clear, well-structured, authoritative content with strong contextual framing performs best. Long-form content supports models, while concise summaries and explanations serve human readers.No. Links still matter, but their role is changing. They are now signals for credibility, authority and sourcing inside model responses, rather than the primary mechanism for driving traffic.Models actively search, compare, cross-reference and validate sources. They assess consistency across the web, recent updates, topical depth and external credibility signals—not just what lives on a single site.CX increasingly begins before a site visit. When users receive answers inside AI tools, brands must design experiences that reduce friction between the model’s response and the human’s next step.Success moves beyond pageviews and CTR toward visibility within model outputs, brand inclusion in answers, citation frequency and downstream influence on consideration and trust.Search is shifting from a click-based journey to a conversation-based experience. Instead of scanning links and visiting multiple sites, users increasingly receive synthesized answers directly from models, often without leaving the interface.Humans skim, seek clarity and value brevity. LLMs ingest everything, prioritize structure and evaluate depth, freshness and authority. Effective content must satisfy both simultaneously.
Clicks Are Not as Relevant
Clicks and links were the core elements in the journey for the Web up to now. The Web was built on links, and the link framed the experience it provided us. This is no longer the case. Clicking on things is not as important now, as we get our answers directly. Something else, the model, does the "clicking" for us. This is not to say that links will die (they won't!), but their role is changing for the first time since the advent of web browsers.
Related Article: Waving Goodbye to SEO: Why AI Visibility Optimization Is the Future
Traffic Is No Longer the Prize
We built nice-looking, friendly, smart websites because we wanted people to interact with our content. We tried to grab attention so people would come to our digital real estate. We measured our success by counting the number of visitors to our properties. It was thrilling to see traffic numbers rise.
Today, that's no longer the case. Traffic is plummeting on many sites, and that's not necessarily a bad thing.
We Have 2 Audiences: LLMs and Humans
The focus of our content strategy was on customer experience. But now, our customers are both humans and machines, and we need to please them both. Humans read slowly, and bots read everything. Humans skim, bots are happy to ingest tons of content. Humans look for clarity, bots for structure and authority.
This reality is creating a creative tension, as we need to come up with content that is friendly for humans to consume while, at the same time, carrying the signals that AI engines pick up as they form their answers.
Models Are Doing Their Own Research
The significant shift is happening underwater. AI models are not passively crawling and waiting for their content. They search, crawl, compare and validate. They keep learning on top of your updates. A simple prompt can trigger dozens of checks behind the scenes. Your content is competing in a network of sources, where it is evaluated for depth, clarity, freshness and credibility.
The New Rules of AI-Driven Discovery
How core assumptions about search, content and customer experience are shifting as LLMs become primary discovery engines.
| Old Web Assumption | What’s Changing | What It Means for Brands |
|---|---|---|
| Clicks drive engagement | Models retrieve and synthesize answers directly | Optimize for quotability, clarity and authority—not just CTR |
| Traffic equals success | Answers increasingly stay inside AI interfaces | Measure influence, inclusion and model visibility over visits |
| One audience: humans | Humans and LLMs consume content differently | Create parallel content strategies for people and machines |
| Crawling is passive | Models actively compare, validate and refresh sources | Freshness, depth and credibility become continuous requirements |
| CX starts on your site | First impressions happen inside AI answers | Design experiences that reduce friction between model and human |
| Stack follows strategy later | Composable, readable systems enable visibility | Build stacks that publish fast, stay open and remain legible to models |
So, in the light of these new trends, what can we do about them? What can organizations do to prepare themselves for this new reality?
Related Article: SEO vs. AISO: What AI Search Optimization Means for Brand Strategy
Look Beyond the Horizon
Don't build or optimize for today's models and standards. Think about the new generation of models that are being built as I write, which will interact with your content in richer ways. Context windows are increasing, big players are adding "memory" to the experience, and contextualization is emerging.
Think about how your content will be considered, not only in your properties but also outside them. What are people talking about your brand in other places? How active are you on Reddit? How much attention are you paying to the outlets you didn't care about before, but that now give LLMs the signals they are looking for?
Come Up With Different Flavors of Content
Write for them both. And this likely means that you will have long-form content for bots, and shorter versions for humans. Verbose product pages will make bots happy, but will distract humans. You will need both.
Rethink Your Metrics
As we mentioned above, traffic patterns are evolving, and thus, how we measure success has to grow as well. Successful campaigns are no longer measured the same way, or by new site launch page visitors, CTR or the like. Not changing the way we measure would be like judging the success of a new movie by DVD sales in a streaming world.
Test Across Models
There are more players in this new world. People will search for content in different places using different mechanisms. Many big players are offering large LLMs, but smaller models will gain relevance. The only way to test what works for your organization is to be testing continually. And not only testing as in QA, but testing as in experimentation. You will have to be very intentional about this experimentation, as it will be a bit messy and not easy (and likely expensive).
Webinar
On demand
Empowering Non-Profits: Smarter Crisis Communication and Community Engagement
The cost of miscommunication is measured in more than words. Learn to deliver outreach that's fast, clear and trusted.
Watch Now
Webinar
On demand
Roundtable: Turning Real-Time CX Signals into Business Results
Four big brands. One live, unscripted discussion on how modern CX teams move from dashboards to real impact.
Watch Now
Webinar
On demand
Unlock Connected Service: How to Forecast, Staff & Support Every Channel
Stop juggling tools. 73% of CX leaders say silos damage CX. Build a seamless service operation instead.
Watch Now
Webinar
On demand
Beyond Composability: How Modern Marketers Build Connected Experiences
Ready to launch campaigns faster, personalize smarter and prove your marketing ROI? Discover the power of a modern DXP.
Watch Now
Webinar
On demand
Rebrand. Migrate. Optimize. How to Do It All (Without Slowing Down)
Cresta leveled up site speed, design flexibility and marketer sanity (in record time). Find out how.
Watch Now
Webinar
On demand
From Manual to Magical: How AI Transforms CX Teams
Learn how to replace manual support processes with automation that actually delivers.
Watch Now
A New Moment for CX
When a potential client receives an answer in an AI chat, they skip entire steps in the old funnel. This will force us to rethink what a good experience looks like, especially when a larger portion of it will be out of our control. Brands that understand this will design experiences that lower the friction between the model and the human.
Prepare Your Stack
I intentionally put this at the end to reinforce that a tech stack will not solve your problems, but it is just as important as your strategy. Start thinking about a stack that will enable you to move quickly in this uncertain terrain. Think again about simplicity, components and composable content. Your stacks need to be more open, as information that sits behind slow or unstructured formats becomes invisible to models, and invisible brands do not get chosen. Your stack must enable you to publish faster, be readable, quotable, and trustworthy.
We are entering a moment where brands win or lose before a client even lands on their site. The first impression happens now inside a model. The shift is significant, but so is the opportunity. Brands that adapt early will shape how model talk about their category. They will set the tone of the questions people and models ask. This is our moment to rethink the way we work, the stacks we assemble, and the story we share.
Learn how you can join our contributor community.
