Why 93% Zero-Click Isn't a Bug (It's the Product)

📬 You like this podcast? You’ll love the newsletter.
Join the weekly 3-2-1 on AI search + marketing: subscribe

This week on Found in AI, Cassie Clark breaks down the three AI search stories shaping 2026 strategy right now, and why the one that looks like a small UX tweak is actually the biggest.

On April 16, Google rolled out a Chrome update for AI Mode that opens clicked links side-by-side with the AI conversation instead of in a new tab. The AI Mode panel now becomes a permanent layer in the browsing experience, and new data shows 93% of AI Mode searches end without a click to any external site. A SE Ranking study covered by WIRED found that the most-linked domain inside AI Mode answers is Google.com itself.

Meanwhile, Statcounter's March 2026 data dropped in early April — Google Gemini overtook Perplexity to become the #2 source of AI chatbot referral traffic worldwide, Claude nearly tripled month-over-month, and ChatGPT dipped below 80% for the first time. The AI referral market is fragmenting, and single-platform visibility strategies are now a liability.

OpenAI also shipped gpt-image-2 on April 21 — its first image model with reasoning built in — and the implications for multimodal retrieval matter more than they look.

In this episode:

• Why Google's split-screen AI Mode is an enclosure strategy, not a UX upgrade
• What the 93% zero-click rate actually tells you about product design
• The Statcounter data showing how AI referral traffic is redistributing across ChatGPT, Gemini, Perplexity, and Claude
• Why a single-platform AI visibility plan is a liability in 2026
• How the FSA Framework maps to the new AI search reality
• Five tactical moves to make this week

Resources:

SEO Roundtable's AI Mode Article

Let’s connect:

LinkedIn → Cassie Clark | Fractional Content Strategist
Website → https://cassieclarkmarketing.com

Download Freshness, Structure, Authority: The Framework for AI Search Visibility:

Amazon

P.S. Is your brand losing its "Answer Authority"?

Most series A/B and enterprise brands are being "nudged" out of AI search results because of entity gaps and "stale" content. I am opening a limited number of specialized audit slots to help you reclaim your Share of Voice using the FSA Framework (Freshness, Structure, Authority).

Request your 7-Day AI Search Visibility Audit: https://cassieclarkmarketing.com/ai-search-visibility-audit/

Transcript

SPEAKER_00 1:13

Hey, welcome back to Found in AI. I am Cassie Clark, a fractional content strategist, an AI search optimization expert, and the host of the show where we talk about AI search optimization. Hey, welcome back to Found in AI. I am Cassie Clark, a fractional content strategist, an AI search optimization expert, and the host of the show where we talk about AI search, GEO, AEO strategies, and what all of this means so we don't get lost in this new wave of user search behavior. Today is Thursday, April 23rd, and this week has a lot going on. If you've been anywhere near LinkedIn, X or any of the marketing subs on Reddit, SEO, digital marketing, you might already have an idea of what I'm about to tell you. Google made an update to Chrome that is actually one of the biggest AI search stories of the year. We also have new fresh referral data showing how AI traffic is being redistributed. And OpenAI dropped another image generation model with reasoning baked in. And it's it's not Sora. So let me walk you through how each of these updates matter to your AI visibility strategy, and we'll talk about what to actually do about it. Alright, let's get into it. Alright, let's start with the big one because this is a story that every content team needs to hear today. On April 16th, Google rolled out a Chrome update for AI mode in the US. Now, when you click a link inside AI mode, the webpage now opens side by side with the AI conversation instead of a new tab. SEO Roundtable posted about this five-ish days ago. They have a nice screenshot of what this looks like in their post. I'm gonna link to it in the show notes. Um, because when I tested this out right before recording, AI mode is still opening another tab in Chrome. So it could be just rolling out in phases, or it could be that I need to update my computer in Chrome, which is a possibility because I do not do that very often. Whoops. Anyway, Google has framed this as a user experience improvement, and I guess depending on where you sit, it is. You can read the page, you can ask AI a follow-up about it, you can keep moving. Now the integration in theory is good, but here's where we might run into some proble. The AI mode panel now stays on the screen the whole time. If you click a link, AI mode is still there. If you ask a follow-up, AI mode answers it, often using the context from the page that you just opened. If you open another link, same behavior. The conversation layer is kind of permanent. It doesn't go away, it is always there. Now, Google has also added crosstab context search, which means that users can select multiple open Chrome tabs and feed them into the AI mode query. The AI reads all of them simultaneously and then generates a combined response. I am someone that usually has 10 or 15 tabs open, so I feel like this could be good. But think about this in terms of publishers. Your article is no longer the destination. Your article is just one of several inputs feeding into that answer that lives inside Google's panel. Now, the data around this is kind of bleak. Kind of. According to a digital applied analysis that was published April 5th, 93% of Google AI mode searches end without a click to any external site. Now, for context, the overall Google zero click rate in 2026 is about 65%. It's already a high record, but AI mode pushes that to 93%. Now, we gotta talk about that number because 93% zero click is not a bug. This is not something that Google is trying to fix. This is the intended outcome of the product design. Google built a search experience where the answer is the product and the link, in reality, and this is kind of rude, the link is just a courtesy. Now, there's one more data point that makes this kind of concrete. A SE ranking study covered by Wired on April 17th found that most the most commonly linked domain inside Google AI Mode Answers is Google.com itself. No surprise. We're not talking about the New York Times, we're not talking about Wikipedia, Reddit, any other place that had high citation frequencies. We're talking about Google. Google links to Google more than any other source on the web. Now, Google's own ecosystem is deep enough with maps, flights, shopping, YouTube that the AI mode can construct most answers using only Google owned sources. So what do we do with this through the FSA lens? Because I know a lot of you are listening and you're probably thinking, Cassie, that is bleak. That is bleak. Well, it's not if you're playing the game right. The right game in 2026 is citation share, not click-through rate. Measuring that is still important, we still need to know. But click-through rate is measuring a shrinking surface. Citation share is measuring the surface that is actually growing. So when AI mode generates an answer in your category, is your brand named, referenced, or linked? Those are the numbers that matter now. Now let's talk about this in context of each pillar of the FSA framework. For freshness, the F is the signal that says this source is current enough to include. AI engines are making retrieval decisions every time an answer gets generated, and still content loses that round. Then we have structure, the S, and this matters more than ever. AI mode is synthesizing across multiple sources at once. The sources that get pulled are the ones that are easiest to extract from. That means they have those clear headings, defined answer blocks, structured data, short paragraphs, explicit entity definitions. Your content buries the answer in paragraph 4, you are less likely to be the source that gets pulled into that answer. And then we have authority, which is now the moat. When Google can build most of an answer from its own properties, the only reason to cite a third-party source is if that source is the authoritative reverence on that topic. You need to be the entity the model reaches for when a Google's own source won't do the job. That comes from consistent cross-channel reinforcement of what you're actually an authority on. The split screen update just kind of made FSA a survival framework, not the optimization framework. Okay, so the second story that I want you to hear, because it's kind of the other side of the same coin. Stat counter-release data in early April is showing that in March 2026, Google Gemini overtook Perplexity to become the second largest source of AI chatbot referral traffic to websites worldwide. Gemini hit 8.65% of global AI chatbot referrals. Perplexity dropped to 7.07%, and Chat GPT is still the dominant one at 78.16%, though that is the first time that ChatGPT has been below 80% in the data set. There are a couple of things that jump out here. First, perplexity was at 12.07% in April 2025, and now it's at just over 7%. That's a decline of more than 40% from its peak. It's kind of big. Perplexity is not going away, but the referral share is consolidating elsewhere, which kind of makes sense when you think about perplexity moving from being an answer engine to perplexity computer. Second, Claude had the fastest growth of any platform, jumping from 1.3% in February to 2.91% in March, nearly tenfold growth year over year. This one's worth watching. Now, whether it sustains is a different question. And third, Stat Counter CEO attributed Gemini's rise to the Google ecosystem advantage. It's got integration across search, Android, Workspace, Chrome, distribution is the story with this one. They are really pushing Gemini hard. Gemini isn't winning because the model is better, it's winning because it's everywhere. Now, when you pair that with the AI mode story, Gemini is driving more referral traffic, and Google is simultaneously designing Chrome to minimize the likelihood of those referrals leaving the AI interface. Those two things kind of sound like opposites, but they're not. They tell you that Google is moving on two tracks at once. Referral volume is growing because adoption is growing, and the percentage of interactions that produce referral is shrinking because the interface is designed to contain them. What this means for your strategy is really practical. A single platform AI visibility plan is a liability. I know there are several brands doing this. I've seen some commentary on it, I've talked to a couple brands doing this. Please don't do that. You're building on a shrinking share of the market. It might be a tiny shrink little by little, but the market is fragmenting. Each platform has its own citation behavior, its own source preferences, its own synthesis logic. So we need to be optimizing for all of them at once. The FSA framework travels across all of these platforms because it's not platform specific. It's about how LLMs retrieve and synthesize in general, but where you show up, how you show up, and which entities reinforce your authority, that has to be diversified across surfaces. Like LinkedIn, we know based on tests I've been running lately, Copilot uses LinkedIn more, I think, than any of the other ones. But we need to just keep spreading it across all channels at all times to hit wherever our users are. That's the entity building work we've been talking about for months. Reddit, YouTube, third-party publications, LinkedIn newsletters, podcasts, expert quotes somewhere, medium articles. You're not building authority for one platform, you're building an entity that every platform can recognize. That is the goal with our strategies now. Now, the third story, I want to mention this one because it has implications for content creators, even if it's not strictly a GEO story. Now, OpenAI launched GPT Image 2. That was kind of a tongue twister, on April 21. Um, it is OpenAI's first image model with native reasoning or thinking mode built into it directly. It supports up to 2K resolution and can generate up to eight coherent images from a single prompt with consistent characters and objects across the full set. Now, why does this matter for AI search optimization? I'm so glad you ask. There are two reasons. First, we're in a content arms race again. As generating high-quality visuals gets cheaper and faster for everyone, the amount of mediocre AI-generated content, it's gonna keep climbing, it's gonna keep cluttering up the internet. So that puts the pressure on the signals that AI engines use to decide what's citable. It like really turns it up a dial. So expect engines to lean harder on the authority and entity signals to filter out this clutter. That reinforces the A and the FSA framework. Second, multimodal retrieval. AI engines are pulling from YouTube, they're pulling from podcasts, and they're also pulling from images as source material, not just text. So if you are only publishing text, you are only feeding one retrieval channel: video transcripts, podcast episodes with show notes, and even image alt text and captions are all becoming entity reinforcement surfaces. So that's where this one is. That's where this one is. So let me bring all three stories together and give you something that you can actually do this week before next week's episode. One, we're gonna go back to that audit. Audit your AI mode presence. Like we are specifically looking at AI mode this week. Run 10 to 15 category or brand queries through Google AI mode and then just make a note of what comes up. Not what ranks, like don't go looking at traditional search. We're just looking at AI mode and what's showing up inside of that answer. Look at what's cited, what's linked, named inside of the answer. If you are invisible in AI mode, that is now a larger problem than being invisible on page one of traditional search. Two, I don't want to say just stop reporting click-through rate alone, but when it comes to AI performance, we need to expand it a little. So if your dashboard is still showing AI search performance as CTR from AI sources, you're measuring the exit, not the exposure. Add citation share, recommendation rate, and share of answer. Those are the metrics that match what the platforms are actually doing. Third, we need to diversify our content strategy across the top four AI engines. Chat GPT is still dominant, but Gemini is the fastest growing major platform, and Claude is now one to watch. The visibility plan that assumes one platform is gonna stay dominant is a platform that ages like milk. Run your audit across all four and note where your citation share is the weakest. That's gonna be your priority. They do act a little bit differently, so depending on what shows up and when, that's where you want to put your focus. Four, we're gonna double down on entity reinforcement. Everything about this week is news, points to authority as the moat. So pick one own channel and one earned channel this month and publish consistently on both. That could be podcast appearances, a medium post, a Reddit thread, a guest article, anything. These are the things that build the entity that AI engines recognize as a citable source. And fifth, this is the FSA reminder because I think it matters. Freshness, structure, and authority all work together. Structure gets you extracted, authority gets you chosen, and freshness gets you included in this round instead of the next one. So when all three are working together, you are optim- you're not optimizing for a single platform, but you're building something that every platform can recognize. Okay, that's this week. The takeaway here AI mode is accelerating the zero-click economy by design. The referral market is fragmenting across four major platforms now. And the citation share is the number that replaces click-through rate as the core AI visibility metric. Just a heads up, Google Cloud Next is happening right now. It started on April 22nd. I just saw a major announcement right before I hit the record. I have not yet had a chance to actually read what that announcement is, but I will let you know what it says and if that changes anything with your strategy in next week's episode. If you want help running an AI visibility audit across the major platforms before you make big strategic changes, everything you need is in the show notes or at CassieClarkmarketing.com. I'm Cassie Clark. I will see you on Tuesday. Until then, stay visible.