If your organic traffic dropped this year and your rankings held steady, you're not imagining things and you're not alone. A February 2026 Ahrefs study analyzing 300,000 keywords found that the presence of an AI Overview now correlates with a 58% lower average click-through rate for the top-ranking page. For every 100 clicks a position-one page could historically earn, Google now keeps 58. That's not a penalty, a ranking drop, or a content problem. It's a structural change in how search works.
The instinct to treat a traffic drop as a performance problem, to audit the site, revisit the keyword strategy, check for penalties, is understandable but increasingly misdirected. We're seeing this pattern regularly with clients who come to us concerned about declines that turn out to have nothing to do with their site's quality or their SEO work.
What's Actually Happening
Google's AI Overviews now appear in roughly 13% of all U.S. desktop searches, up from about 6.5% in January 2025, according to Semrush. When they appear, they answer the query on the page. The user gets what they came for without clicking anywhere. Similarweb data shows zero-click searches for news content rose from 56% to 69% between May 2024 and May 2025. Across all search types, 60% of Google queries now end without a click to any external site, according to a Bain survey from early 2025.
The Ahrefs data makes the direction of travel clear. Position one CTR for AI Overview keywords dropped from 0.073 in December 2023 to 0.016 in December 2025. Even informational keywords without an AI Overview saw CTR fall from 0.076 to 0.039 over the same period, a sign that zero-click behavior was already growing before AI Overviews accelerated it. The finding is corroborated by research from Seer Interactive, Kevin Indig, and Authoritas, all pointing to CTR reductions in the 47% to 65% range.
This isn't a temporary rollout issue. It's Google keeping users inside its ecosystem longer. The business model is working as designed.
What this does to your analytics is real: sessions drop, page views drop, the numbers look bad. What it doesn't necessarily mean is that fewer people are seeing your content, knowing your brand, or moving toward a decision. Visibility and clicks have started to decouple in ways that standard last-touch reporting doesn't capture. That decoupling is one of the more significant measurement challenges we're helping clients navigate right now.
The Part About Traffic Quality Is More Complicated Than People Are Saying
There's a narrative circulating that AI-referred traffic is higher quality than traditional organic traffic, that the visitors who do click through are more informed, more intent-driven, and more likely to convert. There's some evidence for this. Ahrefs found that AI referral traffic drove 12.1% more signups for their own platform despite representing less than 1% of total visits.
But that's one company's data on their own product. Other 2025 studies don't consistently replicate it. The honest answer is that it depends heavily on what you're selling, how your content is positioned, and which AI tools are sending the traffic. ChatGPT referrals tend to look different from Perplexity referrals, which in turn differ from those from an AI Overview.
What does seem consistent: visitors arriving via AI referral have already had a version of your content summarized for them. They know roughly what you do before they land. That context either accelerates the conversation or filters out people who were never going to convert anyway. For most B2B sites, that's probably a net positive. But it's not guaranteed, and reporting it as settled fact is ahead of the data. We'd rather give clients an honest read on what's confirmed versus what's still emerging.
What This Means for Your Content
The shift doesn't make content less important. It changes what good content is for.
Content that exists primarily to capture informational search traffic, the "what is X" and "how to do Y" queries, is taking the hardest hit. AI Overviews answer those questions directly. The click doesn't happen. If your content strategy was built around high-volume informational keywords with thin conversion paths, that model is under real pressure.
Content that demonstrates specific expertise, reflects a genuine point of view, or addresses a question that can't be answered with a synthesized paragraph holds up considerably better. Case studies, detailed technical walkthroughs, content that takes a position. These are harder for AI systems to summarize usefully because the value isn't in the answer, it's in the reasoning behind it.
The same applies to content tied to a specific perspective: your clients, your process, your observations from actual work. An AI Overview can explain what WordPress maintenance is. It can't explain how we've seen a specific class of plugin conflict behave across a dozen enterprise sites, or why we make a particular architectural decision for clients in regulated industries. That kind of specificity is still yours. And increasingly, it's the content that gets cited in AI answers rather than replaced by them.
Brand Visibility and EEAT Matter More Than Ever
One thing the zero-click shift has made clearer is that brand visibility and content authority were always the real game. Traffic was just the metric we used to measure them because it was easy to track.
AI systems don't cite sources randomly. The content showing up in AI Overviews, ChatGPT responses, and Perplexity answers tends to come from sources those systems have determined to be authoritative on a topic. That determination overlaps significantly with what Google has evaluated through EEAT for years: demonstrated expertise, a consistent and credible point of view, content tied to real named experience rather than generic information assembled for search volume.
The brands appearing in AI answers aren't there by accident. They built topical authority over time. Their content reflects actual knowledge, not keyword targeting dressed up as insight. And that content gets cited even when it doesn't get clicked, which means brand recognition is building in places that standard analytics don't capture.
This matters for a practical reason. A buyer researching a vendor category might encounter your brand name in three separate AI-generated answers before they ever visit your site. By the time they do arrive, they're not cold. That's a different kind of top-of-funnel than organic search produced, and it rewards the same things EEAT always did: depth, specificity, a recognizable perspective, and content that couldn't have been written by anyone other than you.
For businesses, this means the content investments worth making right now are the ones that demonstrate what you actually know. Real client scenarios, specific technical observations, honest takes on how things work in practice. Not because Google told you to, but because that's the content AI systems pull from when they're assembling an answer about your topic.
What to Watch Instead of Sessions
If organic sessions are no longer a reliable indicator of content performance, the question is what to measure instead. This is something we work through directly with SEO clients, because the answer varies depending on the business model and buying cycle.
Branded search volume is one of the more useful signals. If people are encountering your content through AI-assisted discovery and later searching for you by name, that shows up in branded impressions in Search Console. It's a lagging indicator, but for service businesses with long sales cycles it's often more meaningful than session counts. Assisted conversions tell a similar story. A visitor who encountered an AI summary of your content, visited briefly, and converted three weeks later through a direct session probably doesn't show up cleanly in last-touch attribution. Looking at multi-touch paths for actual leads often reveals activity that standard reporting misses entirely. And increasingly, AI citation tracking is worth adding to the mix. Knowing whether your content is being pulled into ChatGPT, Perplexity, or Google's AI Overviews, and for which queries, is a real indicator of topical authority building in your space even when it doesn't produce a direct click.
The Practical Reality
For most of our clients, this shift is real but manageable. The sites most exposed are those that built traffic primarily around high-volume, low-specificity content. The sites holding up better are the ones producing content that reflects actual expertise: specific, opinionated, tied to real work.
That's always been what good content is. The difference now is that the gap between it and everything else is showing up faster and more clearly in the data. A 58% reduction in clicks for position one isn't a rounding error. It's a signal worth taking seriously.
Zero-click isn't the end of content strategy. It's a filter. The content that was genuinely useful to a human reader is still useful. The content that was optimized to appear useful is having a harder time.
How Curious Minds Can Help
Measuring content performance in a zero-click environment requires a different reporting approach than most teams currently have in place. We work with clients on SEO strategy that accounts for AI visibility alongside traditional rankings: tracking branded search trends, auditing which content is being cited in AI tools, and building content that earns authority rather than just traffic.
That last part is where the work has shifted most. Getting content to rank is one thing. Getting it cited, recognized, and associated with genuine expertise in your category is a longer play but increasingly the more durable one. If your content strategy was built around traffic volume and that volume is declining, the question worth asking isn't how to get the old numbers back. It's what you're actually building toward.
