# Why Does It Feel Like Fewer Buyers Are Finding You Through Google Lately? > Why rankings can stay flat or even improve while actual clicks decline, what AI Overviews are doing to search behavior, and why ranking and visibility have quietly become two different things. Focentra AI · July 17, 2026 · 7 min read Source: https://www.focentra.ai/learn/why-fewer-buyers-find-you-on-google.html Fewer buyers are clicking through to your site because Google is increasingly answering their question directly on the results page, not because your rankings dropped. Zero-click search has been rising for years, but AI Overviews have accelerated it sharply: pages holding the exact same position they always have are seeing meaningfully fewer visits, because the click that used to follow a top ranking increasingly never happens at all. Ranking well and being visible have quietly become two different things, and most teams are still only measuring the first one. ## How AI Overviews Are Rerouting the Click Google Used to Send You An AI Overview synthesizes an answer directly on the results page by pulling from a wider set of sources than the ten blue links below it, then shows that synthesized answer before a user ever has to click anywhere. If the answer is satisfying on its own, the click that used to reliably go to the top-ranking page just doesn't happen. This isn't a small effect at the margins. SparkToro's research (https://sparktoro.com/blog/in-2026-less-than-one-third-of-google-searches-still-send-a-click/) found that 68.01% of US Google searches in early 2026 ended without any click to a website at all, up from 60.45% just two years earlier, and roughly triple the zero-click rate from a decade ago. Zero-click search itself isn't new, featured snippets and knowledge panels have been quietly absorbing clicks for years. What's changed is the scope. Those earlier formats mostly handled narrow factual questions. AI Overviews now synthesize answers across commercial and consideration-stage queries too, the exact kind of question a buyer types while actually evaluating a purchase, which is precisely the traffic a ranking strategy used to be able to count on. ## Why Your Rankings Can Stay the Same While Your Clicks Fall Ahrefs' analysis (https://ahrefs.com/blog/ai-overviews-reduce-clicks-update/) of 300,000 keywords, comparing Google Search Console click-through data from December 2023 against December 2025, found that the presence of an AI Overview now correlates with a 58% lower average click-through rate for the page ranking in the top spot. That's up sharply from the 34.5% reduction Ahrefs measured just eight months earlier, meaning the effect isn't stabilizing, it's compounding. The reason this is so easy to miss is that ranking position and click-through rate are measured by two completely different systems. A rank tracker only tells you where a page sits in the traditional results, it says nothing about whether an AI Overview is now sitting above that result and intercepting the click. A dashboard showing "still ranked #1" can be technically accurate and still be masking a real decline in the only thing that ranking number was ever supposed to be a proxy for: an actual visit from an actual buyer. This is also why the usual first response, refreshing content, building more backlinks, chasing whatever the last algorithm update supposedly rewarded, so often fails to move the number that actually matters. None of that work touches the actual cause. A team that spends a quarter improving a page's on-page SEO while the real gap is an AI Overview sitting above it will see the rank tracker confirm the work paid off, and the traffic dashboard tell a completely different story, because the two are no longer measuring the same underlying reality they used to. ## The Buyers Who Are Still Searching, Just Not Clicking Through These buyers haven't disappeared, they're getting an answer without leaving the results page. Not every one of them behaves the same way once they've read it. A buyer early in research, still forming a general understanding of a category, is often satisfied by a synthesized answer and has no reason to click further, that traffic was always going to be lower-intent. A buyer further along, actually trying to vet a specific vendor or verify a specific claim before a decision, is a different case: they're more likely to click through to a cited source precisely because they want more than a summary, they want to check the underlying evidence themselves. That second group is the one worth paying attention to, because it's both smaller and more valuable than it used to be. As the overall click-through rate compresses, the clicks that do still happen skew more heavily toward buyers who are actively verifying something, which makes each one of those remaining clicks a stronger signal of intent than an equivalent click was two years ago. Losing visibility with that specific group costs more than the raw click-through numbers alone suggest. The practical implication is that a falling click-through rate on an informational query isn't necessarily bad news, a buyer who got a satisfying synthesized answer to "what does this category even mean" was never going to convert anyway. A falling click-through rate on a comparison or vendor-evaluation query is a different signal entirely, because that's the exact moment a buyer is deciding who to shortlist, and an AI Overview that answers the comparison without naming a specific vendor is a lost opportunity that a traffic dashboard alone won't distinguish from harmless informational drop-off. ## Why Being Cited Is Now a Different Outcome Than Ranking The deeper shift is that the source an AI Overview cites is frequently not the page that would have ranked highest for the literal query typed. Ahrefs found (https://ahrefs.com/blog/ai-overview-citations-top-10/) that only 38% of AI Overview citations now come from pages ranking in the traditional top 10, down from roughly 76% a year earlier. Over 31% of citations come from pages ranking 11 to 100, and another 31% from pages ranking beyond 100 entirely. Google's own retrieval process breaks a single question into several sub-questions behind the scenes and pulls sources across a much wider pool than a simple keyword match would suggest, so a page ranking far outside the top 10, if it answers one specific sub-question especially directly, can get cited over a page holding the #2 ranking spot for the main query. That single stat is the clearest evidence that ranking and citation are no longer the same competition. A strategy built entirely around climbing traditional rankings is optimizing for a result that citation selection increasingly ignores. ## What This Means for How You Should Measure Visibility Now Rank tracking alone is now an incomplete signal of visibility, not because it's wrong, but because it only measures half of what actually determines whether a buyer sees you. The other half, whether your content is the one an AI Overview actually cites for the questions your buyers ask, has to be tracked separately, against a defined list of the specific questions that matter to your business, checked on a recurring basis rather than assumed from a rank report. It's also worth watching organic click volume and conversion as their own line, independent of ranking position, since a widening gap between "still ranked well" and "getting fewer actual visits" is exactly the pattern this whole mechanism produces. Treating that gap as a mystery, rather than as the AI Overview effect showing up in your own numbers, is how teams end up making changes to a page that was never actually the problem. In practice this means running two separate checks on a recurring basis rather than one combined report: a rank check against target keywords, the way it's always been done, and a citation check against a defined list of the actual questions buyers ask, tracking whether your content is named as a source in the synthesized answer. The two numbers can diverge in either direction. A page can lose rank while gaining citations, or hold rank while losing citations entirely, and only tracking one of the two means missing exactly the shift this whole pattern is about. ## How to Get Cited Instead of Just Ranked Getting selected as a citation source starts with the same foundational work traditional ranking has always rewarded, clean crawlability, valid schema, a technically healthy site, but adds a second requirement on top: content structured to answer one specific question directly and completely, rather than ranking broadly for a topic while making a reader dig for the actual answer. A page that opens with a direct answer to a narrow question is a much better citation candidate than a page that opens with company background before eventually getting to the point. This also means the unit of work shifts. Traditional SEO often optimizes one page to rank for one broad keyword cluster. Getting cited across the questions that actually matter to a buying decision means treating each distinct question, "how does X compare to Y," "what does X actually cost," "how long does X take," as its own answerable unit, even when several of them could plausibly live on the same page. A page trying to be the answer to five different questions at once is a weaker citation candidate for any single one of them than five pages each built to answer one question completely. The harder part isn't writing one page this way, it's doing it consistently across the full set of questions buyers actually ask, and then rechecking which sources get cited over time, since query fan-out means the winning source for a given question can shift as the retrieval pool changes. A single optimized page is a starting point. Staying cited as that pool shifts is the ongoing work, closer to what AEO actually involves (https://www.focentra.ai/learn/what-is-answer-engine-optimization.html) than a one-time content fix, and it follows a similar early-signal-then-compounding pattern to what shows up in the realistic timeline for AEO results (https://www.focentra.ai/learn/aeo-results-timeline.html): foundational fixes first, then a widening gap between the sites still treating this as a ranking problem and the ones treating it as a citation problem. --- Stop chasing rankings. Start earning the citation. Book a 20-minute revenue audit at https://cal.com/danydalal/revenue-strategy-call, we'll show you which of your buyers' actual questions you're currently cited for, and which ones you're invisible on.