AI search is no longer a gateway—it is becoming the destination
Google AI search citations describe how often Google’s AI Mode and AI Overviews attribute their answers to specific domains, and the growing share of those citations pointing back to Google’s own properties shows that AI search is shifting from a referral system into a self-contained experience where Google pages—not publisher websites—are the default destination for many queries.
That shift is happening while Google’s search leaders argue that AI search still helps the open web. Nick Fox says AI features in Search now send “billions of clicks to websites every week,” and that overall Google Search continues to send billions of clicks daily. Liz Reid echoes the idea that AI “expands curiosity rather than limits it.” But traffic data tells a harsher story: a recent zero-click study found Google now sends only 27.6% of its clicks to the open web, with click‑through rates falling. If AI overviews traffic impact is supposed to reassure publishers, those numbers suggest the opposite: AI is doing more of the work, and the web is getting less of the reward.
The Google self-citation problem: Business Profiles and Product Panels
Behind the rhetoric about supporting the web sits a blunt fact: Google is citing itself more than ever. A new AI-visibility report shows google.com’s citation share in AI Mode rose 8.4x between April 15 and June 30, making it the model’s second most-cited domain after YouTube. This surge is driven almost entirely by two Google-hosted surfaces: Business Profile cards and Product Knowledge Panels.
These cards now dominate local and product-intent queries. For queries about restaurants, home services, real estate, healthcare, hospitality and travel, AI Mode often cites a Business Profile card instead of a brand’s website. For product comparisons and compatibility checks, Product Knowledge Panels display specifications and merchant options directly inside Google’s interface. One data set found that earlier AI Overview responses already referenced Google’s own results 43% of the time, while another found Google properties made up 17% of 1.3 million citations, rising to about 20% when including YouTube. In other words, much of what AI Mode links to for local and product queries is Google’s own properties, not independent websites.
AI overviews traffic impact: clicks, zero-clicks, and missing data
Google insists AI Mode is designed to connect people to websites and claims it has been improving how links show up inside AI features so users can go deeper when they want. Yet publishers report the opposite: they see less traffic from Google Search than before AI features launched. When a study shows that only 27.6% of Google clicks now reach the open web, the “great decoupling” feels less like a theory and more like lived reality.
The more AI answers inside the results page, the more queries end in zero-click sessions. That pattern is compounded by Google’s decision to ship AI performance reports in Search Console without including clicks. Officially, the company says it is still working with website owners to decide which insights matter most—and apparently, those insights are not clicks. Leaving clicks out looks less like a gap and more like a strategy: if AI overviews traffic impact were truly neutral or positive for publishers, Google would have every reason to show the numbers. The silence is itself a signal.
SEO strategy for AI search: your “homepage” moved to Google
For brands, the harsh news is that the old playbook—rank a blue link, optimize the landing page, win the click—is no longer enough. As one AI-visibility researcher puts it, “Your homepage is no longer the first touchpoint. The Google-hosted card is.” In AI Mode, Business Profiles and Product Knowledge Panels often become the effective landing page for your brand, even though they sit on google.com.
That means SEO strategy for AI search starts with data feeds and profiles, not only content hubs. Profile completeness, fresh photos, correct hours, and rich fields move from hygiene to growth levers. Reviews are no longer tucked away; they sit front and center in the card, shaping user perception before anyone reaches your site. The AI system is trained on the data you provide, and the resulting card becomes the surface users interact with first. The bottom line: winning the click now requires winning the card first. These cards are at least built from data you control—your Business Profile for local and your product feed for shopping—but control is not the same as ownership.
What publishers should do next in an AI-first search world
Publishers now face a dual challenge: stay visible in traditional organic results while adapting to an AI layer that rewrites how information appears and which entities are credited. AI search visibility requires different tactics than classic SEO—schema, feeds, profiles, and review management matter as much as keywords and backlinks. Ignoring AI Mode because “it sends some clicks” is wishful thinking; citation share shows where AI attention goes, and increasingly, that is toward Google’s own surfaces.
The uncomfortable truth is that Google’s incentives and publishers’ interests are no longer neatly aligned. Google is making AI Mode the default search experience while its own domain’s citation share soars. Meanwhile, third‑party studies highlight falling CTRs and a shrinking share of clicks reaching the open web, even as Google publicly stresses stability. In this environment, publishers need a two-track strategy: treat Google’s AI surfaces as critical distribution channels to be optimized, and at the same time, invest in direct relationships—email, apps, communities—that do not depend on a search engine whose AI is learning to cite itself.
