68% of Google Searches Ended Without a Click. Here's Where B2B Visibility Went

68% of U.S. Google searches ended without a click from January to April 2026. A 380,000-answer study maps how AI surfaces cite Reddit, YouTube, G2 and vendor pages differently — and what teams should build first.
By Marcus Bennett
3 min readAt a glance
- 68% of U.S. Google searches ended without a click from January to April 2026
- Reddit appeared in 52% of AI Mode's citing answers but 0% of Claude's; ChatGPT's Reddit citation rate fell from 21% in July to 10% in August
- 84 of the 100 most-cited URLs were comparison pages or listicles, drawing over 66,000 citations
- 51% of B2B software buyers now start research in an AI chatbot more often than Google, up from 29% a year earlier
- Gemini sends 72% of citations to vendor and product domains; review citations range from 8% to 25% by category
From January to April 2026, 68% of U.S. Google searches ended without a click, because Google's AI features now answer questions directly on the results page. That's the headline number from Foundation's State of B2B AI Discovery 2026 report, built with AirOps from roughly 380,000 AI answers across 81 enterprise software topics and six AI surfaces.
The study splits visibility into what it calls two internets: Google's AI surfaces and standalone models. Google's AI Overviews (2.5 billion monthly users) and AI Mode (another billion) ground answers in live Search results. ChatGPT, Claude, Gemini and Perplexity run their own retrieval, and the citation patterns diverge sharply along that line.
What does each surface actually cite?
- Vendor sites, integrators and tool directories take the largest share everywhere: 52% on Perplexity, 72% on Gemini. That share includes competitors and directories, not your pages by default.
- Reddit appears in 52% of AI Mode's citing answers and 39% of AI Overviews, but only 21% on ChatGPT and zero on Claude.
- YouTube reaches 47% of AI Mode answers and 38% of AI Overviews — yet 0.3% of Gemini's citations, roughly 17x less, despite both being Google products. The gap comes from query fan-out (Google running multiple related searches per answer), which pulls in videos that already rank.
- G2 lands between 23% and 30% on five of six surfaces. It's the one domain every surface agrees on. For Claude, third-party citations are almost entirely review platforms: G2 in 30% of citing answers, Capterra 13%, Gartner and PeerSpot about 9% each.
ChatGPT's Reddit citation rate fell from 21% in July to 10% in August, the study found, while Google's surfaces kept citing Reddit. Monthly reporting would have missed the drop — the report recommends weekly citation tracking.
What does this change for marketing teams?
Buyer behavior is moving first. Citing G2 research, the report notes 51% of B2B software buyers now start research in an AI chatbot more often than Google, up from 29% a year earlier. Still, Google handles roughly three-quarters of desktop searches, and ranking matters — it just no longer guarantees a citation, as AI Overviews pull fewer links from top-ranking pages.
Comparison content is disproportionately powerful: 84 of the 100 most-cited URLs were comparison pages (59) or listicles (25), together drawing over 66,000 citations. "Editorial gets the budget; product pages get the citations," said James Scherer, Foundation's VP of Strategy. His advice: fix the pages AI already cites before writing another blog post.
Reddit earned 159,000 citations across 12,500 unique URLs, the most-cited external domain in the dataset. Its logged-out users, who mostly arrive from search engines, grew 27% year over year in Q2 2026 and now outnumber logged-in users, 77.7 million to 52.6 million. The report warns against treating Reddit as a launch channel: steady, useful answers across 50 threads beat one viral thread.
Review sites are the standalone engines' backbone — Claude and Perplexity draw about one in five citations from review domains. In G2's 2026 research, review sites (38%) narrowly passed AI chatbots (37%) as the top source shaping B2B shortlists.
Where to start
Category mix swings are large: review citations range from 8% in managed databases to 25% in ATS and recruiting software; community citations from 5% for SEO platforms to 14% for EDR/XDR tools. The report advises asking buyers which AI tools they use, pulling your category's source mix before budgeting, and measuring by surface with a fixed set of 100+ prompts — alerting on any 20% week-over-week swing. AirOps' 100-point sentiment scores ranged from 81 on Perplexity to 69 on Claude on identical prompts.
Team-size guidance scales from solo operators (own pages plus one layer) to teams of eight or more (one owner per layer plus a per-surface citation reporter). Expect the first quarter to be slow; the report frames a 12-month horizon before the stack compounds and citation shifts get caught within the week they happen.
Source: searchenginejournal.com
Marcus Bennett
Show full bio
Market editor covering money, funding and market shifts around Social platform news and policy at Reach Brief.
Related articles
AI Assistants Hit 76 Billion Monthly Sessions as Report Drops Search Comparison
AI Search Now Decides Which Creators Get Discovered
Alphabet's 13% Stock Plunge and the Ad Tech Bot Identity Scandal
ChatGPT holds 81.4% of global chatbot traffic as growth returns
Alphabet’s $127.6B Ad Quarter, With AI Taking Center Stage