Week of August 25, 2026

A new analysis from Search Engine Land found that Reddit's share of ChatGPT Search citations fell 86% in just four days, dropping from an average of 3.83% to roughly 0.52% between August 14 and 17. An earlier, more gradual decline had already begun August 8, when Reddit's share slipped from the high-3% range into the mid-2% range.

The drop looks specific to ChatGPT. Google's AI Overviews showed no comparable one-day collapse, with Reddit's citation share there declining gradually instead, from about 2.5% in early July to roughly 2.1% in August. The analysis, based on tracking millions of AI responses, could pin down when the shift happened but not why, and the researchers caution the findings remain provisional while they rule out possible data-collection issues.

For brands and publishers watching platform-specific citation share, it's a reminder that answer engines can reweight entire source categories overnight, with no warning and no public explanation from the platform.

Read the full article: https://searchengineland.com/reddit-chatgpt-search-citations-fall-report-485473

Search Engine Land published a nine-month analysis tracking more than 51,000 AI Overview events and 1,661 cited snippets, one of the more detailed looks yet at how Google's AI Overviews perform once a page actually gets cited.

The data shows AI Overview-driven sessions peaked at 16 to 17% of organic traffic in February and March 2026 before settling to a more typical 2 to 4% recently. More strikingly, the analysis found that AI Overview clicks are being misattributed as "Direct" traffic in Google Analytics an average of 22.4% of the time, rising as high as 29.3% in a single month. Structured, specific content — times, prices, routes, comparisons, and HTML tables in particular — outperformed other formats, and cited snippets showed identifiable lifecycles tied to freshness.

The takeaway for practitioners: standard analytics setups are likely undercounting AI Overview-driven traffic, meaning AEO's real referral impact may be larger, and harder to prove, than dashboards currently show.

Read the full article: https://searchengineland.com/ai-overview-data-51000-tracked-events-485080

A new Search Engine Land survey finds that 81% of marketers still refer to their AI search work internally as "SEO," even though 82% have already committed budget to AI search visibility, averaging 24% of total search and content spend.

Terminology hasn't caught up with the shift: only 27% of marketing teams have formally adopted a new term, 42% have decided against one, and the rest remain undecided. Where new terms do show up, usage skews senior — C-suite and directors use "GEO" and "AEO" roughly three times as often as individual contributors. The survey also found ChatGPT is the clear priority platform at 34%, ahead of Gemini (16%), Claude (6%), and Copilot/Bing AI (5%), with only 1% of marketers prioritizing Perplexity. When evaluating vendors, marketers ranked measurable case studies as the top credibility signal and "excessive buzzword usage without clear explanation" as the top red flag.

For AEO/GEO vendors and consultants, the message is blunt: don't let the acronym carry the pitch.

Read the full article: https://searchengineland.com/ai-search-marketers-call-it-seo-485574

Search Engine Land laid out a three-phase framework for earning citations in AI search, built on the finding that up to 85% of top-of-funnel B2B brand mentions in AI answers come from third-party sources rather than a brand's own site.

Phase one is making pages accessible to language models — clean navigation, machine-readable structure, and proper technical tagging across product pages, blog content, and social profiles. Phase two is producing fresh, gap-filling content, mixing original research and seasonal editorial with refreshed formats like infographics, short-form video, and hub-and-spoke clusters. Phase three shifts to earned media: building citations through journalist relationships, niche publication placements, podcast appearances, and micro-influencer partnerships.

The framework's core argument is that the phases reinforce each other and skipping one undermines the rest: "AI visibility compounds when technical accessibility, useful content, and third-party authority work together."

Read the full article: https://searchengineland.com/phases-build-authority-earn-ai-citations-485470

Google's John Mueller said on Bluesky that "from our POV there's nothing really special you need to do for generative AI responses in search," as reported by Search Engine Roundtable. Mueller noted he wasn't entirely sure he understood the original question, but the comment lines up with Google's consistent position that standard SEO fundamentals cover AI-generated search results without a separate optimization playbook.

The statement puts Google's public stance at odds with Microsoft's Bing team, which has previously described more deliberate steps for AI-driven search surfaces.

For practitioners, it's a useful data point on where the platforms disagree: Google keeps saying good SEO is good AEO, while positioning elsewhere in the industry keeps treating them as distinct disciplines.

Read the full article: https://www.seroundtable.com/google-generative-ai-search-responses-todo-41916.html

Interested in advertising with AEO River?
Reach out to explore sponsorship options, formats, and availability.

Talk to our team to learn more.

Please note that posts in this newsletter marked “sponsored” contain sponsored content.