The world of ChatGPT has tilted decisively from asking to doing. OpenAI released its first country-by-country Signals dataset, showing how over 1 billion people put ChatGPT to work. This is essential reading for GEO practitioners tracking adoption shifts.<\/p>\n
The headline: people are more than twice as likely to use ChatGPT to complete tasks at work as they are outside work. Multimedia is the fastest-growing use case, hitting 7.8% of messages globally. Interestingly, in Brazil and Colombia, multimedia exceeds one in ten messages. The adoption gap is closing too. Peru, Uruguay, and Costa Rica jumped the most in per-capita rankings. Usage among people over 35 rose in nearly every country. France and Czechia each saw over 10 percentage-point gains.<\/p>\n
I recommend downloading the Signals data for audience modeling. The doing-versus-asking split gives clear direction for content strategy. Build assets that help users complete tasks, not just find answers. The shift is global, and it is accelerating.<\/p>\n
\ud83d\udd17 OpenAI Newsroom<\/a><\/p>\n If you care about improving GPT-5.6 output for AI search, OpenAI’s latest update matters. The company shipped GPT-5.6 Sol to Plus and Pro users. It also expanded GPT-5.6 Luna access to Free users with unlimited text chats. I think GEO practitioners should treat this as a signal.<\/p>\n Sol delivers more focused answers and stronger factual reliability. OpenAI tested it on financial, medical, and legal prompts. A reasoning slider lets users adjust effort per response. Luna now includes a Think button for harder questions. OpenAI reports 1 billion weekly ChatGPT users.<\/p>\n I recommend studying the before\/after examples in the original post. They show how model tuning compresses answers and prioritizes the core conclusion. That changes how AI search appears in SERPs and how content should be structured. Brands that provide direct answers will gain visibility.<\/p>\n \ud83d\udd17 OpenAI Newsroom<\/a><\/p>\n OpenAI’s update on advancing responsible AI in Europe gives GEO practitioners a clear view of AI content provenance regulation. I recommend reading it even though it is not a direct SEO guide.<\/p>\n The article explains OpenAI’s alignment with the EU AI Act. OpenAI endorsed two Codes of Practice: the GPAI Code and the Transparency of AI-Generated Content Code. Both signal where content labeling standards are heading.<\/p>\n I find the provenance section most relevant for GEO. OpenAI pairs Content Credentials (C2PA) with SynthID watermarks. They now expand this to audio outputs, with text next. This matters for anyone tracking how AI content gets identified in search results.<\/p>\n The piece also covers OpenAI’s Preparedness Framework and Frontier Governance Framework. These map safety practices to legal requirements.<\/p>\n I recommend monitoring these transparency standards. They will shape how search engines and AI assistants treat labeled content.<\/p>\n \ud83d\udd17 OpenAI Newsroom<\/a><\/p>\n OpenAI’s “Building abundant intelligence” post gives SEO and GEO teams a rare view of the economics shaping search’s future. I recommend reading it as strategic context, but do not expect tactical optimization advice.<\/p>\n The data matters. OpenAI cut GPT-5.6 Luna prices by 80 percent. GPT-5.6 Sol improved speculative decoding, lifting token-generation efficiency by over 15 percent. ARC-AGI-3 scores jumped from 13.3 to 38.3 percent while using one-sixth as many output tokens. The model stayed the same. The system around it changed.<\/p>\n I think the GEO takeaway is direct: falling intelligence costs expand what becomes practical to automate. Agentic work through Codex now drives 99.8 percent of OpenAI’s weekly output tokens. That shifts content consumption patterns and SERP behavior.<\/p>\n The article shows the strategy behind AI abundance, but it offers no actionable playbook. Read it to understand the market direction, then build your own response.<\/p>\n \ud83d\udd17 OpenAI Newsroom<\/a><\/p>\n OpenAI’s newest education plugins reveal new ways to learn and teach with ChatGPT Work and Codex. They give SEO practitioners a concrete preview of agentic AI workflows entering the mainstream.<\/p>\n This article introduces three packaged plugins for K-12 educators, college educators, and college students. Each bundles role-specific skills, instructions, and app connections so users skip complex prompt engineering. I think the scale matters most. OpenAI partners with the American Federation of Teachers to train 400,000 K-12 educators. That’s roughly one in ten US teachers. The K-12 plugin integrates with Learning Commons to align materials with academic standards. The College Educator plugin automates syllabi, course websites, and LMS packaging. The College Student plugin creates study guides, flashcards, and quizzes from sources students choose.<\/p>\n For GEO professionals, these plug-and-play agents show where AI-assisted content production is heading: context-rich, role-specific, and institution-controlled. I recommend monitoring these plugins as early signals for enterprise AI adoption and future search behavior.<\/p>\n \ud83d\udd17 OpenAI Newsroom<\/a><\/p>\n This article earns my recommendation because it shows how AI referrals are recreating the oldest conversion optimization mistake \u2014 sending specific intent to generic pages. The author backs the claim with three independent datasets that point to the same split: the machine reads deep, the traffic lands shallow.<\/p>\n Similarweb’s 2026 report found 65% of ChatGPT-cited URLs sit two to three folders deep, yet 58.8% of referral traffic lands on homepages. Previsible’s analysis of 6.77 million AI-referred sessions across 166 websites adds the sleeper stat: 28.8% of ChatGPT referrals land on internal search results pages. Ahrefs reported the same pattern on its own site, with over 80% of AI traffic hitting homepages, product pages, and free tools instead of editorial content.<\/p>\n I think the internal search finding deserves the most attention. That page is usually the platform default, untouched by marketers, because Google handled intent-matching before. Now it receives a third of AI referrals. The visitor arrives pre-convinced and gets asked to search again. That is Let Me Google That For You without the joke.<\/p>\n I recommend treating AI referrals as a landing page problem first. Map the deep pages ChatGPT cites, then route the click to that page or build a midpoint page that matches the promised intent. Similarweb found ChatGPT-recommended brands get 2.5 times more visits in seven days. Those visitors are ready to convert. Give them the page they were promised.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n This article delivers the most useful data I’ve seen on top stories inside AI Overviews. It should force publishers to rethink their opt-out strategies. John Shehata of NewzDash tracked Google’s July 2026 rollout. The numbers are striking.<\/p>\n Top stories inside AI Overviews appear on 15.5% of U.S. trending news queries. The UK sits higher at 17.46%. Entertainment leads both markets. U.S. entertainment queries exceed 35%. World News reaches nearly 32%. Google shows the embedded carousel instead of a standalone module. The two placements are mutually exclusive.<\/p>\n I recommend reading this because it dismantles a harmful myth. Blocking Google-Extended still leaves publishers inside AI Overviews. Google’s documentation limits Google-Extended to AI training and grounding; blocking it leaves Search inclusion untouched. The real control is the generative AI exclusion in Search Console. That setting removes content from AI Overviews, AI Mode, and generative Discover. It likely kills embedded top-story spots too. Google still withholds layout confirmation, and Shehata honestly labels that as an interpretation.<\/p>\n Interestingly, this mirrors 2006’s vertical creep. Google pulled news into web results before Universal Search arrived. I think the bigger story is trust. Embedding named publishers inside AI answers rebuilds credibility. Users see real bylines and direct links. That beats synthetic text with thin sourcing. This is a genuine step forward. I recommend pairing this article with your Search Console settings review. The data changes news SEO budgets immediately.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n I helped scale Google Ads through billions in advertiser spend, and this article shows how those lessons map to AI search. Jason Shafton argues that AI assistants changed what a result is. Ranking stops mattering when the answer becomes the result.<\/p>\n The Semrush stat frames the shift. Nearly 90% of ChatGPT-cited pages sit at position 21 or lower. Twenty years of chasing page one barely registers with assistants. Client audits confirm the pattern. One brand won citations through Reddit and trade press despite middling rankings. Another held page-one positions and earned no AI citations.<\/p>\n The article breaks AI visibility into six signals: brand authority, content freshness, entity recognition, extraction-ready structure, cross-platform consensus, and earned media. Most of these signals live outside your website. Brand and PR work now drives performance marketing.<\/p>\n The 90-day sprint is the practical core. Days 1-30 run 20 query tests across ChatGPT, Perplexity, Gemini, and Google AI Mode. If you show up in fewer than 12 runs, you have a citation problem. Days 31-60 run one scoped experiment per channel. Tools like Profound, Otterly.AI, and Peec AI turn progress into a trendline.<\/p>\n I recommend this article for anyone still optimizing for the old entry points. The evidence is concrete. The framework is actionable.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n This Semrush study shows how topical authority spreads in ChatGPT and where it stops. It gives SEO and GEO teams a data-backed roadmap for AI visibility. I consider this a must-read because it settles a question most of us have been guessing at: does winning one category earn you the next one?<\/p>\n The research tracks two separate signals across 283,215 citations and 76,493 brand mentions in 1,094 ChatGPT categories. Citations spread widely, even into distant categories. Brand mentions stay close to core expertise. In close categories, brands earned citations in 74% of appearances and named mentions in 44%. In distant categories, those shares dropped to 50% and 25%. Cited-and-named appearances were nearly 4x higher in close categories.<\/p>\n Depth beats breadth for brand recognition. Brands appearing in only one of five prompts saw their mention share drop. That penalty lifted only after appearing in three of five prompts. Showing up in just one of 20 categories built nothing. Industry shapes the pattern too. Finance and real estate rewarded broad coverage with more citations. Legal and healthcare demanded more proof before ChatGPT named brands.<\/p>\n I recommend using this data to focus your GEO strategy. Build depth in a few core categories first. Then expand into semantically adjacent ones. Topical authority works best as a launchpad into nearby territory, with far less carryover into unrelated worlds.<\/p>\n \ud83d\udd17 Semrush Blog<\/a><\/p>\n Brodie Clark delivers the most practical guide on utilizing Bing product feeds for ecommerce AI SEO. It includes step-by-step setup instructions and a compelling case study. I rate this 9\/10 for actionable depth and current UCP insights.<\/p>\n Clark walks through the full Microsoft Merchant Center workflow. You need a Microsoft Advertising account first. Merchant Center lives inside the advertising platform. Importing feeds from Google Merchant Center is straightforward. Clark warns that verifying the correct business entity is critical. He shares a client example where the wrong entity caused weeks of delay, setting their progress back considerably.<\/p>\n The case study on ads and free product listings stands out. Clark demonstrates that ads must be enabled for free listings to appear in the Bing Shopping tab. That finding changes how I advise ecommerce clients. The experiment lends confidence to the conclusion.<\/p>\n I appreciate his transparency about platform bugs. Free product listing reporting has been stuck since July 6th. Outdated documentation remains an issue. He also flags Bing’s new UCP setting within Microsoft Merchant Center. That development deserves close attention from ecommerce teams.<\/p>\n I recommend this for any ecommerce SEO practitioner. Clark grounds his advice in real experiments rather than theory. The insights on ChatGPT using product feeds make this timely for AI-focused strategies. Product validation in Merchant Center alone rarely secures free listing visibility. Clark’s case study shows why.<\/p>\n \ud83d\udd17 Brodie Clark<\/a><\/p>\n Moz’s Dr. Peter J. Meyers shares a free 50k query fan-out dataset that reveals exactly how brands bias AI search. This is the most actionable GEO research I have read this year. His team generated 1,000 subtopics across 20 verticals, then fanned each into 50 prompts using the Gemini API. The result: 50,000 fan-out prompts plus 5,333 grounding queries.<\/p>\n Three insights matter most. Brand bias concentrates in just two fan-out types. Only 12.8% of prompts mentioned a brand, yet Entity and Comparison fan-outs generated 97% of those mentions. Brands missing from comparison queries become invisible in AI answers. Relevance and diversity need balance. Alignment scores ranged from 0.22 to 1.00 with a mean of 0.67. Even a low 0.22 score remains topically useful, while 0.90 matches nearly every keyword. That sweet spot keeps fan-outs connected while preserving variety. The medium long-tail dominates. Prompts averaged 8.2 words under a 12-word cap. Content teams should target those mid-length queries instead of chasing ultra-long variations.<\/p>\n I recommend downloading the raw dataset now and auditing your brand through Entity and Comparison fan-outs first \u2014 that is where AI visibility is won and lost. The data covers 20 industry verticals, so most brands will find usable benchmarks. This original, repeatable research deserves a place in your GEO toolkit.<\/p>\n \ud83d\udd17 Moz Blog<\/a><\/p>\n I recommend this article because it delivers a practical framework for AI visibility measurement. Traditional SEO metrics are breaking down as AI answers reshape search. David Khim cuts through social media noise and gives operators a clear playbook.<\/p>\n Three ideas matter most. First, prompts are the opening domino. Most AI visibility tools auto-recommend prompts by mapping existing pages back to LLM outputs. That assumption fails when your current pages are built around the wrong strategy. Track the prompts your customers actually ask, not the ones your site already covers. Second, track visibility per model. ChatGPT skews toward consumer; Claude skews toward enterprise. B2B teams need Claude visibility despite its lower user counts. Google AI Mode also matters because Chrome dominates the browser market. Third, use self-reported attribution. Add “How did you hear about us?” to lead forms and CRM records. One client discovered that 5% of registrations came from ChatGPT despite having done zero AI visibility work.<\/p>\n Interestingly, the data explains why standard analytics can be misleading. 80-90% of AI-platform leads were tagged as organic or direct. That makes referral traffic nearly useless as a KPI. Citations belong on the monitor list, not the KPI dashboard.<\/p>\n I recommend using prompts, cross-model visibility, and self-reported attribution as your core KPIs. Then tie self-reported data to pipeline and closed revenue. That connects AI visibility measurement to business outcomes. Stop chasing citations. Start asking your buyers.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n Google now reports AI search impressions separately in Search Console. That changes how GEO practitioners measure generative visibility. I recommend reading this breakdown before building your next AI search report.<\/p>\n The new Generative AI report shows impressions from AI Overviews and AI Mode. You can segment by page, country, device, and date. It lacks CTR, average position, citation placement, and passage data. Google calls this a diagnostic lens, not a scoreboard. I agree.<\/p>\n Key insight: AI impressions differ from traditional impressions. One property-level impression can represent multiple URLs. Page-level rows each count separately. The math looks broken, but it isn’t. Avoid blending these numbers with organic data.<\/p>\n I find the comparison framework valuable. Map organic visibility against AI visibility per page. You will spot pages with broad visibility, pages ranking without AI retrieval, and retrieval outliers. Google is also testing an opt-out control for AI features. Establish a baseline before you use it.<\/p>\n This article earns its place in your GEO reading list.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n AI search data finally has a home in Google Search Console. Neil Patel’s latest article walks through the new Generative AI report. I think this is the most actionable update for SEO and GEO professionals this year.<\/p>\n The report shows how your content performs in AI Overviews and chatbot-style search results. You can see impressions, clicks, and the exact queries that trigger AI citations. This closes the biggest blind spot in GEO measurement. Before this update, we had no reliable way to track AI visibility.<\/p>\n Patel explains how to interpret the new metrics. He also compares AI search performance against classic SERP data. I recommend setting up this report immediately. Establish your baseline this week. Track which content formats earn AI mentions. Use the query data to guide entity optimization and source credibility signals.<\/p>\n This is measurement, not guesswork. Review the report monthly. Adjust your GEO strategy based on real behavior. The tools have finally caught up with this shift.<\/p>\n \ud83d\udd17 Neil Patel<\/a><\/p>\n This is the practical playbook for SEOs who want to build full SEO analysis dashboards without touching code.<\/p>\n The Moz API MCP server connects your AI assistant directly to Moz’s SEO data using open protocols. I recommend this guide for anyone tired of spending months building custom reporting apps. The author tested it with a single prompt comparing bose.com to jbl.com and sennheiser.com. That one sentence produced a full competitive analysis, including keyword metrics, site metrics, and Brand Authority. He then used the Fetch Link Intersect endpoint to surface backlink opportunities, getting exact domains and content types to target. Any MCP-compatible tool works: Claude Desktop, ChatGPT, Gemini CLI, Cursor, or VS Code. My favorite tip: end every session with “Build me a dashboard from all of this.” That turns scattered files into one shareable deliverable. Interestingly, the MCP approach also cuts token waste by focusing the model on specific data rather than broad analysis. I think this is a smart way to get senior-level reporting speed from any team.<\/p>\n \ud83d\udd17 Moz Blog<\/a><\/p>\n I recommend this podcast episode for anyone tracking the state of GEO conversation in AI search. Josh Blyskal delivers data-rich insights that cut through industry noise.<\/p>\n Prompt frequency obsession wastes time. Running a prompt 10x daily vs once barely moves visibility scores. The real hard problem remains attribution.<\/p>\n Advanced teams now use a new model. They treat agent attention as a proxy for human attention. They heavily discount citations into “clicks.” ChatGPT referral traffic is up 60% since mid-March. CPG brands benefit most.<\/p>\n GEO is fundamentally a relationship business. Look at the Reddit-Google negotiations. Claude rarely cites Reddit at all. The walled-garden problem is real.<\/p>\n I appreciate the honest breakdown of slop vs signal. Product pages and battle cards get fresh analysis too. Interestingly, listicles still do the “caloric work” of teaching models an industry.<\/p>\n The ChatGPT vs Claude comparison adds practical depth. If you build AI search optimization strategies, this episode offers attribution thinking you can apply today. I recommend giving it a full listen.<\/p>\n \ud83d\udd17 Siege Media<\/a><\/p>\n I tested 15 brand monitoring tools to find the best options for tracking AI, social, and media mentions. The Moz article ranks 8 picks. I think this list deserves attention from SEO and GEO practitioners.<\/p>\n The article evaluates each tool on channel coverage, context accuracy, query flexibility, trends, reporting, and setup effort. Moz AI Visibility leads for tracking brand mentions in LLMs like ChatGPT, Gemini, and Google AI Mode. It includes prompt suggestions, prompt tracking, citations, and sentiment analysis starting at $99\/month. Brandwatch dominates enterprise social listening. Meltwater suits PR teams. Brand24 adds AI-enhanced listening. Google Alerts remains a solid free option.<\/p>\n I recommend reading this before buying any monitoring tool. The comparison table saves hours of research. The author tested each product with the same methodology, so the rankings feel fair. For GEO work, the Moz AI Visibility section delivers the most value. It shows how to build a prompt bank and find citations that shape AI answers. This is a practical reference for building your 2026 monitoring stack.<\/p>\n \ud83d\udd17 Moz Blog<\/a><\/p>\n The future discoverability of publisher content depends on winning the passive phase of discovery, not optimizing solely for rankings.<\/p>\n Harry Clarkson-Bennett, SEO Director at The Telegraph, argues that discoverability now spans social platforms, AI assistants, Google Discover, podcasts, newsletters, and creators. The old search-led model no longer dominates every stage of the funnel. I think his framework of three discovery phases \u2014 passive, active, and owned \u2014 gives SEO teams a practical lens for rebuilding growth strategy. Getting users into the owned phase requires consistency across the other two.<\/p>\n The piece delivers hard data. Up to 86% of Google clicks are discovery-led, per iPullRank’s Q3 2026 zero-click behavior analysis. Teenagers spend just four minutes daily on publisher websites versus 19 minutes for over-55s. And 52.3% of searches from 18-to-24-year-olds end without a click, compared to 42.7% for those aged 55-64. Interestingly, older audiences are seven times more likely to click an ad, while Gen Z remains more discerning. These numbers confirm the shift is structural, not a temporary dip.<\/p>\n I recommend this article for any SEO lead managing media or publisher properties. Clarkson-Bennett explains why platforms evolved into walled gardens, how answer engines commoditized informational content, and why publishers must build habits through consistent engagement. His “break the scroll” concept is the actionable takeaway \u2014 create moments compelling enough to interrupt passive consumption. This is the clearest thinking I’ve seen on future discoverability this year.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n When AI takes the click, click worthiness should guide investment strategy.<\/p>\n That is Bill Hunt’s central argument in this Search Engine Journal piece. I believe it is the most actionable planning framework for zero-click search. Hunt’s Click Worthiness model measures the remaining business value after AI answers the query. Search volume still reveals customer intent, yet AI shifts where the value lands. He contrasts two United Airlines queries. The fact-based question ends at the AI answer. The decision-based question starts a customer journey. The second query justifies continued investment. The first delivers little incremental value. Interestingly, commercial proximity alone rarely creates engagement value. I recommend applying this test to every content brief. Ask whether the query leaves a decision worth making. Completeness is becoming the cost of participation. Competitive advantage comes after AI answers. Organizations that create meaningful engagement will outperform those publishing interchangeable content. Practical, sharp, and directly applicable to budget planning.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n A report warns that Cloudflare AI bot blocking can stop Googlebot from indexing your site. I think every SEO should check their settings today. A Reddit user in r\/SEO enabled Cloudflare’s AI Training block. Googlebot and Bingbot then received HTTP 403 errors on their sitemap. Disabling the block restored access immediately. That is a timely warning.<\/p>\n Cloudflare classifies Googlebot as a “Search + Training” bot. Blocking AI training also blocks search indexing. Cloudflare’s documentation confirms mixed-purpose crawlers get blocked by all AI-blocking configs starting September 15, 2026. The Reddit user saw this before the official deadline. Whether this is a fluke or user error remains unclear. I think the risk is real enough for immediate action.<\/p>\n The poster confirmed Cloudflare’s dashboard showed Googlebot and Bingbot as blocked in the AI Crawlers section. They weren’t dealing with spoofed user agents. Google’s John Mueller asked for a DM. That tells me the issue has traction at Google’s level.<\/p>\n I recommend three steps. First, audit your Cloudflare Bot Fight Mode and AI Crawler settings. Second, verify Googlebot can access your sitemap in the URL Inspection tool. Third, watch Cloudflare’s September 15 default change if you run ads. The new defaults block Training and Agent bots on ad-supported pages. Search stays allowed. Mixed-purpose bots like Googlebot get blocked under any AI-blocking config.<\/p>\n This report serves as a useful reminder. AI bot blocking has a hidden cost. Test your settings now. Otherwise Google notices your indexing problems first.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n I recommend listening to The Page 2 Podcast episode featuring Yoast Principal SEO Alex Moss. It delivers one of the clearest roadmaps I have seen for getting brands selected by AI agents.<\/p>\n The conversation tackles four practical areas. Markdown outperforms HTML for LLM comprehension. llms.txt offers a portable way to create machine-readable sources of truth. Entity maps and structured data help AI connect content to the right context. Businesses should prioritize these signals over traditional click metrics.<\/p>\n Alex Moss joins hosts Jon Clark and Joe DeVita to explain the shift from rankings to AI selection. I think the entity relationship discussion is the strongest segment. It shows exactly how to build knowledge graph foundations that AI systems trust.<\/p>\n For SEOs experimenting with AI search, this episode provides actionable next steps. I recommend treating it as a practical checklist for the age of AI agents.<\/p>\n \ud83d\udd17 Yoast SEO Blog<\/a><\/p>\n The Semrush Blog’s roundup of the 7 best AI visibility tracking tools for agencies gives SEO practitioners a practical benchmark for comparing AI tracking platforms. I recommend it for agency-side teams exploring tooling, even though the content carries obvious promotional bias toward Semrush’s own toolkit.<\/p>\n The author personally tested most tools. The comparison spans Semrush, OtterlyAI, Peec AI, AthenaHQ, Profound, Scrunch, and Brandlight. Each entry includes pricing, key strengths, and the agency need it fits best. OtterlyAI starts at $29\/month for budget-conscious teams. Peec AI suits fluctuating client rosters. Semrush’s AI Visibility Toolkit costs $99\/month per domain.<\/p>\n I find the selection criteria most useful. The article evaluates tools on cost at scale, multi-client monitoring, actionable recommendations, AI platform coverage, and white-label reporting. The at-a-glance table makes agency-specific strengths easy to scan.<\/p>\n My main criticism: depth is moderate, and Semrush ranks itself first. Still, for a fast vendor scan, this is a solid resource. Pair it with independent tests before committing.<\/p>\n \ud83d\udd17 Semrush Blog<\/a><\/p>\n Reddit CEO intends to show more reviews and recommendations across feeds and search, and I think this shift matters for every SEO professional tracking long-tail content. Reddit currently restricts feed recommendations to one week of content. That constraint leaves a massive corpus of timeless posts\u2014parenting advice, book or movie reviews\u2014unsurfaced. Huffman called the improvement potential an “order of magnitude” opportunity and said Reddit will tackle the work over the next year.<\/p>\n COO Jen Wong tied search directly to the shopping experience. She outlined rich media modules with product visuals plus Shopping Listing Ads that combine multiple retailers in one unit. This confirms Reddit is moving beyond fresh-content feeds toward a discovery engine for evergreen queries.<\/p>\n I recommend SEO teams monitor Reddit’s feed and search changes closely. Once Reddit fixes its recommendation models, it could compete directly with publishers ranking for review and “what should I buy” keywords. The platform’s trajectory deserves a spot in your 2026 content strategy discussions.<\/p>\n \ud83d\udd17 Search Engine Journal<\/a><\/p>\n Tinuiti’s Q2 2026 benchmarks challenge the easy story that paid search absorbed organic’s collapse forever. I think this piece is a must-read for SEO and PPC teams planning H2 budgets.<\/p>\n The data is specific and useful. Google paid search clicks grew 14% YoY in Q1 2026. Q2 spending growth slowed to 14%, versus 15% in the same quarter of 2025. YouTube spending growth fell from 20% to 15% YoY. Instagram dropped to 17%. Amazon Sponsored Products jumped 38%, though Prime Day’s shift from July to June inflated that figure.<\/p>\n Interestingly, the article links paid growth to organic decline. Pew found users clicked a traditional result on 8% of visits with AI summaries, versus 15% without. Ahrefs tied AI Overviews to 58% lower CTR for top-ranking pages.<\/p>\n I recommend reading this before finalizing your Q3 forecasts. The Q2 slowdown partly reflects tougher comparisons, but it signals paid growth is getting harder to sustain.<\/p>\n
\nImproving GPT-5.6 Sol in ChatGPT\u2014and expanding access to GPT-5.6 Luna for free users<\/h3>\n
\nAdvancing responsible AI across Europe<\/h3>\n
\nBuilding abundant intelligence<\/h3>\n
\n
\nNew ways to learn and teach with ChatGPT Work and Codex<\/h3>\n
\n\ud83e\udd16 GEO\u00b7SEO Highlights<\/h2>\n
AI Referrals Are Recreating The Oldest Mistake In Conversion Optimization<\/h3>\n
\nWhat Top Stories Inside AI Overviews Means For Publishers And Brands In 2026 And Beyond<\/h3>\n
\nI Helped Scale Google Ads To Billions \u2013 Here’s How I’d Build An AI Search Strategy Today<\/h3>\n
\nHow topical authority spreads (and where it doesn’t) in ChatGPT [Study]<\/h3>\n
\n
\nUtilizing Bing Product Feeds for Ecommerce AI SEO<\/h3>\n
\nWhat 50k Query Fan-Outs Reveal About Brands (+ Free Data!)<\/h3>\n
\nAI Visibility Measurement: What To Track & What To Ignore<\/h3>\n
\nGoogle Now Reports AI Search Impressions. Here’s How To Read Them<\/h3>\n
\nAI Search Data Is Now in Google Search Console: Here’s What You Need to Know<\/h3>\n
\nHow to Build a Full SEO Analysis Dashboard with the Moz API MCP Server<\/h3>\n
\nThe State of GEO w\/ Profound’s Josh Blyskal<\/h3>\n
\nI Tested 15 of the Best Brand Monitoring Tools: Here Are My Top 8 Picks<\/h3>\n
\nThe Future Of Discoverability<\/h3>\n
\nWhen AI Takes The Click, Click Worthiness Should Guide Your Strategy<\/h3>\n
\nReport: Cloudflare AI Bot Blocking Can Prevent Googlebot From Indexing Sites<\/h3>\n
\nThe Page 2 Podcast: What SEOs must know to get chosen by AI<\/h3>\n
\n7 best AI visibility tracking tools for agencies by need<\/h3>\n
\nReddit CEO Intends To Show More Reviews And Recommendations<\/h3>\n
\nPaid Search Absorbed Organic’s Collapse \u2013 Then Q2 Slowed Down<\/h3>\n