GEO - Forschungsstand KI-Suche Oktober 2025

1758728680529.pdf

The document discusses the shift from keyword-based SEO to a topic-based approach, referred to as Topic-First SEO. It explains that keywords are no longer the primary focus, but rather a part of a larger content ecosystem centered around topics. The document highlights the limitations of keyword-only thinking, including surface-level content, cannibalization, and blind spots. It then outlines the benefits of Topic-First SEO, such as deeper content, stronger user trust, and more durable visibility. The document also provides a strategic framework for implementing Topic-First SEO, including building content systems, auditing and grouping content by topics, filling gaps across personas and intents, and tracking topics rather than just keywords.

1759330448175.pdf

The document discusses the limitations of traditional keyword-based SEO and advocates for a topic-first approach. It highlights the problems with keyword optimization, such as fragmented rankings and cannibalization, and presents a four-step solution to operationalize topic-first SEO. The steps include building a topic map, adding audience lenses, expanding with subtopics, and optimizing links by topic. The document also introduces the concept of a 'Topical Authority Flywheel' and provides metrics to measure the effectiveness of topic-first SEO, such as tracking performance by topic and advanced metrics like Topic Coverage Score and Topical Authority Score. The approach promises a measurable lift in 10-30 days after cluster refresh.

AI Search Guide_DE_Rev-2.pdf

Die zunehmende Verwendung von KI-gestützten Suchmaschinen verändert die Art und Weise, wie Menschen nach Informationen suchen. Unternehmen müssen ihre SEO- und Marketingstrategien anpassen, um in dieser neuen Landschaft wettbewerbsfähig zu bleiben. Der Leitfaden erläutert die Bedeutung der KI-Suche, stellt wichtige Akteure vor und gibt Handlungstipps, um Inhalte für KI-gestützte Plattformen zu optimieren. Er behandelt Themen wie semantische Suche, Kontextbewusstsein und multimodale Suche sowie die Notwendigkeit, Inhalte so zu strukturieren, dass KI-Modelle sie leicht interpretieren können.

Operationalizing Topic-First SEO | Kevin Indig

Kevin Indig, a Growth Advisor and Hypergrowth Partner, shared a post about operationalizing topic-first SEO, emphasizing the importance of mapping topics for sustained visibility beyond 30 days. He linked to a detailed memo on the subject, which provides frameworks and resources for implementing this strategy. The post garnered significant engagement, with professionals appreciating the valuable insights and resources shared. Commenters noted the strategy requires a mindset shift but expressed enthusiasm for its potential impact.

🔥☝️Should I create comprehensive holistic content or short specialized content for AI Search?👇 I wrote an article about this question in my blog. Here's a summary of the key points from the… | ✌️Olaf Kopp ☀️

The document discusses the optimal content strategy for AI search optimization, debating between comprehensive holistic content and short specialized content. Olaf Kopp argues that Large Language Models (LLMs) struggle with processing long, unstructured text, particularly information in the middle. While comprehensive content covers the entire 'intent space' of a topic, it can be challenging for AIs to process if not structured correctly. The proposed solution is a hybrid approach: being comprehensive in scope while structuring content into short, digestible, 'passage-ready' segments. This involves using clear headings, subheadings, bullet points, and short paragraphs to make content scannable for both humans and AIs. The recommended strategy is to create content that is comprehensive in coverage but highly structured and chunked for easy processing by LLMs, favoring shorter, specialized content when in doubt.

#geo #ki #content #marketing #seo #contentmarketing | Benjamin O'Daniel | 28 comments

☝️The Chunk length at Gemini is 8172 characters.👇 I talked a lot about generative passage-based retrieval according to AIOverviews, GEO ... Here my comprehensive article… | ✌️Olaf Kopp ☀️

The author discusses the importance of understanding the maximum length of chunks or passages for LLM readability optimization, particularly for Gemini models. According to Google's AI Developer documentation, the maximum number of tokens per chunk is 2043, and since 1 token is equivalent to about 4 characters, the text window for chunks is approximately 8172 characters. The author suggests creating chunks of not more than 8000 characters to prevent the 'lost in the middle' problem. The discussion also touches upon hierarchical chunking and its implications for passage-based retrieval.

#seo #googlev #ai #ki | Christian Kunz | 20 comments

Google is testing AI-generated descriptions in search results, marked with a star symbol, currently limited to Reddit results. This development may eliminate the need for website owners to optimize meta descriptions, as Google can generate context-specific descriptions. The change could impact SEO practices and the role of meta descriptions in search engine results.

Reddit's market cap drop: a Google-OpenAI story | Kevin Indig posted on the topic | LinkedIn

Reddit's market cap dropped $6.5 billion after ChatGPT citations decreased, attributed to Google removing the num=100 parameter from search results, affecting OpenAI's access to Reddit content. This change highlights the control Google has over information pipes in the AI era and the vulnerability of platforms like Reddit. The shift is accelerating changes in AI search, with OpenAI testing direct checkout and Reddit renegotiating deals with both OpenAI and Google. User-generated content remains dominant in LLM answers, but structured sources like Wikipedia may gain prominence.

#ecommerce #ai #chatgpt #instantcheckout #agenticcommerce | Fabian Jaeckert | 21 comments

Fabian Jaeckert's LinkedIn post discusses how ChatGPT differs from Google by using organic relevance instead of paid advertising for product recommendations. OpenAI's model earns revenue only when a purchase is completed, charging a small transaction fee. This approach eliminates bid auctions, click prices, and opaque campaigns, focusing on genuine recommendations based on relevance, quality, and context. For businesses, this means visibility is based on relevance rather than payment, and they only pay when they generate sales. The post highlights a significant shift in how ecommerce could operate, emphasizing the importance of becoming relevant rather than 'buying' visibility.

#ai #ki #seo #gemini #chatgpt | Christian Kunz

Christian Kunz's LinkedIn post discusses the growing traffic share of Google Gemini in the generative AI landscape. According to Similarweb data, OpenAI remains the leader with 73.8% traffic share, down 4.3 percentage points from the previous month. Google Gemini increased its share from 9.1% to 13.7%, while Deepseek followed with 3.9%. The overall traffic for generative AI tools surged in September, with Google Gemini experiencing the most significant growth. The post attributes Gemini's success to the quality of its results and new features like Gemini Pro 2.5 and 'Nano Banana'.

🔥☝️The Transition Towards Understanding User Intent and Providing Direct, Comprehensive Answers👇 The landscape of search has undergone a significant transformation, moving beyond the simple… | ✌️Olaf Kopp ☀️

The landscape of search has transformed from simple keyword matching to understanding user intent and providing comprehensive answers. Traditional search engines indexed documents based on individual words, failing to capture conceptual relationships and often returning imprecise results. Modern search technology has evolved to classify queries as 'answer-seeking,' develop processes to select and score explanatory answer passages, and present high-scoring passages as potential answers in formats like 'answer boxes.' This transition aims to provide direct answers to specific questions rather than just a list of resources. Olaf Kopp's article discusses the technologies aligned with this transition, including generative passage retrieval and optimizing LLM readability.

Usability study of Google's AI Mode: insights and observations | Kevin Indig posted on the topic | LinkedIn

A usability study of Google's AI Mode was conducted with 37 participants across 250 tasks. The study revealed that clicks were rare, with a median of zero external clicks per task. When clicks occurred, they were mostly transactional. The AI Mode held users' attention, with most participants staying within the panel. 88% of users interacted with the text, and product previews functioned like mini product detail pages, allowing for quick spec checks with few exits. The study's findings have implications for SEO and content strategies, as users are getting answers directly on the results page, potentially reducing the need for clicks.

Verändert AI die Suche grundlegend? Ja. Verschwindet SEO deshalb? Sicher nicht. Aber es wird Zeit, sich auf die nächste Ära vorzubereiten. Im Podcast mit Torsten Schwarz, live vom OMT, habe ich… | Johannes Beus | 15 comments

Johannes Beus discusses how AI is changing search and SEO. He argues that while AI fundamentally alters search, SEO won't disappear but will evolve. Beus outlines four key points: returning to content fundamentals with new measurement tools, leveraging customer service and sales as content sources, utilizing genuine user sources like Reddit, and striving for excellence in content to rank among the top 3%. He emphasizes that authenticity and expertise will become crucial ranking factors.

#seo #google #aioverviews #aimode | Christian Kunz

Liz Read, Google's Vice President of Search, discussed the importance of creating high-quality, in-depth content to succeed in Google's AI Overviews and AI Mode. She emphasized that users want more than just quick answers and are looking for unique perspectives and comprehensive information. Read also stated that the same mechanisms used for traditional search rankings are applied to AI-generated results, and that the goal of AI is to complement, not replace, traditional search. The article highlights the need for content creators to focus on providing a good user experience and to differentiate their content from AI-generated summaries.

#google #aimode #kimodus #suche #ki #ai #deutschland | Philipp Justus | 80 comments

Google has launched its AI mode search feature in Germany, called Kaimo Dos. This new search mode combines the speed and accuracy of traditional search with advanced AI capabilities, allowing for more conversational and complex queries. Users can interact with AI mode through text, voice, image, or live video. The feature is currently being rolled out in Germany and across Europe. Some users have expressed concerns that AI mode may negatively impact content creators by reducing website traffic and not compensating them for their content.

Google E-commerce SERP Features | Kevin Indig

Kevin Indig's LinkedIn post discusses Google's role as both a competitor and distributor for e-commerce businesses, highlighting the importance of understanding Google's SERP features. The post references a detailed analysis on Growth Memo, suggesting that SEO is becoming more like 'shelf placement' than content discovery. Comments from industry professionals like Lily Grozeva and Gideon Ugbeh discuss the implications for SEO strategies and regional variations in e-commerce SERP features.

5 ways to stay visible when Google stops sending clicks | Kevin Indig

Kevin Indig shares five strategies to maintain online visibility in a changing search landscape. First, optimize content for AI overviews by targeting longer, more specific queries and structuring content like featured snippets. Second, influence users before they search through advertising, quality content, PR, and brand collaborations. Third, invest in platforms like YouTube and Reddit where users validate information. Fourth, make content easily skimmable with key information, bullet points, and FAQs. Additional commenters suggest diversifying traffic sources, strengthening brand authority, optimizing for zero-click searches, and providing original data.

How to identify AI Overviews impact on SEO with GSC and RegEx | Kevin Indig posted on the topic | LinkedIn

Kevin Indig shares how to use SEO Gets and RegEx to identify sites losing clicks due to AI Overviews in Google Search Console (GSC). For large sites, creating separate GSC properties for subdirectories helps compare AI Overviews' impact on different page templates. SEO Gets offers features like 'people also ask' filter preset using similar RegEx. Commenters praise SEO Gets for its utility in tracking SEO changes and analyzing AI Overviews' effects.

#seo #ai #ki | Christian Kunz | 13 comments

Christian Kunz discusses the importance of optimizing for AI search despite its current low traffic share. He argues that AI search is the future and that optimizing for it is crucial for visibility. Kunz emphasizes that classic SEO remains important as it benefits presence in AI environments. The post sparks a discussion among SEO experts on the balance between optimizing for traditional search and AI-driven search.

#seo #google #sea #googleads | Christian Kunz

The document displays LinkedIn's content categories and a sign-in page. The content categories include Career, Productivity, Finance, Soft Skills & Emotional Intelligence, Project Management, Education, Technology, Leadership, Ecommerce, and User Experience. The sign-in page prompts users to log in or create a free account to access more content. It also includes links to LinkedIn's User Agreement, Privacy Policy, and Cookie Policy.

Is AI Cutting Into Your SEO Conversions? | Kevin Indig

Kevin Indig's LinkedIn post discusses how AI is affecting SEO conversions, suggesting that AI-driven changes are altering attribution models and potentially making SEO appear less effective. The post links to a detailed memo on the topic. Comments from industry professionals like Luca De Berardinis and Kanaar Bell discuss the implications of AI overviews on attribution and the need for a mindset shift in understanding SEO effectiveness beyond traditional attribution models.

#wikipedia #ki #marketing #pr | Benjamin O'Daniel

The document discusses the importance of Wikipedia for AI systems, particularly in providing information to chatbots like ChatGPT. Jörg Niethammer, an experienced SEO expert and Wikipedia contributor, shares insights on how companies should manage their Wikipedia presence. The post highlights that Wikipedia is often used as a top source by AI systems when searching for company information. It warns against companies using Wikipedia as a PR channel, as such attempts are usually quickly reverted by the community. The discussion emphasizes the need for companies to understand Wikipedia's rules and the value of having experts manage their Wikipedia presence.

#seo #llmo #aioverviews #chatgpt | Christian Kunz | 14 comments

Christian Kunz shares insights from an analysis by Mike Korenugin on factors influencing citations in Large Language Models (LLMs) and Google AI Overviews. The study found that domain traffic and the number of linking domains are key factors for both ChatGPT and AI Overviews. For AI Overviews, content length is also significant, while for ChatGPT, Interaction to Next Paint (INP) is crucial. The post discusses the implications for SEO strategies, emphasizing the importance of strengthening domain presence, technical optimization, and creating in-depth content.

#seo #ai #ki #llmo | Christian Kunz | 20 comments

A recent study by AirOps reveals that brand mentions in AI searches and Large Language Models (LLMs) are dominated by third-party websites, with own websites playing a minor role. The study found that mentions of brands in AI searches and LLMs are 6.5 times more likely to come from third-party websites than from the brand's own website. The frequency of mentions varies significantly between different models. The study emphasizes the importance of onsite authority and offsite presence for marketing professionals to achieve visibility in AI search. It suggests that strategies for owned and earned content must be treated as a comprehensive package to improve recognition and discoverability.

"AI Mode studies reveal zero-click dominance and influence" | Kevin Indig posted on the topic | LinkedIn

A meta-analysis of 10+ studies on AI Mode reveals that 92-94% of sessions result in no external clicks, with users spending 52-77 seconds reading AI answers per task. The studies show that brand familiarity drives decisions without triggering clicks, and that domain authority correlates stronger than individual page optimization. The findings indicate a shift from traffic to influence, with attribution becoming increasingly complex. The research suggests that investment in AI visibility tracking tools is becoming essential, and that the next decade will belong to those who master algorithmic influence without clicks.

Die KI-Browser-Schlacht wird heiß(er): OpenAI schickt Atlas ins Rennen 🚀 Und plötzlich ist er da. Heute hat OpenAI global seinen Browser "ChatGPT Atlas" gelauncht – eine Mischung auf Browser und… | Jörg Schieb | 12 comments

OpenAI has launched ChatGPT Atlas, a new browser that integrates AI capabilities directly into the browsing experience. Atlas features a 'Ask ChatGPT' sidebar that can summarize webpages, compare products across tabs, and navigate the web autonomously in 'Agent Mode'. The browser has a memory feature that can recall previous activities and derive industry trends. While basic features are available for free, 'Agent Mode' is limited to Plus, Pro, and Business users. Atlas competes with Perplexity's Comet browser, which also offers AI-driven features. The launch signals increased competition in the browser market, potentially challenging Google Chrome's dominance.

#atlas #browser #openai #chatgpt #geo | Fabian Jaeckert | 18 comments

Fabian Jaeckert discusses OpenAI's new Atlas Browser and its potential impact on GEO (Google Search Engine Optimization) and SEO. The browser integrates AI capabilities, eliminating the need to switch between different applications and browsers. It allows users to interact with web pages directly within the browser using AI agents. This development is seen as a significant change in how users interact with the web and could potentially disrupt Google's dominance in search. The post highlights three main implications: 1) the elimination of the 'media break' in AI usage, 2) a frontal attack on Google's search dominance, and 3) the emergence of AI agents as new users, which will change how websites are optimized and interacted with. The post also includes comments from various professionals discussing the potential benefits and drawbacks of the new browser, including concerns about privacy, functionality, and its potential impact on SEO and LLMO (Large Language Model Optimization).

#seo #google #ki #ai #aimode | Christian Kunz | 15 comments

Multiple studies show that Google's AI Mode significantly reduces clicks on search results, with 92-94% of sessions having no external clicks. The number of clicks is expected to drop as AI Mode becomes standard. This change affects SEO strategies, emphasizing domain strength and visibility in AI platforms over traditional rankings.

#seo #google | Christian Kunz

According to BrightEdge, Google's market share in search engines has risen for the first time in over nine months. Previously, Google's market share had been declining, with reports suggesting it dropped below 90% earlier this year. The increase is potentially linked to Google's AI Mode, as indicated by growth in long-tail queries on Google.de. Despite this, the broader AI search ecosystem, including ChatGPT and Perplexity, has seen a loss in market share, accounting for less than 1% of referral traffic from search. Smaller players like Claude and Grok continue to grow, but at a slower rate, with Claude's referrals being less than 2% of ChatGPT's and Grok's at 0.5%. Google's dominance remains significant, emphasizing the importance of optimizing for Google, including its evolving features like AI Overviews and AI Mode.

Google AI Mode: Mehr Links, weniger Klicks Seit einigen Wochen stellt Google den neuen AI Mode auch in Europa zur Verfügung. Seitdem haben wir uns die AI Antworten auf viele Millionen Prompts… | Johannes Beus | 23 comments

Johannes Beus, Founder & CEO of SISTRIX, analyzed Google's new AI Mode, now available in Europe. The AI Mode provides more links than traditional search results, with averages ranging from 15.5 links per answer in Germany to 27.2 in the UK. However, most links are hidden behind small source symbols, potentially reducing click-through rates. This shift may increase the importance of visibility in search results while decreasing clicks on individual websites. The post sparked discussions among SEO professionals about the implications for traffic, the role of E-E-A-T in AI-generated results, and the need for new metrics to measure success.

KI liebt strukturierte Daten? ⛔ Im Gegenteil, sie schmeißt sie raus ✂ Es wird ja viel darüber geredet, strukturierte Daten, z.B. im JSON-Format, wäre der große GEO Hack. Mich hat das aus… | Fabian Jaeckert | 25 comments

Fabian Jaeckert discusses how AI models, particularly Large Language Models (LLMs), process and utilize data during training. Contrary to the common belief that structured data in formats like JSON is crucial, Jaeckert presents findings from a study by Chart Deep Research indicating that AI models heavily filter out structured data, favoring unstructured text instead. The study reveals 'strong filtering on free text', 'aggressive cleaning', and removal of structured data as 'noise'. Jaeckert suggests that unique, unstructured information is more valuable for AI training. The post sparked a debate among professionals, with some supporting Jaeckert's claims and others questioning the methodology and conclusions.

#seo #google #seotools | Christian Kunz | 15 comments

Google has blocked another method for SEO tools to retrieve current search results and rankings through the Search Lite API. Previously, SEO tools used a workaround to fetch 100 results with just two requests, but Google has now deprecated this API, limiting it to 10 results per request. This move is part of Google's efforts to protect its search results from unauthorized use, particularly by AI platforms. The change affects SEO tool providers, who must now find alternative methods to obtain current rankings efficiently. Some users are discussing potential workarounds and the implications for SEO tool pricing.

🚨 LLMs Can Now "Deep Read" Your Content at Scale. Google is calling it BLOCKRANK. Google DeepMind's research on BlockRank solves the biggest problem in AI-driven search: efficiency. It makes… | Nick LeRoy | 25 comments

Google DeepMind's BlockRank technology improves semantic ranking efficiency by 4.7x, making 'deep read' of documents linear rather than quadratic. This advancement changes content optimization strategies, focusing on relevance, immediate answer provision, and top 10 ranking. Marketers must adapt to these changes to remain competitive.

#seo #ecommerce #ai #ki #chatgpt | Christian Kunz

A recent analysis of organic Large Language Model (oLLM) traffic in e-commerce reveals that it lags behind traditional channels in key financial metrics such as conversion rate and revenue per session. However, oLLM traffic shows favorable bounce rates, indicating that the content is relevant to users. Despite current limitations, oLLM traffic has shown positive development over the observation period. The average order value (AOV) of oLLM decreased in the first year, partially offsetting gains from improved conversion rates. The analysis compares oLLM traffic to other channels, including paid social media, and suggests that oLLM content may be more effective as an 'upper funnel' touchpoint rather than a direct sales driver.

Im Rahmen der ersten GEO-Konferenz von Claneo in Berlin, war ich vor einigen Wochen auch Gast im Marketing Transformation Podcast von Erik Siekmann. Der Podcast ist jetzt veröffentlicht, den Link… | Johannes Beus | 11 comments

Johannes Beus, Founder & CEO of SISTRIX, discusses the impact of ChatGPT on Google's dominance in search. He highlights that ChatGPT has become the second-largest search engine with 4-9% of Google's volume, decoupling visibility from traffic. AI answer systems are changing how users interact with search results, making 'mentions' in AI responses the new key metric. The era of traditional organic traffic is ending, and businesses must adapt to this new paradigm.

I just finished reading the new Semrush × Statista Ecommerce report - and it’s a reality check for anyone betting too early on AI shopping replacing search. Here’s what stood out to me: 1/ 61% of… | Kevin Indig | 17 comments

Kevin Indig discusses the Semrush × Statista Ecommerce report, highlighting that while 61% of U.S. consumers have used Large Language Models (LLMs) for online shopping, AI-driven traffic remains 300× smaller than organic search. The report suggests AI is reshaping ecommerce by enhancing discovery, inspiration, and comparison, rather than replacing traditional search. Over 50% of consumers use AI shopping assistants, indicating a shift towards hyper-personalized, conversational discovery. The report concludes that organic traffic remains the primary channel with defensible intent and scale.

AI-Search Verbraucherumfrage von McKinsey | Artur Kosch | 15 comments

A recent McKinsey consumer survey reveals that AI search is now used more than traditional search engines, with 44% of respondents using AI search engines like ChatGPT as their first stop for product searches. This shift in consumer behavior has significant implications for brands, with those not optimizing for AI search potentially losing up to 50% visibility. The study also found that by 2028, $750 billion in consumer spending is expected to flow through AI search systems. McKinsey recommends that companies build AI search capabilities as a core competence, including AI-optimized content and interdisciplinary teams.

#seo #sea #google | Christian Kunz | 15 comments

Christian Kunz discusses a mysterious drop in search traffic for some websites in late September, despite no official Google update. He suggests that Google's new ad labeling, which makes ads less distinguishable from organic results, may be the cause. SEO expert Cyrus Shepard shares similar observations, noting that Google's ad layout changes may be shifting clicks from organic results to ads. The change, combined with the upcoming Google AI Mode, may further pressure websites' ability to attract visitors through search.

How to Get on More "Best" Lists | Glen Allsopp posted on the topic | LinkedIn

Glen Allsopp shares strategies for getting featured in 'best' lists that dominate search results. He analyzed ~750 lists and provides 7 tips: finding lost mentions, being on platforms like G2/Clutch, winning awards, checking for affiliate links, sponsored listings, updating content, and creating own lists. Allsopp warns against over-optimizing for listicles and suggests a balanced approach.

ChatGPT ignoriert strukturierte Daten/JSON beim Grounding Hier ist der Beweis (Video) Strukturierte Daten werden aktuell als wichtige GEO-Maßnahme empfohlen. Beweise? Fehlanzeige. Also habe ich… | Fabian Jaeckert | 28 comments

The author tests whether ChatGPT uses structured data (JSON) for grounding and finds that the ChatGPT crawler does not directly access JSON data. The author concludes that ChatGPT may not use JSON for grounding, but notes that other crawlers like OAI-Searchbot might. The discussion involves various SEO experts commenting on the relevance of structured data for ChatGPT and other AI models.

Diese 8 Social Media Quellen kannst du nutzen, um in der Google AI Suche zu ranken Google hat vor wenigen Wochen den AI Mode eingeführt - die neue Form der Suche als Chatbot. Zusammen mit dem… | Felix Beilharz | 41 comments

Felix Beilharz discusses how to improve visibility in Google's AI search by leveraging eight key social media platforms identified by Johannes Beus of SISTRIX. Google has introduced two AI-driven search features: AI Mode and AI Overview. While good SEO remains crucial, these AI searches also draw from external sources, including eight significant social media platforms: YouTube, Facebook, Reddit, Instagram, Quora, LinkedIn, Pinterest, and TikTok. Being active on these platforms not only generates reach but also enhances visibility in Google's AI search. Comments from professionals like Johannes Bertuleit and Christa Goede highlight the importance of authentic engagement and the dominance of US tech giants.

1761764779560.jpg

The image presents a list of the top 50 websites in Germany according to AI Mode, a ranking system. The top 10 websites are highlighted, with youtube.com leading at 40.08%, followed by google.com at 31.69%, and wikipedia.org at 17.83%. Other prominent websites include amazon.de, reddit.com, facebook.com, ebay.de, otto.de, idealo.de, and instagram.com. The list continues with various German websites, including news sites, e-commerce platforms, and social media. The ranking is presented in a table format with the website names and their corresponding percentages. The image also features a blue bar at the top with the title 'AI Mode: Top 50 Quellen in Deutschland' and a search bar. The background of the image is light blue.

28% of ChatGPT’s most-cited pages have ZERO organic visibility in Google 🤯 my awesome colleague Louise Linehan analyzed the 1,000 most-cited pages in ChatGPT, via our ChatGPT index in Ahrefs Brand… | Ryan Law | 58 comments

Ahrefs' analysis of ChatGPT's 1,000 most-cited pages found that 28% had zero organic visibility in Google. The research, conducted by Louise Linehan, suggests that these pages were likely heavily represented in ChatGPT's training data but have since been impacted by Google updates. The findings raise questions about the differences between ChatGPT's citation practices and Google's ranking algorithms.

1761661996255.jpg

The document presents a pie chart analyzing the SEO visibility of the top 1000 ChatGPT-cited pages. It reveals that 28.3% (283 pages) have zero organic keywords, indicating no traditional search visibility, while 71.7% (717 pages) have organic search presence with a median of 279 keywords per page. This data suggests a significant portion of highly cited pages by ChatGPT lack organic search visibility, potentially impacting their discoverability through traditional search engines.

- YouTube

Sneak peek from my upcoming article on LLM citations 👇 This is what, I believe, made our study on SEO pricing one of the most frequently cited pages by various AIs. I know what you're thinking… | Mateusz Makosiewicz | 40 comments

Mateusz Makosiewicz shares insights from his upcoming article on LLM citations, highlighting that a study on SEO pricing became frequently cited by AIs due to its well-structured content. The page was created by Joshua Hardwick before GEO was a consideration, demonstrating that timeless content principles are effective for AI citations. Makosiewicz suggests that the best practices for AI citations aren't new, but rather established content principles done correctly. The post sparks discussion among professionals in the field, with comments touching on the relevance of traditional SEO practices and the potential differences in how LLMs treat newer content.

1761734379095.jpg

The image illustrates the key elements of a highly citable page for LLM-generated answers, using an example article about SEO pricing. The page features a clear and concise answer to a common question, a timestamp with a last update date, and a structured format with key takeaways and subheadings. The content is built on original data and includes a clear expert author byline. The article also explores the topic from different angles, providing a deeper breakdown of costs and including visual aids like bar graphs.

100+ Questions That Show AEO/GEO Is Different Than SEO - AI & SEO Fundamentals

The article discusses the differences between AEO/GEO (AI search optimization) and traditional SEO, highlighting 100+ questions that arise when analyzing AI search systems like ChatGPT and Perplexity. The author, having reverse-engineered ChatGPT's ranking system and analyzed Perplexity's ranking factors, identifies fundamental differences in how these systems operate compared to traditional search engines like Google. Key differences include the use of Reciprocal Rank Fusion (RRF), vector embeddings, temperature settings, and token limits, which create new challenges and opportunities for optimization. The article also touches on issues like hallucination rates, citation fabrication, and the probabilistic nature of AI search results.