Shay Ohayon

Following great feedback on Fanout Explorer, I went 2 𝘀𝘁𝗲𝗽𝘀 𝗳𝘂𝗿𝘁𝗵𝗲𝗿 and built free versions for Claude and Gemini too 🤩: 𝗗𝗼𝗲𝘀 𝗥𝗲𝗱𝗱𝗶𝘁 𝗴𝗲𝘁 𝗰𝗶𝘁𝗲𝗱 𝗼𝘂𝘁𝘀𝗶𝗱𝗲 𝗖𝗵𝗮𝘁𝗚𝗣𝗧, 𝗼𝗿 𝗼𝗻𝗹𝘆 𝗰𝗼𝗻𝘀𝘂𝗹𝘁𝗲𝗱? I compared all three on “𝗯𝗲𝘀𝘁 𝗦𝗘𝗢 𝘁𝗼𝗼𝗹𝘀 2026.” 👀 👇 I ran ChatGPT Fanout Explorer vs the versions I created for Claude and Gemini (all available for free download below) to deep-dive into their fan-outs, what they have in common, what differs and what can be learned from it. Across this and my first test, I looked at the following prompts across the main 3 LLMs: ✍ 🤖 AI live chat tools (I want real user opinions, not just vendor pages) ✍ 🤖 Shopify vs WooCommerce for a small ecommerce business ✍ 🤖 best SEO tools 2026 👀 1️⃣ Asking specifically for user opinions changes the source mix: If you ask for real user opinions then Reddit/reviews enter the picture. If you ask for comparisons or “best” tools then ChatGPT leans much more towards first-party, while Claude and Gemini lean far more on third-party blogs, roundups and comparison pages 2️⃣ Reddit matters, but not for every prompt or LLM: ChatGPT fetched and cited Reddit when asked for real user opinions, Claude searched for it but (Brave) returned none, Gemini searched for it but cited none, and for “best SEO tools 2026” none of them used Reddit at all 3️⃣ Your route into the answer differs by LLM: ChatGPT often rewards your own product, pricing and help content (has to be fresh from the last 30 days), while Claude and Gemini can be shaped much more by what third-party publishers, comparison sites and ranking roundups say about you 4️⃣ Small niche sites can outweigh big publishers in Claude and Gemini Both surfaced smaller third-party sites and long-tail comparison pages that ended up shaping their answers, showing that relevance to the specific fan-out query can matter more than publisher size 5️⃣ Sometimes the prompt = fan-out: Claude and especially Gemini often run fairly direct searches, practically matching the prompts, searching them as they are, then work with what Brave or Google gives them first. In those cases, the fan-out looks much more like a traditional search query than a deep chain of thought 💡 My 2-pence: Don’t optimise for AI search (GEO/AEO or however you call it) as one channel: Test prompts your customers are likely using, see where each LLM goes and make sure your brand is visible across both the first-party (fresh help pages and assets, refreshed every 30 days) and third-party surfaces (reddit and review sites big and small) that shape those answers Huge thanks to @aleyda , @glenngabe, @suganthan and many others for featuring me, testing the bookmarklet and sharing it with others, means a lot! 🙏 Get more insights + download all 3 free bookmarklets (#ChatGPT Fanout Explorer, #Claude Fanout Explorer and #Gemini Fanout Explorer) here👇 Download all 3 Fanout Explorer Bookmarklets here: oritmutznik.com/wp-content/u… GitHub: github.com/oritsimu-new/fano… Full write up: oritmutznik.com/ai-workflows… #seo #ai #geo #aeo

Interrupting my summer holiday to share that my AI Search Audit Framework has been featured on my no. 1 newsletter @aleyda's #SEOFOMO🐐🤩 One of my 2026 new years resolutions was to revive my blog and document my SEO and AI search frameworks, skills and tactics completely for free, in full and in plain English because industry friends who know me know that this is how I roll and I do not gatekeep. Some or these friends are why I do this - some people with years of experience in SEO have taken career breaks and they come to this platform and feel like the train has left the station without them. Nope. Not on my watch, and if this in any way resonates, see the bonus piece I wrote just for you. Watch this space👀 How to Run an Al Search Visibility Audit: Framework, Tools and Free Skill Included ---> oritmutznik.com/ai-search-op… My bonus piece: Al Search Optimisation 101: From SEO to AEO, GEO and Search Everywhere Optimisation ---> oritmutznik.com/ai-search-op… Sign up to Aleyda's #SEOFOMO right now --->seofomo.co/ #seo #ai #aeo #geo

@Hostinger published their full AI SEO playbook. Free. No lead-gen form. Here it is and how to leverage it for your brand 👇 🤓 TL;DR: A practical and transparent AI search case study: 100 articles re-engineered leading to a 52% AI citation share growth in 3 months - their 6-step framework and how to apply it to your brand: 1️⃣ Semantic SEO structure Answer-first intros. Every H2 as a direct question. First sentence under each heading must answer it. 💡 Audit your top 10 articles. Can the intro answer the query in one sentence? Can every H2 become a question? 2️⃣ Topical authority - start with comparisons Build topic clusters but prioritise comparison content first. AI prefers pros/cons, features, pricing, quantifiable data over vague descriptions. 💡 Source comparison topics via Google Autocomplete, Reddit, and your SEO tool. 3️⃣ Make your authority legible to AI (EEAT) Detailed author bios with real credentials, linked to their profiles. Give AI a reason to trust the source, not just the content. 💡 Check your bylines. If they don't show expertise, fix them first. 4️⃣ Make commercial pages citable Informational summaries, HTML tables (features/specs/pricing), and FAQ sections. AI prefers Q&A for direct citation. 💡 Pick your top 3 commercial pages and add one table and one FAQ each. 5️⃣ Remove crawler friction Schema markup, fix 404s, and "create an llms.txt file giving AI platforms explicit crawl permission" 💡 Removing crawler friction - yes, llms.txt - nah. I specifically tested this and no search engine (not AI nor traditional) specifically fetched it once according to log files. 6️⃣ Optimise for AI agents AI agents leave slow, unstructured pages. PageSpeed is no longer just a ranking signal, and no longer just for Google. 💡 Run your top pages through PageSpeed Insights. Under 70 on mobile needs fixing. How they tracked the success of their AI SEO Framework - KPIs: 📍 Visibility - track 5-10 customer questions monthly across AI tools. Monitor movement in 6-8 weeks. 🔗 Citation share - test 10-15 questions in Perplexity and ChatGPT etc. Competitor appearing where you should? Content gap. Allow 2-3 months. 💬 Brand sentiment - ask AI: "What are common complaints about [Brand]?" Narrative off? Publish corrective content. Track quarterly. 🪙 My 2-pence 💡 This is one of the most openly shared AI SEO frameworks I've seen from a brand doing this at scale - and it holds up. 💡 But it's not plug-and-play. It requires structured experimentation to isolate what works for YOUR brand - customer research for real intent, competitor analysis to find gaps you can credibly win. The playbook gives you the method. The work is in testing and iterating. 💡 Timelines matter. 6-8 weeks for visibility, 2-3 months for citation movement. Anyone promising faster is trying to sell you something. 💡 Genuinely one of the rare times I'd say: read the source, not just the summary. Links (looks like there might be more in this series, worth following): hostinger.com/uk/tutorials/a… hostinger.com/uk/tutorials/a… #ai #seo #hostinger #aiseo #llms

Finally someone woke up and chose AI-rank-tracking violence & put together actual data to be able to say(and I quote) that "any tool that gives a "ranking position in AI" is full of baloney"🍿 My thoughts on this excellent research by @sparktoro here linkedin.com/posts/oritsimu_… #seo