How LLMs changed SEO: From search to solution engines

We’ve seen a similar pattern at BoloSign. When we publish long-form guides or help docs that are structured for search (clear headings, FAQs, concise summaries), they don’t just perform better on Google — they start showing up inside AI-generated answers too. Here’s what’s really happening behind the scenes: AI models aren’t scraping the web randomly. They’re referencing trusted, indexed content — and in many cases, pulling directly from what Google already considers the best version of an answer. That means the old SEO advice of “just write valuable content” isn’t enough anymore. Now, value has to be sa tructured value: Clear intent-driven sections. Snippet-ready answers. Consistent brand signals across your site. When we optimized BoloSign’s knowledge base and comparison pages this way, the impact was immediate. Not only did organic traffic jump, but queries like “DocuSign alternative for small businesses” started surfacing BoloSign in AI chat results too. This is where SEO and AI search fully converge — where clarity, freshness, and entity consistency become your ranking currency. If your brand creates educational or product-led content, this is the perfect time to rethink how you structure it. AI won’t just find your pages — it will quote them. Kudos to SEO Stuff for breaking this down so clearly. This is the future of discoverability, and it’s already here.

The new SEO isn’t on Google. It’s inside ChatGPT. And most digital businesses have not caught on to that yet. Right now, when someone types “best CRM for agencies” into ChatGPT... There’s a shortlist of 3–5 tools that get mentioned. If you’re not on it, you essentially don’t exist. This is a massive opportunity for businesses that move early. And a serious threat to those who don’t. If you want to rank inside ChatGPT, and get true visibility, This is what you need to do: Step 1: Understand How ChatGPT Pulls Information ↳ ChatGPT doesn’t crawl the live web like Google. It ranks content based on: - Training data - Plugin integrations - Custom GPTs GPT-4o prioritises structured, high-authority content from its training set. Step 2: Reverse Engineer Prompts Your ICP Is Using ↳ Ask: “What would my ideal customer type into ChatGPT?” Use keyword tools like: - Ahrefs - Semrush - AnswerThePublic - AlsoAsked - Keyword Insights AI Prioritise prompts with clear buying intent. Step 3: Build Structured, High-Authority Content ↳ Create SEO-friendly landing pages, rather than useless blogs. Focus on: - “Best X” lists - Product comparisons - Clear how-tos - Decision trees Use formats like bullet points, tables, and pros/cons. Step 4: Get Referenced on GPT Training Set Sites ↳ Want to rank in answers? Be where GPT learned from. Target backlinks and mentions on: - Wikipedia - Reddit - Quora - Medium - Government sites - High DR media This helps with both SEO and visibility inside AI answers. Step 5: Create a Custom GPT Around Your Brand ↳ If you want full control, build a Custom GPT trained on your resources. ↳ Make it public and usable so it shows up in the Explore GPTs section. Example: Use “Compare CRMs” so your custom GPT can show up in the results. Step 6: Maintain and Optimise ↳ Regularly ask ChatGPT (incognito) what prompts you show up for ↳ Update content to match evolving prompt structures ↳ Get added to new “best of” or “top tool” lists quarterly This is the next great opportunity for visibility. Those who rank early will win big.  But those who delay won't be found at all. Have you started optimising for AI-driven search yet? Let us know in the comments. We built Searchable to: - Show where your brand is visible (and invisible) in AI search - Turn insights into higher rankings + AI citations - Optimise automatically for SEO, AEO, GEO It will get your site up to speed without burning you out. Learn more and join the waitlist here 👇 https://lnkd.in/gTfCj6Ht ♻️ Repost to help others learn about AI search optimisation.  Follow Searchable for more on SEO/AEO/GEO.

Visibility of your brand message and solution offering won’t come from Google search anymore. Machine customers are already outpacing human buyers. They retrieve answers, score options, trigger flows. No scrolling. No searching. No decision journey. If your content isn’t structured to be found, cited and selected: you’re out. Competition wins. Not because your offer is weak. Because you’re unreadable to the systems that choose and present. #MachineCustomers #AIVisibility #Retrievability #B2BMarketing #AgentEconomy

Our in-house SEO team has been hard at work for the last 18 months on incorporating Generative Engine Optimization (GEO) and Answer Engine Optimization (AEO) into our SEO service for clients. Searchable has created a great guide on the steps, but the work will take time. I especially like the reminder to reverse engineer what your Ideal Client Profile (ICP) would be asking, and ensuring your content is answering those questions. As they mention, Semrush is a great tool for that and the main tool our agency uses.

The inbound landscape is definitely shifting. As ChatGPT, Claude, and other AI tools join search engines as major discovery channels for prospective customers, we need to think beyond traditional SEO. Here's a few examples of how to optimize for LLM users (who ask longer, more context-rich questions, usually as part of an ongoing conversation) vs. search engine users (who use shorter, context-free queries that often aren't even proper sentences). 🎯 Structure content for direct answer extraction - LLMs are aiming to pull precise information from well-organized content. Use clear headings, definition lists, and structured data. Put key facts in the first paragraph of sections and always think "how would an AI tool summarise this?" rather than "how would Google rank this?" 📚 Go deep, not wide - LLMs cite pages that fully answer questions in one place. Unlike search engines' preferences for keyword-targeted pages, LLMs favor depth. For example, a thorough guide on "Python data visualization" should be cited far more than 10 thin posts on matplotlib, seaborn, etc. LLMs are designed to pull more often from pages that fully answer questions in one place - this is a big difference to many SEO tactics. ✅ Make context explicit - schema markup, publish dates, and credentials matter more than ever. LLMs actually read your methodology and sourcing. Add high-quality schema and metadata (which are also SEO best practices) along with explicit statements like "This guide covers X, Y, and Z." LLMs use this context to assess relevance and authority. And unlike humans who skim, LLMs actually read your "About" sections and methodology explanations on any given page. 🔗 Write for citation-worthiness, not 'clickability' - LLMs cite content they can paraphrase and attribute confidently. Use clear, factual language with proper sourcing and avoid clickbait, ambiguity, or sensationalism. Think academic paper, not viral headline. The goal is to be the source LLMs trust to quote and also remember that SEO engagement factors won't be taken into account in an AI tool either. 💡 Semantic clarity > keyword density - LLMs are designed to understand concepts, not just keywords. Use precise terminology, define acronyms on first use, and explain relationships between concepts explicitly. Write "X causes Y because Z" rather than stuffing "X Y Z" throughout. Natural, clear explanations win over keyword optimization. This means you can repeat key concepts fewer times and explore more of the related ideas - another key difference to on-page SEO. Ultimately, if you can become a trustworthy source that directly answers questions, alongside optimizing specific pieces of content to appear high up in a ranked list, you can generate interest from both LLMs and search engines simultaneously. Have you started adapting your content strategy for the AI era?

SEO is becoming very stupid... 😱 Disney shows on Google Search Black Hat SEO Packages!!! 80% of SEOs try to follow Google guidelines, read/listen to John Mueller saying... "Spam links wont affect rankings", "forget about disavow file", "dont build links", "build Authority... EEAT...etc.". The truth is the ALGO is rigged, BlackHat, Parasite and Repurposed, rule SERPS and for the last 30 months Google has been: 1. Destroying their own Algorythm, update after update. 2. They parked their MAIN source of trust -> giving users a reliable reply to their queries 3. Put their Secondary source of revenue (sponsored), below AI overviews, in an attempt to fight a new market (LLM's), but forgetting of the market they were leading. 4. Recently, after huge drops on Revenue, on an attempt to quickly regain what they lost after 30 months of wrong decitions, decided to place AI overviews on 2nd place, pull up Sponsored, and leave the component generating TRUST: Organic Results, as far as possible from their users. Lets forget about SEO, and think about UX: - New SEARCH users, will choose between tools like: Tik-tok, ChatGPT, Meta, and even Google, to run their searches, and perform their transactions. - Old loyal users like ME, may prefer going to Google, but: +3 years ago, I used to search on Google for answers, scrolled 1-2 sponsored results, and find a few reliable sources of information. Nowadays you need to get to page 3 of Google to find something decent... SO Google is pushing me, a loyal user, to go use ChatGPT, Copilot, etc. because despite I dont like AI's, the difference is huge, scrolling through x3 result pages, and reading useless info, vs getting what CHATGPT wants me to do. Obviously its about time, that LLM's will start getting "monetary influence" to tell you B is better than A. So I dont want to do follow AI advice. But the G. Giant is giving no other choice to their loyal users. Sad time for Google, exciting time for new marketers, because that 93-94% market share Google used to have will spread accross soooo many new platforms R.I.P. Google, and thanks for your service.

Read this if you need to learn about AI and SEO: What's amazing about our view from the balcony with Exit Five is that we see hundreds (maybe thousands) of questions from marketers every year, so I feel like we have a strong sense of what topics are important and top of mind. There are a million questions about AI, but the loudest one over the last year has been about AI and SEO. - I am noticing my personal habits changing to use ChatGPT more - My boss is asking me to look into it and how we rank in AI Search - Google is just answering questions directly vs. sending traffic - Our organic traffic has gone down... - The board has some opinions about AI and SEO and I need to get smart on the topic So our team has put together a free AI SEO course; 100% free (just with your email) and has a bunch of vides you can flip through. Now, a lot of this is changing in real-time, so there are no secret timeless hacks here - but if you want to get SMARTER about this topic, this is a great place to start. In this free course cover: 1. How To Get Featured in AI Search. Learn the signals for getting featured in AI-Driven search (ChatGPT, Perplexity, Google AI Overviews). Make sure your organization is easy for AI to recognize, trust, and quote-- so your brand shows up in AI answers. 2. The AI-Driven Research Stack. Watch the exact workflow to turn raw research into a focused topic map: who you’re targeting, who you’re up against, and where the gaps are, ending with a ready-to-ship content plan. 3. SERP → Brief → Draft System. Turn SERP analysis into production-ready content. Build briefs that capture intent, outline angles, and prompt AI for safe first drafts, without breaking SEO best practices. 4. How to Track AI Traffic in GA4. Set up reports to spot AI-generated sessions and measure impact. Attribute which pages, topics, and formats AI engines surface most, and what’s driving sign-ups and revenue. 5. Plug-and-Play Templates + 30-Day Plan. Copy the exact worksheets, briefs, and reporting frameworks the experts use. Follow a 30-day quick-wins plan to improve AI search visibility while building a sustainable long-term strategy. Here's a link to access this resource now: tinyurl.com/39xu854b

October 21st 2025. A brand new search engine is presented to the world. I will treat this like when Google was released on September 4, 1998. I just asked it this question: "How do I go about getting seo ranked on Chatgpt atlas?" Mark my words, I’m doing everything I can to dominate these search results. Headed to find my goldmines. First: What “SEO on ChatGPT Atlas” Actually Means Unlike Google, ChatGPT Atlas doesn’t have traditional “search result pages.” Instead, it draws from: Open-web content that’s publicly indexable (like your blog posts, YouTube transcripts, and product pages). AI-search integrations (OpenAI’s web search and Bing index). Structured, trusted sources — companies and creators with clearly verifiable websites, authorship, and expertise. So “ranking” here means: Being cited or surfaced inside AI answers, summaries, and snippets when people ask related questions. Step-by-Step: How To Get Ranked on ChatGPT Atlas (and AI search generally) 1. Make Your Site Crawlable to AI Ensure your site is publicly accessible, fast, and indexable (no noindex, robots.txt blocks, or gated content). Include an updated sitemap.xml and schema markup for pages, articles, FAQs, and products. Use clean HTML — AI models read text first, not fancy design. 2. Focus on “How-To” and “What-Is” Content ChatGPT ranks answers that directly solve a user’s question. - Add 3-4 “How to” or “What” questions (and short answers) to every blog. - Use conversational, clear writing — avoid keyword stuffing. - Include your name + company (“Author: Brad Smith, AutomationLinks”) so the AI can credit you. 3. Connect Your Content Ecosystem Atlas uses OpenAI’s own search + Bing + YouTube + social signals. That means: Publish transcripts of your YouTube videos as blog posts. Cross-link to them on Substack, Medium, and LinkedIn (all publicly indexed). Add canonical tags back to your main domain so the authority flows to your site. 4. Optimize for AI Snippets AI prefers clear, scannable formatting: Use short paragraphs (1–3 lines). Add bullet points and headers with clear intent (“Step 1,” “Key Benefits,” “Example”). Provide structured data for FAQs and How-Tos. Include definitions and comparisons (e.g., “Automation vs Manual Follow-up”). 5. Build an AI-Authority Graph To appear in AI answers, your online presence must match across: Website “About” page LinkedIn / X profile YouTube channel Press or podcast appearances When those all reference the same person + brand + domain, ChatGPT Atlas treats it as a trusted expert source. 6. Submit and Monitor Use tools like Perplexity, ChatGPT, and Bing Copilot to ask: “Who is [your name] from [brand]?” or “Best AI marketing agencies.” Track when your site appears in answers or citations — that means it’s indexed. Keep publishing fresh educational content; recency matters to AI. Anyone else going to take advantage of this once in a lifetime opportunity? This could make me more money than Bitcoin!

I just read this line from and 'SEO' industry latest report: "The goal is no longer to rank #1 on a list of links. The goal is to be cited in the AI-generated answer." This is wrong. The goal is absolutely NOT to be cited. The goal is to be recommended. For the best part of the last 3 years I've said 'traffic is vanity, profit is sanity'. Well, citation as a goal is stark raving madness. I'll give you an example. I arrived in Amsterdam for the Spotlight event by Semrush on Monday and it was cold, wet and I was hungry. And I had no clue about where I was in relation to anything. I didn't read any blogs. I didn't watch any TikToks. I just went to ChatGPT, and it gave me a list of fantastic restaurants, fantastic locations to visit, and even step-by-step guidance on exactly how to find them from walking right outside my hotel. ChatGPT recommended businesses to me and I purchased from those businesses. The goal of generative engine optimisation is to increase the likelihood that a business is recommended to AI users in buying situations. It 100% is NOT to have your blog content cited. Now, if your goal is to have blog content cited, all you need is SEO. And if you think that's all GEO is, then yes, SEO is GEO. A citation is an acknowledgment of a source you have used in your work. A recommendation is an act of suggesting or endorsing someone or something as worthy or suitable. Now, where a citation is useful if it's going to give you the potential to influence an AI response on a subject or topic. And I've seen AI responses influenced by content of others where that content is false. Because AI doesn't fact-check. So there is a lot of manipulation that can be done using AI on any subject. So, for online reputation management or to influence a worldview, citation could be of use. But it’s not the citation that matters; it’s the synthesised response. Now, did those restaurants do any SEO or GEO to be recommended for me in Amsterdam? Of course not. They did brand marketing, the very thing that most people have forgotten about for the last decade. And largely that is GEO. GEO could defined as supplying the 'network' with relevant information about your business/ brand. In essence, GEO is all about the supply of relevent information in the right volumes, and in the right places online. I know SEOs want to desperately keep hold of their jobs, roles and importance. But that has to start with commerciality. And there is literally zero commerciality to be cited by AI. The goal is to be recommended to people in buying situations. That's where all the value in GEO is. That's why it's different from SEO, and that's why it's going to matter more and more.

How to rank in AI search (yes, it’s already a thing) AI + SEO Workflows was the clear winner in last week’s poll. So let’s start with one of the biggest shifts happening right now: Semrush predicts that by 2028, traffic from AI tools like ChatGPT and Perplexity might beat Google. If that feels far away → remember 2022? That was 3 minutes ago. So… how do you actually show up in AI answers? ➤ Here are 6 proven tactics from industry experts: 1. Track and benchmark your AI presence Alex Birkett recommends tracking: → Visibility and sentiment in AI tools like ChatGPT, Perplexity, and Google AI Mode → Referrals from AI-specific sources (check GA4: chatgpt. com / referral, etc) → AI bots in your server logs (like ChatGPT-User, PerplexityBot) → Any link between AI mentions and signups, demos, purchases 2. Build brand authority and be present where it matters Kevin Indig found that brand search volume is a top predictor of AI mentions → Contribute to trusted sites → Collaborate on webinars, reports → Be active in forums like Reddit and LinkedIn → Repurpose your strongest content across formats 3. Structure your content for AI extraction → Use proper heading hierarchy (H2, H3) → Keep sentences short and context clear → Add semantic HTML and schema (like FAQ Page, How To) → Use bullet points, tables, and Q&A blocks to make answers easy to lift 4. Refresh your content regularly → 90% of AI citations go to content updated in the past 3 years → Update data, tools, examples, screenshots, and broken links → Reflect recent trends and reoptimize layout 5. Build content around real customer questions → Use sales calls, support tickets, chat logs, and GSC to find recurring queries → Include “AI search” in attribution questions (“How did you hear about us?”) Anj Viray used this to create prompt-driven content that ranks for niche queries 6. Use video as a citation magnet → YouTube is the 3rd most cited domain in AI tools → Create how-tos, product demos, and tool comparisons → Target high-intent questions that match AI prompts This isn’t just SEO anymore. It’s SEO ➝ PR ➝ analytics ➝ brand marketing. ➤ Tools I use to track all this: ↳ GA4 ↳ @OtterlyAI → 🎁 Comment “OTTERLY” and I’ll DM you my link to get 10 FREE keywords ↳ Semrush ↳ 50 free SEO tools I found in the @AI SEO Academy What about you? What tools do you use? What strategies do you test? Let’s exchange ideas and grow together!