SEO AI Hasn’t (Really) Changed SEO CT Moore SEO 8 mins read October 13, 2025 Blog SEO AI Hasn’t (Really) Changed SEO Table of Contents The Evolution of Search What AI Has Changed for Search What AI Has NOT Changed for Search GEO for Ecommerce: ChatGPT Shopping Search SEO is Dead. Long Live SEO The rise of AI Search is ushering in a collapse in CTRs from the Organic SERPs. As LLMs are trained off of our content, and AI Overviews and Google’s AI Mode scrapes it directly into the SERPs, we’re having to completely rethink what SEO means for business. But the more SEO changes, the more it stays the same. And Organic Search is still the most qualified source of traffic of all digital channels. After all, when you type a query into a search bar, you are actively searching for what you need, and in that sense, you’re already one step down the conversion funnel. It just happens that those queries now include prompts, and the search bar extends to the chat window. So the question becomes how do you target these high-intent users in the age of SGE (Search Generative Experience)? How do you build a GEO (Generative Engine Optimization) strategy that ensures your content not only stands out in the SERPs, but pushes through the veil of the AI overviews? Well, the answer is in how online search and discovery are evolving. The Evolution of Search AI search and discovery is really just solving the same problem that Google (and other search engines) set out to solve in the first place. In the early days of the web, we found things through human-curated Web Directories (like ALIWEB and DMOZ). Users would “submit” their sites for review, and then human editors decided which pages were added, and how they were categorized by topic, geography, and subject matter. It was an obvious solution to an emerging problem (i.e. how to find things online), but it was hardly an elegant one, and it certainly wasn’t scalable. So then search engines came along, and they actually crawled websites and indexed them based on the keywords. Of course, the presence of keywords says nothing about the quality of a site’s content, and search results were chockfull of spammy sites. But then two students at Stanford figured out that we can use backlinks to rank sites, and that eventually became Google with the 200 search ranking factors it uses today to measure a site’s relevance and authority. All these innovations improved user search experience, but that experience remained very algorithmic. And interacting with that algorithm feels clunky and cumbersome — and not quite natural. LLMs are changing that. They’re making our search and discovery experience more conversational and social — more natural feeling. And they do that by taking on the initiative of the search experience. They aggregate search knowledge for us, more quickly than we can, and make the experience much more interactive. What AI Has Changed for Search If you’re in the business of selling Page Views and Impressions, you’re probably cooked. I mean, you can fight fire with fire for a while, using AI to scale production as you follow closely behind the internet on its slow march towards death — biding time until the day that those Impressions aren’t worth the pixels they’re rendered on. But if you have a more transactional business model, this disruption to the search landscape represents new opportunities. For starters, a conversion for a transactional site is an actual sale (or subscription), and that’s something that AI can’t replace — at least not for now. Add to that the average LLM visitor being worth 4.4x the average traditional organic search visitor and how half of ChatGPT links pointing to business/service sites, and the SEO/GEO opportunity for transactional sites is significant. What AI Has NOT Changed for Search While the search experience is evolving to be more conversational and interactive, what hasn’t changed is what makes a site relevant or authoritative on any given topic. Building E-E-A-T by creating people-first content is just as important as ever (or even moreso). LLMs still source content from what already ranks in the SERPs, and LLM optimization guidelines share a lot of best practices with traditional SEO, including: Getting brand mentions on trust/cited website Establishing a unique value-proposition Creating original and useful content Making your content accessible (to users, search crawlers, and LLMs alike) Incorporating image and video content Optimizing content at the passage level Tailoring content to target audiences and Managing your brand’s online reputation So ranking in the age of AI, whether it’s through AI Overviews or directly in an AI chat thread, is incredibly similar to ranking in traditional search. The LLMs are simply trying to provide its users with the most useful and relevant content available. So GEO is really just by helping those LLMs provide their users with that level of content. GEO for Ecommerce: ChatGPT Shopping Search Of all transactional sites, none have a greater opportunity to optimize for GEO than shopping sites. After all, they tend to have a wide range of SKUs, often across multiple product categories, each of which represents an opportunity to engage LLM users through their value-added content. Specifically, ChatGPT Shopping Search now offers Instant Checkout so users can complete purchases without ever leaving the chat. And there are 8 key steps merchants can take to optimize for ChatGPT Shopping Search. 1. Add Structured Schema Markup to Your Site Use schema (JSON-LD etc.) to explicitly mark up product info: name, description, brand, images, SKU/GTIN, offers, reviews, etc. This helps ChatGPT (and other AIs) “understand” your products more reliably. 2. Create and Maintain a High-Quality Product Feed A product feed (e.g. for Google Merchant, Shopify, etc.) is a structured data set of all your products. ChatGPT may draw on such feeds. Keep it accurate, refreshed, and in sync with your site. 3. Ensure AI Bots Can Read Your Site (No Hidden or JS-only Content) Don’t hide key product information behind JavaScript; put essential data in raw HTML so crawlers (including GPT’s) can see it. Also, make sure your robots.txt doesn’t block ChatGPT’s crawlers (e.g. “GPTBot” or “OAI-SearchBot”). 4. Write Product Copy That Matches Real Buyer Prompts AI matches by language and phrasing, not just keywords. Study how your customers talk (tools like “People Also Ask,” keyword tools, prompt research). Use that language naturally in titles and descriptions. 5. Feature Your Value Proposition Prominently Make clear, compelling statements about “why this product is better” early on. Reinforce that promise throughout the page (features, specs, reviews). Consistency helps AI connect the dots. 6. Build Strong Authority Beyond Your Site ChatGPT considers external signals: reviews, ratings, community sentiment (Reddit, forums), editorial mentions, awards, etc. Strengthen your brand’s presence and reputation across the web. 7. Sign Up for ChatGPT Instant Checkout / Integrate with OpenAI’s Protocols To enable users to buy within ChatGPT, you’ll need to integrate (if eligible) or apply via OpenAI’s Merchant program or Agentic Commerce Protocol. Be ready with structured feeds, API hooks, and certification. 8. Track Your ChatGPT Visibility Use UTM tags (OpenAI appends utm_source=chatgpt.com) to see traffic from ChatGPT. Monitor which prompts surface your products, how often, how traffic behaves (bounce, conversion), and track trends over time. Also classify prompt types into four buckets (price-based, use-case, feature-based, problem-solution) and test across each of them. SEO is Dead. Long Live SEO Just as the world has been ending since the day it began, SEO has been dying since the day it was born. From the earliest days of search, new players have been entering the search landscape to offer users a new and better way to find the best possible information online. The rise of AI hasn’t meant an end to search engines. Rather, it’s just marked a milestone in their evolution. Indeed, AI has brought search closer to its full potential — i.e. helping users discover the most relevant and useful sources of information online. Share This Article Facebook Twitter LinkedIn Email
Answering 14 SEO Questions You’re Probably Asking Yourself Gregory Caltabanis SEO 12 mins read March 9, 2025