Marie Haynes reviews Google's new documentation on ranking in AI search. She explains RAG, query fan-out, the shift away from commodity content, and the growing importance of experience and agentic search. The video also covers technical SEO myths and practical next steps for e-commerce sites.
A short editorial from the SEONIB team on why this content matters.
Google's new documentation confirms ranking in AI search requires original, experience-driven content over technical SEO tricks.
Unlike generic advice, Google explicitly downplays structured data and technical optimizations, urging a shift to unique viewpoints and agentic readiness.
Every SEO should audit their content for originality and e-commerce sites should set up their business agent in Merchant Center now.
AI technique that uses web content to ground generative answers in factual information.
Process where Google's AI generates multiple concurrent questions from a user's single query to surface diverse content.
Content offering a unique viewpoint, original experience, or analysis that cannot be easily replicated by AI.
Experience, Expertise, Authoritativeness, Trustworthiness; Google's quality framework emphasizing real-world experience.
Search that uses AI agents to perform tasks like booking or comparing products on behalf of users.
Google's Merchant Center feature that lets AI assistants chat with your brand using your site data for accurate grounding.
Model Context Protocol for enabling AI agents to interface with website functionality, still in development.
AI-generated summaries displayed at the top of Google search results, pulling from multiple sources.
What is query fan-out in Google AI search?
Google's AI generates a list of related questions from a user's original query to retrieve a broader range of relevant content.
Does technical SEO matter for AI overview rankings?
Less important than content quality. Core web vitals and minor technical issues are unlikely to hurt rankings unless they prevent crawling or indexing.
Should I use FAQ schema to get into AI overviews?
Probably not necessary; studies show no clear ranking benefit. Google can understand unstructured content effectively.
Is it worth optimizing for AI overviews if clicks are decreasing?
Yes, because AI overviews can drive higher-quality engagement and conversions, even if overall clicks drop.
What is non-commodity content?
Content that provides original information, reporting, research, or a unique personal viewpoint, not rehashed information available everywhere.
Do I need an LLMs.txt file for my site?
No, Google says it is not required. Language models can read regular HTML well; producing markdown versions is unnecessary.
What is Google's stance on AI-generated content?
Allowed as long as it is high-quality and does not violate spam policies. The focus should be on originality and user value.
How does experience (E-E-A-T) help in AI search?
AI cannot replicate real-world experience. Content based on personal experience stands out and is more likely to be cited by AI answers.
What is the business agent in Merchant Center?
A feature that allows AI to interact with your brand using your site data. You can test it by clicking 'Test your agent' in Merchant Center.
Should I focus on Web MCP for agentic search?
Not yet. Google advises focusing on current best practices; Web MCP is still clunky and under development.
SEONIB automatically generates AEO-optimized pages, embeds schema, and submits them to IndexNow — so your content gets discovered faster by Google and AI answer engines.