AI Search Relies on SEO – And It Knows It
▼ Summary
– AI summaries are compressing organic search traffic, reducing referral traffic to publishers, while Google reports all-time high search queries.
– Publishers expect traffic to halve in the next three years, despite Google’s updates aimed at sending traffic back to websites.
– Large language models use retrieval-augmented generation (RAG) to fetch documents from a search index, relying on the semantic HTML and logical site hierarchy built by SEO professionals.
– Modern SEO includes both traditional site health maintenance and AI-readiness strategies, such as optimizing for RAG extraction and strengthening brand entity signals.
– SEO remains foundational for AI search, as it provides the structured data and signals that LLMs need to verify facts and attribute sources.
As AI-generated summaries shrink the space once reserved for traditional organic listings, referral traffic to publishers is facing a serious squeeze. Search is being used more than ever, yet the interface itself now holds users longer, shifting away from its historical role as a simple switchboard directing clicks elsewhere.
The outlook from publishers is grim. Traffic is projected to drop by half within the next three years. In stark contrast, Google recently reported that search queries have reached an all-time high, signaling what some might call a golden age for digital visibility.
AI has supercharged Search. According to Google, “People are searching on Google more than ever before. Last quarter, we saw an all-time high in Search queries.” The company has also announced updates intended to drive traffic back to websites, though whether this is a genuine effort or a PR move to deflect antitrust scrutiny remains unclear.
What this all reveals is that despite claims that SEO and search engines have been overtaken by GEO, AEO, and LLMs, the opposite is true. Optimizing for search engines remains essential, and technical SEO is the foundation for AI search.
Large language models are probabilistic text-generation engines, not databases or reasoning tools. They don’t retrieve stored facts; they calculate the statistical likelihood of word sequences. To make answers current and grounded, retrieval-augmented generation (RAG) pulls documents from a search index and feeds them to the model before it writes a response. As Jess Peck explained in a December 2024 YouTube video, “Oh my god, ChatGPT is not a search engine.”
For an AI search engine to answer a query using RAG, it depends on a high-quality data pipeline. That requires an organized, easily navigable, and authoritative data source, made possible only through the semantic HTML, logical site hierarchy, and clean indexing that SEO professionals provide.
Who builds, structures, and maintains that data source? The SEO community. They are the ones labeling data, cleaning clutter, and ensuring machines can read what humans write. Without that foundational architecture, AI search engines encounter inefficient paths and poor website structures. SEOs are not bystander victims of the AI revolution.
Modern SEO now includes both legacy site health maintenance and specific AI-readiness strategies: optimizing for RAG extraction and strengthening brand entity signals across the knowledge graph. By structuring data so machines can interpret context, SEO professionals deliver the signals AI search engines use to verify facts and attribute sources. Technical SEO ensures that the “information gain” of a page is accessible to the models that need to cite it. If you want an AI to recommend your product, your digital footprint must support that.
Optimization does not disappear in the age of AI; it becomes the baseline for trust.
Can you optimize for LLMs without an SEO program? The smartest brands are not abandoning SEO for AI search. They are aggressively using SEO to fuel their AI readiness. They understand that AI search does not replace the need for traditional SEO and information retrieval practices; it amplifies that need.
It is easy to dismiss SEO as a relic, but the reality is that SEO has built the product that LLMs now productize and charge for. As Jamie Indigo summarized on LinkedIn, “We should be clear-eyed about what happened – and intentional about what we build next.”
SEO runs the engine room that powers the ship. As AI shifts the digital landscape, the engine room is the best place to be prepared for the future. Looking ahead to an era of GEO and AEO, we must ask: Can you optimize for LLMs without that grounding in SEO knowledge and expertise?
(Source: Search Engine Journal)
