The New Search Landscape: Why AI is Deprecating SEO, Not Killing It
A recent panel discussion featuring Lily Ray (Amsive) and Mike King (iPullRank), moderated by Nick Lafferty (Profound), tackled the question every marketer is asking: Is SEO (Search Engine Optimization) dead?
The consensus suggests a more complex reality: SEO is being deprecated and forced to evolve into a more holistic discipline, often dubbed AEO (AI Engine Optimization) or GEO (Generative Engine Optimization). The key takeaway is that the traditional tactics that guarantee success in Google no longer reliably translate to visibility in platforms like ChatGPT or other LLMs.
The Divide Between Classic Search and AI Search
The core challenge facing marketers is the disconnect between traditional search rankings and AI citations.
The 19% Problem: Mike King highlighted data showing that being ranked number one in classic Google search only translates to roughly a 19% likelihood of being cited by an LLM. This low probability necessitates a change in approach, moving beyond single-channel SEO strategies.
The Continued Importance of Fundamentals: Lily Ray cautioned against abandoning SEO entirely, stressing that foundational technical practices—like optimizing for E-E-A-T (Experience, Expertise, Authority, Trust), site speed, and crawlability—are still critical. In many cases, being indexed by Google or Bing is a prerequisite for the LLM's grounding process, meaning SEO is the "cost of entry" for AIO.
A New Name, a New Opportunity: The panelists agreed that the messy emergence of new names like AEO or GEO is an opportunity to reframe the discipline. Mike King argued that calling it something new allows the industry to shed the perception of SEO as "free traffic" and secure greater executive investment for this higher-stakes area of AI visibility.
The Evolving Role of the SEO Professional
The new search environment demands that the SEO professional break out of traditional silos and become a cross-channel strategist.
Cross-Functional Collaboration: The role is evolving to require deep integration with other teams, including social media, video (YouTube), and PR. Since LLMs are multimodal (consuming text, video, audio, and images), SEOs must ensure content is repurposed effectively across all channels.
The Content Shift: Building for Agents: A major tactical takeaway is the need to create content designed specifically for machine consumption, not just human readability. This involves adding an "extra layer of information" to pages, or creating content solely for agents, to make the brand's statements and product facts completely unambiguous and transparent for the LLMs.
Anticipating Queries: The new strategy involves a "query fan out" optimization, which maps out every possible question a consumer might ask about a brand or product to ensure a clear, authoritative answer is available on the site for the AI agent to cite.
Measurement, Risk, and Model Volatility
The biggest unknowns remain attribution and the stability of AI models.
Three Buckets of Metrics: Measuring AIO/GEO success requires a refined framework, stratified into three categories:
Performance Metrics: Traditional metrics like referral traffic and conversions.
Channel Metrics: AI-specific metrics like Share of Voice and Citation Rate.
Input Metrics: Manipulable factors like Passage Relevance Scores and bot activity.
The Need for Data: The panelists stressed the need for LLM providers to offer first-party data (similar to Google Search Console) to help marketers segment, attribute, and understand how people are engaging with AI responses.
Model "Algorithm" Updates: LLM updates—such as data refreshes for the underlying models—function as dramatic "algorithm updates" that can change citation patterns and referral traffic "on a whim". Marketers must anticipate these changes will become routine.
A Word of Caution: While LLMs may not have "web spam" guidelines yet, Mike King issued a clear warning regarding dual-optimization strategies: "Do not ever piss off Google". Brands must ensure any AIO-focused tactics (like cloaking for the AI agent) do not violate Google’s guidelines, as this would compromise visibility in the existing, massive organic search channel.