Global digital marketing strategies are adapting to the rise of artificial intelligence assistants, which are increasingly influencing buyer journeys and redefining traditional search engine optimization and attribution models.
This shift introduces Generative Engine Optimization (GEO), a new discipline focusing on influencing an AI model’s learned knowledge and associations, distinct from conventional SEO that addresses web discovery.
Initial data suggests AI assistants currently account for a small fraction of organic search traffic directed to Google. Marie Haynes reported that ChatGPT sessions represented only 0.62 percent of Google organic sessions across a sample of 100 websites, with Google maintaining dominance in total conversion volume.
Despite this, observers note a growing trend where users initiate searches with AI assistants before moving to Google Search, which complicates the definition of a “Google visit” and attribution.
Generative Engine Optimization (GEO) specifically targets a model’s “parametric memory,” aiming to shape the foundational knowledge an AI assistant uses to generate responses. This differs from traditional SEO, which ensures web content is discoverable and accessible for AI systems to retrieve.
A study on AI citations highlighted the significant influence of parametric memory on user behavior. Brands recommended by an AI model without an initial search query appeared in subsequent generated queries 67 percent of the time, compared to 11 percent when not pre-recommended.
This finding underscores how an AI model’s internal knowledge can initiate a search query, rather than simply responding to one, according to the analysis.
The evolving situation requires a dual approach where SEO continues to focus on web discovery and retrieval, ensuring relevant sources are available for AI answers, while GEO addresses the direct influence on an AI model’s learned knowledge base.
Organizations like Google and OpenAI are central to this transformation, as their platforms define the parameters for both traditional search and generative AI interactions.
Source: Search Engine Journal
