TLDR: A research paper introduces Generative Engine Optimization (GEO) as a new strategy for visibility in AI-powered search engines. It highlights AI search’s strong bias towards authoritative third-party (earned) media over brand-owned or social content, unlike traditional search. The study also reveals differences among AI engines in content sourcing, freshness, language sensitivity, and a bias towards major brands. It provides actionable guidance for content creators to adapt to this evolving landscape, emphasizing structured content, earned media dominance, and engine-specific strategies.
The landscape of information retrieval is undergoing a profound transformation with the rapid adoption of generative AI-powered search engines like ChatGPT, Perplexity, and Gemini. Unlike traditional search engines that present ranked lists of hyperlinks, these new platforms synthesize information into concise, citation-backed answers. This shift challenges established Search Engine Optimization (SEO) practices and necessitates a new approach, which researchers term Generative Engine Optimization (GEO).
A recent research paper, titled Generative Engine Optimization: How to Dominate AI Search, by Mahe Chen, Xiaoxuan Wang, Kaiwen Chen, and Nick Koudas, delves into this evolving environment. The study provides a comprehensive comparative analysis of AI Search and traditional web search (Google), conducting large-scale experiments across various content verticals, languages, and query variations to quantify the differences in how these systems source information.
Key Differences Between AI Search and Traditional Search
One of the most significant findings is that AI Search engines exhibit a systematic and overwhelming bias towards ‘Earned media’ – third-party, authoritative sources – over ‘Brand-owned’ and ‘Social’ content. This stands in stark contrast to Google’s more balanced mix of content types. For instance, in categories like Automotive and Consumer Electronics, AI search results were heavily dominated by earned media (e.g., professional reviews, publisher domains), often with negligible social content, while Google maintained a more diverse distribution including social platforms like Reddit.
The research also highlights that AI Search services differ significantly from each other in several key areas:
- Domain Diversity: Each AI engine tends to draw from a unique set of sources, leading to low overlap in cited domains across different platforms.
- Freshness: While some AI engines provide relatively fresh content in certain verticals, others, particularly in areas like automotive, showed a reliance on older, more static review pages.
- Cross-Language Stability: The study found varying degrees of stability when queries were translated. Some engines, like Claude, maintained a more consistent set of authoritative English domains across languages, while others, like GPT, effectively swapped to entirely different local-language ecosystems.
- Sensitivity to Phrasing: AI engines generally showed less sensitivity to minor rephrasing of queries compared to Google, especially when the changes were primarily formatting-related.
- Big Brand Bias: When presented with unbranded queries for product recommendations (e.g., “most popular cola brand”), AI systems showed a clear preference for major, market-leading brands over niche or indie brands.
- Local Search: Overlap between AI search and Google for local business queries was significantly lower than in broader consumer verticals, suggesting distinct strategies are needed for local visibility.
The Generative Engine Optimization (GEO) Agenda
Based on these empirical results, the researchers formulate a strategic GEO agenda, providing actionable guidance for practitioners:
1. Engineer for Agency and Scannability: Websites must be designed as an “API for AI.” This means rigorous implementation of technical SEO fundamentals and detailed schema markup (Schema.org) for all product specifications, prices, reviews, and availability. Content needs to be structured for easy machine parsing and synthesis.
2. Dominate Earned Media: Given the overwhelming bias towards earned media, brands must shift focus from solely creating owned content to systematically earning third-party validation. This involves investing heavily in public relations, media outreach, and expert collaborations to secure features, reviews, and mentions in authoritative publications and review sites. Building backlinks from these high-authority domains is crucial for AI’s perception of a brand’s trustworthiness.
3. Engine-Specific Tactics: A one-size-fits-all approach is ineffective. For engines like Claude and ChatGPT, focus on securing positions within globally recognized, authoritative domains. For Perplexity, which incorporates more diverse sources including YouTube and retail sites, consider video content and accurate product listings on major retailer domains. Gemini, with its greater propensity to cite brand-owned properties, allows for a more balanced approach leveraging both earned media and well-structured content on a brand’s own domain.
4. Multilingual Strategy: Localize Authority, Not Just Content: For non-English markets, simple translation is insufficient. For engines that localize heavily (like GPT and Perplexity), building relationships with and earning coverage from authoritative local-language publishers is essential. Simultaneously, strengthening English-language earned media presence can improve visibility on engines like Claude across multiple languages.
5. Content Strategy: Justify and Compare for the Shortlist: AI search aims to justify recommendations on a synthesized shortlist, not just provide links. Content must be explicitly engineered to answer comparison questions, featuring scannable pros and cons lists, comparison tables, and clear statements of value proposition (e.g., “longest battery life,” “best for small families”).
6. Niche Brand Strategy: Overcome the Big Brand Bias: Niche brands must over-invest in building tangible, verifiable authority within their specific niche through deep expert content and targeted earned media campaigns in specialty publications. Leveraging strategies that work on engines like Perplexity, such as high-quality YouTube review content and community engagement, can help build grassroots authority.
Also Read:
- Organic Search Retains Dominance as BrightEdge Reports AI Search Under 1% of Referral Traffic
- Evaluating LLM Creativity in Marketing: A New Benchmark Reveals Nuances
The Imperative for Principled GEO
The study concludes that the AI search landscape is fragmented, dynamic, and highly competitive. Success requires a continuous, principled, and strategic discipline, moving beyond periodic SEO audits to a comprehensive GEO operating system. This system must integrate continuous competitive intelligence, structured content frameworks, systematic authority-building, a defensive and offensive ranking defense system, and an integrated, metrics-driven execution platform. In essence, dominating AI search is an ongoing “arms race” where intelligence, impactful content, strong authority, and rapid response are paramount.
