Skip to content

Princeton GEO Paper

Full Title: GEO: Generative Engine Optimization Authors: Pranjal Aggarwal, Vishvak Murahari, Tanmay Rajpurohit, Ashwin Kalyan, Karthik Narasimhan, Ameet Deshpande Published: November 2023 (arXiv)

The Princeton GEO paper is the foundational academic work that formalizes the concept of Generative Engine Optimization (GEO). As generative engines (like Perplexity and SearchGPT) shift search from traditional ranked lists to synthesized, LLM-generated answers, the paper defines how content creators can adapt.

Key Contributions

  • Formalization of GEO: Introduces Generative Engine Optimization as a distinct paradigm from traditional SEO.
  • GEO-bench: A large-scale benchmark of diverse user queries across multiple domains used to evaluate how often specific web sources are cited by generative engines.
  • Optimization Strategies: Identifies several geo-tactics (e.g., Authoritative Tone, Statistics Addition, Quotation Addition) that creators can apply to their content.
  • Empirical Findings: Demonstrates that applying specific GEO strategies can boost a source's visibility and citation rate by up to 40% in generative engine responses. It emphasizes that strategy efficacy is highly domain-dependent.

Relationships

  • Introduces the core geo-tactics used to rank in LLM context windows.
  • Defines the benchmark used to evaluate [[perplexity]] and other target engines.