Generative Engine Optimization (GEO)
Bachelor & Master Thesis
Generative Engine Optimization (GEO) studies how brands, products, websites, and information sources become visible in AI-generated answers. As systems such as ChatGPT, Gemini, Perplexity, and AI-powered search increasingly synthesize information directly for users, visibility is no longer determined only by traditional search rankings. It also depends on which sources are retrieved, cited, mentioned, recommended, and repeatedly exposed across generative-search systems.
This thesis will investigate GEO from an empirical and quantitative perspective, with possible research directions including:
- Which content, source, ranking, or page-level characteristics are associated with retrieval and citation by generative-search systems?
- How stable is visibility across repeated queries, prompt formulations, systems, and time?
- How do traditional SEO signals relate to visibility in generative search, and where do SEO and GEO differ?
- How do prompt specification, information needs, or query characteristics influence which sources and products become visible?
- Which statistical models and experimental designs can be used to explain or predict retrieval, citation, recommendation, or source visibility?
- How can GEO measurements be made reproducible, statistically sound, and comparable across systems?
- What practical implications can be derived for publishers, brands, and organizations seeking visibility in AI-mediated search?
The exact research question will be defined together with the student and can focus on areas such as source visibility, citation behavior, recommendation stability, prompt effects, temporal variation, SEO–GEO relationships, or experimental GEO evaluation.
Preferred Candidate
Candidates should have:
- Mandatory knowledge of statistics, including hypothesis testing, regression or comparable quantitative methods, and empirical data analysis.
- Mandatory prior knowledge of SEO and/or Generative Engine Optimization (GEO).
- Strong programming skills, preferably in Python and experience working with larger datasets.
- Basic understanding of information retrieval, search engines, and large language models.
- Ability to design and execute reproducible empirical experiments.
- Interest in Web research, AI-powered search, digital visibility, and quantitative evaluation.
Experience with causal inference, statistical modelling, web scraping, APIs, search-engine data, or experimental research is a strong advantage.
The topic is particularly suitable for students interested in the intersection of Information Retrieval, SEO/GEO, Generative AI, Web Science, and empirical data analysis.
Mentor: Emilija Gjorgjevska
Doctoral Researcher, Technical University of Munich (TUM)
Research focus: Generative Engine Optimization (GEO), Search & Information Retrieval, Generative AI, Web Visibility, and empirical evaluation of generative-search systems.