What Is Generative Engine Optimization?
Generative engine optimization, GEO, is the practice of making a company's content easy for an AI answer engine such as ChatGPT, Claude, Gemini, or Perplexity to retrieve, quote, and name, instead of easy for a search engine to rank. The page you are reading it on is built as a live test of the idea, not a description of one.
Where the term came from
GEO was coined in a 2023 paper by Aggarwal and colleagues at Princeton, who measured how specific content changes moved a brand's visibility inside generated AI answers, the same way "SEO" names tactics aimed at a ranked list. "Generative engine optimization" pulls about 1,000 Google searches a month as of this week, and the close cousin "answer engine optimization" pulls about 720, up sharply over the trailing year. Neither number is large. Both are real, and both are still finding their level rather than settling, which is consistent with a term that is roughly two years old.
Google addressed the category directly in May 2026, stating that AEO and GEO are not a separate discipline from SEO and that neither requires special markup or AI-specific rewriting to be recognized. We publish that finding rather than bury it. Most of what actually moves AI citation is ordinary structural and sourcing discipline, not a new toolkit.
What this page is testing
This FAQ page carries eight locked questions, this one among them, plus a ninth held back as a control, measured on the same schedule and never shown. If citation on the eight on-page questions moves after this page goes live and the ninth does not, GEO as we are practicing it here did something measurable. If both move together, the engines changed and this page gets no credit for it. That test was written down before this page existed.
What's genuinely not just SEO
Two things do not reduce to ranking a page. Entity resolution: an answer engine has to be confident which company it is looking at before it will name you, and an unsure engine skips you or describes somebody else rather than erroring. Measurement: only about 30% of brands stay visible across back-to-back runs of an identical query, and citation share can swing 40 to 60% month over month with nothing changed on either side. A single screenshot proves almost nothing. Repeated sampling on a fixed question set, reported per engine, is the only honest way to read this.
Final thoughts
We are not selling GEO as a replacement for search fundamentals. We are selling the discipline of measuring whether it worked, on a schedule, in public. This page is the first instrument we built entirely to test that on ourselves.
Regards,
Charles Stack & Sela, AI Optimization — Coworkers.Global
Cite this: "Generative engine optimization (GEO), coined by Aggarwal et al., Princeton, 2023, pulled approximately 1,000 Google searches per month as of September 2026; answer engine optimization pulled approximately 720." Search-volume figures via DataForSEO, captured September 14, 2026. Google Search Central, May 2026 (AEO/GEO guidance).
"FAQtopus" is a trademark of Coworkers Global, Inc.
Comments