GEO: Generative Engine Optimization Guide

GEO (generative engine optimization) is the practice of making content and brands eligible to be retrieved, cited and recommended by AI answer engines such as Google AI Overviews, Google AI Mode, ChatGPT and Perplexity. This page is the hub for the subject on this site: it links every related concept, comparison and industry page, grouped by the question each one answers.

The evidence was reset in 2026. A September 2026 re-test of the on-page levers from the original 2023 GEO paper found that they "move citation on none" of ten modern engine families (Bajemon and Rochet, vendor authors), and a July 2026 survey of 45 GEO studies concluded that "no reviewed technique shows a stable, longitudinal, cross-platform causal effect on organic discoverability" (Martinez). What the 2026 data does support is narrower: only 37.9% of URLs cited in AI Overviews rank in the organic top 10 (Ahrefs, March 2026), 85.7% of citations point to sites the brand does not own (Żatuchin, June 2026), and roughly 65% of cited sources change from one day to the next (Schulte, Bleeker and Kaufmann, April 2026).

Readers new to the subject should begin with the three pages under Start Here. Practitioners can go straight to the section that matches the decision in front of them: which surface matters, what gets cited, whether crawlers can reach the site, or how to measure any of it. Figures across these pages are dated because citation shares move quickly.

Start Here

Three pages give the definition, the landscape and the current evidence.

Generative Engine Optimization — the full definition: making content and brands eligible to be retrieved, cited and recommended by AI answer engines.
AI Search — discovery systems that use large language models to generate synthesized, cited answers instead of ranked lists of links.
State of AI Search Citations: Q4 2026 — a quarterly digest of the published studies, with every figure dated and attributed.

AI Search Surfaces

Where AI-generated answers appear, and what each surface does to clicks.

AI Overviews — Google's AI-generated summaries at the top of search results: when they appear and what they cite.
Google AI Mode — Google's conversational search surface, now the destination behind an AI Overview's "Show more".
ChatGPT Search — how ChatGPT retrieves and cites the live web, and the crawlers that feed it.
Perplexity — the answer engine built around visible citations, with a source mix that differs from Google's and ChatGPT's.
Query Fan-Out — how an engine splits one query into related sub-queries, and why cited pages often sit outside the organic top 10.
Zero-Click Search — searches answered on the results page itself, and the click-through evidence for AI answers.
AI Referral Traffic — the visits AI assistants do send to websites, and how to identify them.

How Citation Works

What determines which sources an engine names, and how much each one shapes the answer.

AI Citations — the source links attached to a generated answer; being cited and shaping the answer are separate outcomes.
AI Share of Voice — the proportion of AI answers for a prompt set that mention a brand, measured against competitors.
Earned Media — press, reviews, listicles and community threads a brand does not control, which supply most AI search citations.
Position Bias — a model's tendency to use the start and end of its context more reliably than the middle.
Training Data Frequency — how often a brand or fact appears in training text, shaping what a model recalls and recommends from memory.
Answer Engine Optimization (AEO) — the overlapping label for getting a brand or page into direct AI answers.

Access and Technical Hygiene

Whether crawlers and agents can reach and read a site at all. Much of this layer is now a publishing-platform feature: LightCMS, the CMS serving this site, automates llms.txt, Markdown copies of every page, per-purpose crawler controls and AI traffic reporting.

AI Crawlers — bots that fetch pages for training, AI search indexing or live user requests, each with its own robots.txt rules.
llms.txt — a curated Markdown map of a site for AI systems; the 2026 log data shows coding agents use it far more than AI search bots.
Schema Markup — schema.org structured data that powers rich results; a controlled 2026 study found no lift in AI citations.
Agent Experience (AX) — designing products, docs and APIs so AI agents can find, understand and operate them without human help.
AGENTS.md — an open Markdown file format that gives coding agents project-specific build steps, tests and conventions.
MCP Registry — the official metadata catalog of public Model Context Protocol servers.

Measurement

AI answers change daily, so visibility has to be sampled repeatedly and reported as a range.

AI Visibility Measurement — quantifying how often, and how accurately, AI answer engines mention, cite and send traffic to a brand.
LLM Optimizer — AI-search visibility measurement tool.

Platforms as Sources

The third-party platforms AI engines cite most, and what earns a citation on each.

YouTube — one of the sources AI answer engines cite most, led by long-form video on Google's AI surfaces.
Reddit — a leading cited domain across ChatGPT, Perplexity and AI Overviews, with a share that moves sharply.
LinkedIn — a growing share of social citations, mostly from individual posts and articles.
TikTok — a search destination for consumers that AI engines rarely cite.
Wikipedia — the top cited source in 11 of 12 languages in a June 2026 study of 128 brands.
Video GEO — making video, mostly on YouTube, more likely to be retrieved and cited by AI answer engines.
Short-Form Video — vertical feed video that AI engines rarely cite but that drives discovery and sales.

Local

How AI assistants recommend nearby businesses, and the listing data behind those answers.

Local AI Search — the use of AI assistants and AI-generated results to find, compare and choose nearby businesses.
Google Business Profile — Google's free listing that controls how a business appears in Search and Maps, and feeds Google's AI answers.

Coding Agents as a Channel

Software is increasingly chosen by an agent in a terminal, not a person reading a results page.

AI Coding Agents — agents that plan, write and run code, and increasingly decide which SDKs, APIs and services a project adopts.

Comparisons

Side-by-side pages for the terms and surfaces most often confused.

Generative Engine Optimization vs SEO — what the two disciplines share, where they diverge, and what is debunked.
Generative Engine Optimization vs Answer Engine Optimization — mostly a difference in terminology, with one useful nuance.
AI Overviews vs Google AI Mode — triggering, sources, volatility, traffic impact and publisher controls.
ChatGPT Search vs Perplexity — citations per answer, per-page influence, most-cited domains and crawler behaviour.
ChatGPT Search vs Google AI Mode — reach, eligibility, crawlers, source mix, controls and measurement.
llms.txt vs robots.txt — one tells crawlers what they may access; the other offers agents a curated guide.
llms.txt vs MCP — a static Markdown briefing an agent reads versus a live protocol an agent calls.
Earned Media vs Owned Media — why studies put the brand-owned share of citations at very different levels, and how to split budget.
Long-Form Video vs Short-Form Video — long-form earns nearly all AI citations of video; short-form wins feed reach and conversion.
AI Coding Agents vs AI Assistants — assistants recommend software to a human buyer; coding agents pick and install it themselves.

By Industry

The same evidence applied to the sources, buyers and regulatory limits of each vertical.

GEO for Gaming — how players and developers ask AI about games, and which sources the answers draw on.
GEO for Retail — how AI shopping answers choose sources, what product feeds do, and where agentic checkout stands.
GEO for Healthcare — which sources AI engines cite for health questions, the accuracy risks, and what providers can do.
GEO for Financial Services — low overlap with organic rankings, comparison-site dominance and compliance limits.
GEO for Legal — how clients use AI to research lawyers, and the ethics limits on what firms can do about it.
GEO for Real Estate — listing portals inside ChatGPT, how buyers use AI, and what agents and brokerages can control.
GEO for Travel & Hospitality — how hotels, destinations and travel brands get named in AI answers.
GEO for Advertising & Marketing — what agencies should sell, and stop selling, as GEO services.
GEO for Publishing — an access and licensing decision: traffic loss, crawler blocking, opt-outs and who gets cited.
GEO for Education — how colleges and course providers appear in AI answers, and what surveys show about student AI use.
GEO for Developer Tools — what AI cites at the evaluation stage, how coding agents pick dependencies, and docs agents can use.
GEO for Local Businesses — third-party lists and reviews, service pages, consistent facts, and the limits of the evidence.
AI Search for Developer Tools — how developers and software buyers research tools with AI assistants and coding agents.
AI Search for Local Businesses — how customers use ChatGPT, Google AI Mode and AI Overviews to find local businesses, and how they verify the answers.
Short-Form Video for Local Businesses — rarely a source of AI citations, but a driver of discovery, proof and conversion.