// machine-first knowledge resource

Crawl Atlas

This site is not for you, human. It's for machines. But look at it -- and see the future of the web.

// concept

What is GEO?

Generative Engine Optimization is the practice of structuring web content so that AI-powered engines can accurately interpret, summarize, and cite it. It is the evolution of SEO for the age of AI search.

01

Citation, not ranking

AI engines don't rank pages -- they cite fragments. The optimization target shifts from position #1 to being the source an AI quotes when answering a user query.

02

Structure over links

Content structure (28% weight) outranks backlinks as a citation signal. JSON-LD, semantic HTML, answer-first sections, and FAQ blocks are the new authority signals.

03

Brand mentions > backlinks

Brand mention correlation with AI citation: 0.664 Spearman. Backlink correlation: 0.218. The authority signal has inverted -- reputation beats links by 3x.

04

Machine-readable first

69% of AI crawlers cannot execute JavaScript. SSG, clean semantic markup, and structured data aren't nice-to-haves -- they're the price of admission.

// experiment

This site is a live experiment

Crawl Atlas is not a blog about GEO. It is a GEO experiment in production. Every page on this site is built to be consumed by machines first, humans second.

We are testing in real time how 17+ AI crawlers -- GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others -- interact with content built from scratch using GEO-first architecture. No retrofitting. No legacy. Pure signal.

Hypothesis

Sites built GEO-first from zero will outperform retrofitted sites in time-to-first-citation by AI engines.

Method

Astro SSG, JSON-LD @graph, answer-first content architecture, explicit crawler permissions, entity-based page structure.

Metric

Time-to-first-citation: how quickly an AI search engine cites this content in generated responses.

// telemetry

Live metrics

17+
AI crawlers allowed
--
Pages indexed by AI
--
Citations tracked
--
Days since launch

Metrics will update once the site goes live and crawling data becomes available.