Seoxpert.io
AI SEO · GEO

AI SEO: optimise for ChatGPT, Claude, Perplexity, and Google AI Overviews

AI SEO — also called GEO (Generative Engine Optimization) — is the practice of structuring a website so AI search engines can find, read, and cite it. The technical surface is narrow and testable: five fixes, none requiring backlinks, content rewrites, or a separate strategy. Website regression monitor for founders, agencies, and developers — SEO, security, performance and compliance checks after every deploy.

Run the audit below — it returns a GEO citability verdict alongside the regular SEO findings. Free first scan, no credit card.

Free tier: 4 full audits / month on a single domain.

The five technical conditions

What AI engines check before they cite

/llms.txt presence

The emerging convention for "robots.txt but for LLMs" — a plain-text file at the root of your site that curates which content AI engines should ingest. Format: # H1 with the site name, > blockquote with positioning, ## sections with curated URL lists. Optional today, increasingly used as a quality signal.

robots.txt rules for AI crawlers

GPTBot, ChatGPT-User, OAI-SearchBot, ClaudeBot, PerplexityBot, Google-Extended, Applebot-Extended. Blanket disallows opt sites out of citation. The fix is targeted Allow rules for runtime crawlers paired with selective blocks for training scrapers.

Organization JSON-LD with sameAs links

Tells AI engines which entity you are. Without it, the engine guesses from page text and can confuse you with similarly-named competitors. The sameAs array (LinkedIn, Twitter, Crunchbase, GitHub) is critical — it becomes the canonical identity link that disambiguates brand mentions.

Answer-first paragraphs on question-titled pages

Pages titled "What is X?" should answer the question in 1–3 short sentences (8–35 words) at the very top, before any preamble. AI engines extract citation snippets from the head of the page; pages that bury the answer under sales copy lose the slot.

Canonical landing pages for the topics you cover

Mentioning a topic across blog posts isn't the same as having a page that ranks for it. AI engines cite landing pages, not paragraphs. The Seoxpert content-gap agent uses an LLM to surface topics covered across the site that lack a canonical page, with cited evidence URLs.

AI SEO vs traditional SEO

Same fundamentals, different decisive signals

AI SEO and traditional SEO share the technical floor — indexability, crawlability, structured data, page speed, security. A site that fails the basics fails for both. Above the floor, the decisive signals diverge:

  • Freshness weights differently. AI engines fetch live at answer time, so a page updated yesterday can be cited tomorrow without waiting for Google to recompute. Traditional SEO is a moving average measured over weeks.
  • Entity disambiguation beats backlink graph. Organization schema with a populated sameAs array tells AI engines which brand you are. Backlinks help traditional rankings; for AI citations, identity links to LinkedIn / Crunchbase / GitHub are higher leverage.
  • Answer-first content beats keyword density. A page titled “What is X?” wins the citation slot if its first sentence is “X is …”, regardless of where the keyword density peaks. Traditional SEO is more forgiving of long preambles.
  • Citation crawler access is binary. If your robots.txt blocks ChatGPT-User, you are not cited. There's no partial visibility — you're in or out. Traditional crawl access has redirects, rendered fallbacks, mobile-first nuance.
FAQ

Common questions about AI SEO and GEO

What is AI SEO?

AI SEO (also called GEO — Generative Engine Optimization) is the practice of structuring a website so that AI search engines like ChatGPT search, Claude, Perplexity, and Google AI Overviews can find, read, and cite it. The technical conditions overlap with classical SEO but include a few unique signals that no traditional audit tool checks for: /llms.txt presence, AI-bot rules in robots.txt, Organization JSON-LD entity schema, and answer-first paragraph structure on question-titled pages.

Is AI SEO different from traditional SEO?

They overlap heavily on technical fundamentals — indexability, crawlability, structured data, page speed, security — but diverge on three things. (1) AI engines fetch pages live at answer time, so freshness signals weight differently. (2) Citation eligibility depends on entity disambiguation (Organization schema, sameAs identity links) more than backlink graph. (3) Answer-first content structure beats keyword density: a page titled "What is X?" wins the citation slot if its first sentence is "X is …", regardless of how dense the keyword is later in the page.

What is GEO (Generative Engine Optimization)?

GEO is the term that's settling as the industry name for AI SEO. It refers specifically to the optimisation work that makes a site eligible for citation in generative-AI answers — ChatGPT search, Claude, Perplexity, Google AI Overviews, Microsoft Copilot. The technical surface is narrow and testable: five fixes, none requiring a backlink campaign. The Seoxpert scanner runs the four hard checks (llms.txt, robots.txt AI bots, Organization schema, answer-first) and an LLM-powered content-gap pass that surfaces missing landing pages.

Do AI search engines actually cite websites?

Yes, and they cite them with attribution links. ChatGPT search, Perplexity, Claude (in research mode), and Google AI Overviews all return URL citations next to the generated answer. Each engine fetches pages live via a runtime crawler — ChatGPT-User and OAI-SearchBot for ChatGPT, PerplexityBot and Perplexity-User for Perplexity, ClaudeBot for Claude, Google-Extended on top of Googlebot for AI Overviews. The technical conditions for citation are testable.

How do I check if my site is AI-search-ready?

Run the Seoxpert audit on any URL — it returns a "GEO citability" verdict alongside the regular SEO findings. Specifically it flags whether /llms.txt exists, which AI bots your robots.txt is currently blocking (often unintentionally), whether the homepage has Organization JSON-LD with sameAs identity links, and which question-titled pages bury the answer under preamble. Free first scan, no credit card.

Do I need to remove "no AI training" rules from my robots.txt?

It depends on your goal. If you want to be cited but not used as training data, you need targeted rules: Disallow GPTBot (training scraper) but Allow ChatGPT-User and OAI-SearchBot (runtime fetchers). Same pattern for Claude (block ClaudeBot, allow whatever Anthropic ships for citation) and Perplexity (block CCBot, allow PerplexityBot). A blanket Disallow for all AI bots opts the site out of citation entirely.

How long until AI engines start citing my site after I fix these?

AI engines re-fetch on each query, so improvements compound from your next deploy — there's no waiting for Google to recompute rankings. The day you ship Organization schema and an answer-first paragraph, the next ChatGPT-Search query for "[your topic]" can lift your URL. Practically, expect a 1-2 week observation window before traffic measurably moves, because the question is "do enough users ask the right query" rather than "did the crawler re-index".

Run the AI / GEO citability check

Free scan · No credit card · Results in under 2 minutes.