The right tool turns a confusing, manual task into a two-minute check. This guide shows you how to get the most out of it.
In this guide you will learn the vocabulary of AI SEO and GEO. We will keep it practical, with clear steps, visual breakdowns, and specific actions you can take today. The first step in any AI visibility project is to run a free AI crawler check on your website so you know exactly where you stand against the 257 bots we track across 8 categories.
Key Takeaways
- GEO and SEO share most of their foundation, but the differences decide AI citations.
- Citability, authority, and crawl access are the three pillars that matter most.
- You can measure progress with an AI Visibility Score and AI referral tracking.
- Start with a free baseline using the free AI crawler check.
Why There Is No Simple Answer on the Vocabulary of AI Search
The vocabulary is unusually unstable, and not because it is new. Several of the most common terms were introduced by vendors describing their own products, so they carry a built-in point of view, and different vendors use the same word for different things. A conversation can proceed for some time with each side confident it understands the other, because the words match and the referents do not.
This has a direct practical cost. Most implementation errors here begin as a definition error: treating crawling and indexing as one action, or training and retrieval as one behaviour, produces configurations that do the wrong thing while looking deliberate. So the distinctions worth learning are the ones that change what you would do, and the rest is naming.
What to Settle About the Vocabulary of AI Search, in Order
Crawling, indexing and access are three separate things
Crawling is fetching, indexing is listing, and access is whether the fetch is permitted at all. robots.txt governs the first, meta directives govern the second, and firewalls govern the third, which is why a page can be disallowed and still appear in results, and why a Disallow provides no protection whatsoever. Collapsing these three is the single most expensive vocabulary error available.
Training and retrieval are different behaviours with different stakes
Training collection gathers text for future model updates and its effects cannot be reversed. Retrieval fetches a page to answer a question being asked now, and it stops when you stop it. The same operator often runs separate agents for each. One decision is close to permanent and the other is close to free, so treating them as one topic guarantees mispricing at least one of them.
GEO, AEO and AI SEO mostly name the same work
The proliferation of labels reflects positioning rather than substance. All of them describe making content findable, interpretable and quotable by systems that generate answers. Where the terms do differ is emphasis, so the useful move is to ask which mechanism someone means rather than which acronym they prefer.
Citation is not the same as a link, and neither guarantees a visit
An AI answer may name you, quote you, link you, or use you without any of the three. These outcomes have different value and only one of them registers in analytics. Any conversation about measuring AI visibility that does not first fix which of these is being counted will produce numbers nobody can reconcile.
The Costliest Way to Get Your Terminology Wrong
Assuming shared terms mean shared definitions
The costly version happens in a meeting where blocking AI is agreed and implemented, and nobody establishes whether that meant training collection, live retrieval or both. Everyone leaves aligned. The configuration blocks every AI agent, live citations stop along with training access, and the decline is attributed to anything except the vocabulary gap that caused it. Asking which mechanism, once, prevents it.
How to Know You Got Your Terminology Right
The check that matters here: Next time a term appears in a decision, ask what would change if it meant the other thing. If the answer is a different configuration, the definition was worth pinning down.
The 50 terms, defined
Grouped by topic so related ideas sit together. Each definition says what the term means in practice, and links to a deeper guide or profile where we have one.
| Term | Plain meaning |
|---|---|
| 1. Crawler | Software that fetches web pages automatically, usually on its own schedule. |
| 2. User agent | The name a client sends in the User-Agent header, for example GPTBot. |
| 3. robots.txt | A text file at your domain root that asks crawlers which paths to avoid. See the robots.txt validator. |
| 4. Disallow | A robots.txt rule that asks a crawler not to fetch a path. It is a request, not a lock. |
| 5. Crawl budget | How many pages a crawler is willing to fetch from your site in a given time. |
| 6. Training crawler | Collects pages that may be used to train future models, such as GPTBot. |
| 7. Search crawler | Builds an index that an AI search product answers from, such as OAI-SearchBot. |
| 8. User-triggered fetcher | Fetches one page because a person asked an assistant about it, such as ChatGPT-User. |
| 9. Personal AI agent | An assistant with its own cloud browser that does tasks for one person. See personal AI agents. |
| 10. Web Bot Auth | A way for bots and agents to sign requests (RFC 9421) so sites can verify who sent them. |
| 11. Verified bot | A bot that a CDN such as Cloudflare has checked and can identify reliably. |
| 12. Spoofed user agent | A client that copies a known bot name to look trusted. See fake AI crawlers. |
| Term | Plain meaning |
|---|---|
| 13. Indexing | Storing a page so it can appear in results. Separate from crawling. |
| 14. noindex | A meta tag or header that asks search engines not to list a page. |
| 15. X-Robots-Tag | An HTTP header that carries robots directives, useful for PDFs and other files. |
| 16. Canonical URL | The version of a page you want treated as the main one when duplicates exist. |
| 17. Rendering | Running a page's JavaScript to see the final content. Many AI crawlers do not render. |
| 18. Server-side rendering | Sending finished HTML from the server, so crawlers see content without running scripts. |
| 19. Soft 404 | A missing page that returns status 200, which wastes crawl and confuses indexes. |
| 20. Sitemap | An XML list of the URLs you want crawled, with optional last-modified dates. |
| 21. Crawl trap | A URL pattern that creates endless pages, for example filter or calendar parameters. |
| 22. Faceted navigation | Filters such as size or colour that create many URL combinations of one listing. |
| Term | Plain meaning |
|---|---|
| 23. GEO | Generative engine optimisation: making content easy for AI systems to find, understand and quote. See GEO for beginners. |
| 24. AEO | Answer engine optimisation. Mostly the same work as GEO, with a focus on direct answers. |
| 25. AI SEO | An umbrella name for SEO work aimed at AI search and assistants. |
| 26. AI Overview | Google's AI-written summary at the top of some search results. |
| 27. AI Mode | Google's conversational search mode that answers with AI and cites sources. |
| 28. Citation | When an AI answer names or links your page as a source. |
| 29. Mention | When an AI answer names your brand, with or without a link. |
| 30. Retrieval | Fetching documents at answer time so the model can ground its reply in them. |
| 31. RAG | Retrieval-augmented generation: a model answers using documents it just retrieved. |
| 32. Grounding | Tying an AI answer to specific sources so it can be checked. |
| 33. Zero-click search | A search that ends without a visit because the answer is on the results page. See zero-click survival. |
| 34. Query fan-out | When an AI system splits one question into several searches behind the scenes. |
| Term | Plain meaning |
|---|---|
| 35. Structured data | Machine-readable facts in JSON-LD, such as Product or FAQPage. Try the schema generator. |
| 36. Entity | A clearly defined thing, such as a company or product, that systems can recognise. |
| 37. Knowledge graph | A database of entities and how they relate, used by search engines and assistants. |
| 38. E-E-A-T | Experience, expertise, authoritativeness and trust: Google's quality framework. |
| 39. Topic cluster | A main page on a topic plus linked pages that cover its parts in depth. |
| 40. Pillar page | The main page in a topic cluster. |
| 41. Citability | How easy a passage is to quote: a clear claim, a source, and little ambiguity. |
| 42. Original data | Research or numbers only you publish, which gives others a reason to cite you. |
| 43. llms.txt | A proposed Markdown file that summarises a site for language models. See the llms.txt generator. |
| 44. Content signals | Machine-readable statements of how content may be used, for example search but not training. |
| Term | Plain meaning |
|---|---|
| 45. AI Visibility Score | Our 0 to 100 score: bot access 70, AI infrastructure 15, technical readiness 15. |
| 46. AI referral traffic | Visits that arrive from AI products such as ChatGPT or Perplexity, seen by referrer. |
| 47. Share of voice | How often your brand appears in AI answers for a set of questions, compared with rivals. |
| 48. Prompt set | The fixed list of questions you re-ask AI tools each month to track visibility. |
| 49. Crawl-to-refer ratio | How many pages a bot fetches for each visitor its product sends back. |
| 50. Server log analysis | Reading raw access logs to see which bots really visit. See tracking AI bots in logs. |
Five term pairs people confuse most
- Crawling vs indexing. A page can be blocked from crawling and still be indexed from links. Use noindex to keep it out of results, not Disallow.
- Training vs retrieval. Blocking a training crawler does not remove you from AI answers. Blocking a retrieval or search crawler does.
- Citation vs mention. A citation can send a visit. A mention builds awareness but leaves no trace in analytics.
- Crawler vs agent. robots.txt is for crawlers. Personal agents act for one person and are controlled with CDN or firewall rules.
- GEO vs SEO. They share most foundations. GEO adds quotable passages, clear entities and access for AI bots on top.
Put the vocabulary to work
- AI crawler checker: tests 257 bots against your robots.txt in one scan.
- AI crawler diagnostic: fetches a page as each AI bot and names the layer that blocks it.
- robots.txt validator: shows allowed, blocked and partial status per bot.
- robots.txt generator: builds a file with presets for search, agents and training.
- batch URL checker: compares up to 20 sites side by side, with CSV export.
- llms.txt generator: writes the file AI assistants read for a site summary.
- schema markup generator: produces JSON-LD for eight common types.
- GEO audit: scores a page for generative engine optimization.
Where to Go From Here
Terminology is one piece of a larger picture. The AI crawler directory documents every crawler we track with its operator, purpose and safety rating, and the batch URL checker audits many sites in one pass if you manage a portfolio.
Tools are only useful in sequence: generate, validate, then verify. Try the robots.txt builder and finish with an AI crawl checker.
Your Terminology Action Checklist
Five concrete steps, specific to what this guide covered. Work through them in order, changing one thing at a time so you can tell which change produced the result.
- □Establish a baseline with an AI crawler test and write down the score before you change anything.
- □Apply the single highest-impact change from this guide, on its own, so you can attribute the result.
- □Validate the change with the robot checker before it reaches production.
- □Re-measure and compare against your baseline rather than against expectation.
- □Schedule a recurring re-check, because redesigns and security updates quietly undo this work.
