The curriculum
Seven sections, in the order to read them
LiveUpcoming, not yet linked
Start with one tool, one real task and about fifteen minutes. This page walks you through seven steps, from understanding why AI makes things up to...
Open this section →AI search, explained
6 of 6 pages liveAI search means the search engine answers the question itself and cites a handful of sources, instead of handing you ten links. This section explains...
1.1What is AI SEO, and how is it different from SEO?1.2GEO vs AEO vs LLMO: do the labels mean anything?1.3What is Google AI Mode?1.4How AI Overviews and AI Mode changed search traffic1.5How do AI assistants decide what to cite?1.6Which AI platforms actually send traffic?Open this section →Getting found by AI
1 of 5 pages liveGetting found by AI means an AI system can reach your pages, read them without running JavaScript, and tell which of them answers a question. It is...
2.1Should you allow AI crawlers on your site?2.2What is llms.txt and do you need one?2.3Does JavaScript rendering block AI crawlers?2.4robots.txt, sitemap.xml and llms.txt: which does what2.5How to check whether AI can actually read your pageOpen this section →Writing content AI will cite
5 pages · October 2026Answer-first writing, information gain, and the page structure an AI actually lifts a quote from.
Structured data and entities
4 pages · October 2026Schema, entities, and the difference between being read by a machine and being understood by one.
Platform playbooks
4 pages · November 2026One playbook each for ChatGPT, Perplexity, Google AI Overviews and Copilot, reviewed every ninety days.
Measuring AI visibility
5 pages · November 2026Tracking mentions, citations and AI referral traffic, and proving whether any of it moved.
Reference
Look any term up
Grouped by subject rather than alphabetically, so related ideas sit together.
AI terms, explained
55 definitions so far. Each in one sentence, then in plain English, with an example, and no jargon used to define other jargon.
Open the glossary →About this course
How it is written, and what it costs
Why this exists
Most AI advice is either too shallow to act on or too technical to follow. This is an attempt at neither: one page per question, written in plain English, complete enough that you do not have to go looking elsewhere afterwards.
I have spent twelve years in search. The part that interests me now is what happens when the search box is replaced by an answer and nobody clicks through. Most of this course is about that shift, and about what a marketer can actually do about it.
How the pages work
Every page follows the same shape, so once you have read one you know how to read all of them:
- The answer, in the first sixty words. If that is all you needed, you can leave.
- Key points, so you can decide whether the rest is worth your time.
- The explanation, with a diagram where a diagram helps.
- Do it yourself, as numbered steps you can follow today.
- Common mistakes, and the part nobody else writes: when the advice does not apply.
- Questions, including the awkward ones.
Every page carries the date it was last checked. If something has changed since, the page is wrong until I fix it, and the date tells you how likely that is.
What it costs
Nothing. There is no sign-up, no email capture and no course to buy at the end. If you would rather have someone do this work than learn it, that is what my consulting is for. It gets one mention, at the bottom, rather than four per page.
Want this done rather than learned?
I run AI visibility audits and consult on AI search for brands and agencies. One conversation, no deck.
- Hub redesigned. Start here rewritten. Glossary grew to include AI Mode, the named crawlers, entities and schema.
- Learn AI opens with Getting found by AI.