The AI SEO Masterclass

1. SEO, AEO, GEO & LLMO — In One Breath

SEO, AEO, GEO, and LLMO are not four separate disciplines — they are the same evolving idea, described at different points in time. SEO earned you a spot on Google's page one. AEO, GEO, and LLMO are what's layered on top of that same foundation today, to earn you a place inside the answer AI gives, instead of just a link below it. SEO has not faded — it is the floor everything else stands on. Here is the one-line version of each term:

2. SEO Is Still the Foundation

Google published its own official AI-search guidance, and its message directly confirms what good marketers already suspected. In Google's own words: "The best practices for SEO remain relevant for AI features in Google Search (such as AI Overviews and AI Mode)." There is no separate algorithm you need to chase — Google's generative features are built directly on top of its core Search ranking and quality systems.

The data backs this up hard: around 76% of AI Overview citations are pulled from pages already ranking in Google's top 10 organic results. If your site is not winning classic SEO, it is statistically very unlikely to be cited by AI at all. SEO is not competing with AI search — it is the entry ticket to it.

But here is the honest, nuanced part a real masterclass has to include: SEO alone no longer guarantees the citation. The correlation between organic rank and AI citation has weakened sharply over the last two years — from a strong relationship pre-2024, down to a much weaker one today. Translation: ranking well gets you in the room, but a second, different layer of optimisation decides who actually gets quoted once you are there. That second layer is everything from Chapter 3 onward.

3. How AI Actually Answers a Question

Every optimisation tactic in this guide only makes sense once you understand the mechanics behind it. An AI answer is not one step — it is five. Tap through each stage to see exactly what happens between someone typing a question and an AI naming you as the source:

4. The Platform Playbook

Not every AI engine works the same way, and treating them as one blob is a mistake. Each platform leans on different sources and rewards different things. Tap through Google, ChatGPT, Perplexity, Claude, and the rest to see exactly how each one decides what to cite:

5. The SEO Foundations That Still Matter

None of the above works without the classic SEO layer underneath it. Click through the three pillars below — the same on-page, off-page, and technical foundations that have mattered for years, and still do:

Search Engine Optimization

Free Customer Acquisition System

1. On-Page SEO

2. Off-Page SEO

3. AI Search (GEO)

1. On-Page SEO

Make your own website easy to read

Core Objective

Organize the words, pictures, and links on your website so search engines can understand what you sell.

Key Action Items

Clear page titlesHelpful headingsFast loading photosSimple page links

Simple Example

Write simple headings and descriptive text. For example, if you sell cakes in Delhi, your page title should say "Fresh Chocolate Cakes in Delhi" instead of just "Home". This helps Google show your site to local buyers.

6. The Technical Overlap: What Still Decides AI Visibility

The single most important technical fact in this entire guide: most AI crawlers do not render JavaScript. GPTBot fetches JavaScript files in only about 11.5% of requests and never executes them; ClaudeBot fetches JS around 24% of the time and also never executes it. Googlebot is the one major exception that fully renders JS. If your content only appears after client-side rendering, most AI crawlers see an empty shell — full stop. Beyond that, here is the classic SEO checklist, with why each item specifically matters for AI visibility today:

• Server-rendered content: the make-or-break item above. Use server-side rendering, static rendering, or incremental regeneration so your main content is in the initial HTML response — not something that only appears after JavaScript runs.
• Page speed & Core Web Vitals: still a real ranking factor, and AI crawlers do not wait around or retry a slow page — they fetch once and move on.
• Mobile-friendliness: around 71% of AI chatbot sessions happen on mobile, so a page that breaks on a phone is failing most of your real audience.
• Crawlability & indexability: no orphan pages, a clean and current XML sitemap, and no accidental "noindex" tags left over from staging.
• Clean heading hierarchy (H1 → H2 → H3): this helps AI "chunk" your page into distinct, quotable pieces, the same way it helps a human skim.
• Semantic HTML — lists, tables, definitions: genuinely easier for AI to extract cleanly than one long, dense paragraph saying the same thing.
• HTTPS, clean URLs, and internal linking: unchanged baseline trust and structure signals that both classic crawlers and AI retrieval systems still rely on.

7. robots.txt & llms.txt: Don't Accidentally Lock AI Out

This is the single most common, most avoidable mistake in AI SEO: a default "block all bots" setting — often left over from a WordPress plugin, a Wix default, or an agency's staging config — silently blocks your AI visibility along with everything else. Most site owners never check.

robots.txt lives at your domain root — https://www.example.com/robots.txt — and tells crawlers which parts of your site they may visit. Here is a permissive sample that explicitly allows the major AI crawlers:

User-agent: * Allow: / User-agent: Googlebot Allow: / User-agent: Google-Extended Allow: / User-agent: GPTBot Allow: / User-agent: OAI-SearchBot Allow: / User-agent: ClaudeBot Allow: / User-agent: PerplexityBot Allow: / User-agent: CCBot Allow: / Sitemap: https://www.example.com/sitemap.xml
llms.txt is a newer, emerging convention — a plain Markdown file, also placed at your domain root (https://www.example.com/llms.txt), that gives AI tools a short, structured summary of your site and its key pages:

# Your Business Name > One-line description of what you do and who you help. ## Key Pages - [Blog](/blog): In-depth guides on your core topic - [Case Studies](/case-studies): Real results and proof - [About](/about): Founder background and credentials
Be honest with yourself about llms.txt: Google has explicitly said it does not use it for Search or AI Overviews — in Google's own words, it "will neither harm nor help." It is a low-cost, emerging convention some AI tools may reference as they mature, not a guaranteed lever. Worth ten minutes to add, not a strategy to build around.

8. Schema Markup & JSON-LD: What Actually Matters Now

Structured data is one of the most debated topics in AI SEO right now — some studies show a big citation lift, others show almost none. The honest, current picture, schema type by schema type, is more useful than picking a side. Tap through each one:

9. Content Tactics That Win Citations

Once the technical layer is solid, this is where citations are actually won or lost. Six proven tactics, click through each one to see exactly how it works in practice:

Generative Engine Optimization

AI Search & Recommendation Architecture

Direct Answering

Expert Citations

Hard Stats

Schema Markup

Brand Association

Info Density

Direct Answering

Answer questions directly

Core Objective

AI engines look for short, direct answers to show in their search result boxes.

Key Action Items

Simple answersClear definitionsParagraph headersDirect language

Simple Example

Write answers in a clear 1-2 sentence block at the top of your page. For example, write: "Generative Engine Optimization (GEO) is the process of styling your website content so that AI search engines can easily find, read, and cite it in their answers."

10. Keyword Research: SEO vs AEO

Traditional SEO keyword research is about search volume — finding phrases enough people type into Google every month to justify writing about them. AEO research flips the question: instead of chasing volume, you are hunting for the exact questions people ask, in natural, conversational language, because that is what an AI engine has to match against when it decides what to retrieve.

A simple four-step process:

1. Start with what you already rank for. Open Search Console and filter your existing queries for question words — what, how, why, can, does, is. These are proof that people already come to you asking these exact things.
2. Map the follow-up questions. A tool like AlsoAsked visualises Google's "People Also Ask" data as a tree, showing what someone asks next after your seed question. AnswerThePublic does something similar, organised by question type.
3. Ask the AI itself. A genuinely useful trick: ask ChatGPT or Perplexity your seed question, then ask it "what follow-up questions would someone ask next?" and "what does a complete answer need to cover?" You are literally asking the model to show you its own retrieval logic.
4. Build one page per real question, not one page per keyword. A page that answers a specific, real question in genuine depth outperforms ten thin pages built around keyword variations of the same idea.

11. The Most Powerful Tactic: Answer Every Real Question

There is one tactic in this whole guide that beats every paid tool: answer the real questions your actual audience already asks you, in your own words. Three places to mine them for free:

• Search Console queries. Filter for question-phrased searches that already bring people to your site, especially ones where you rank on page 2 or 3 — that is real, existing demand you have not fully answered yet.
• YouTube and Instagram comments. The questions people leave under your videos and posts are, word for word, the language your real audience actually uses — unfiltered by any keyword tool.
• Your sales or support team. The questions a support inbox or a sales call gets asked most often are the highest-intent questions that exist, because they come from people already close to buying. If five people asked your sales team the same question this month, that is a piece of content, waiting to be written.

Turn each of these into its own clearly-titled page or section, phrased exactly the way the question was actually asked. This is the single highest-leverage, lowest-cost research method in this entire masterclass.

12. Off-Site AI Visibility

Here is something most people miss entirely: for AI visibility, what happens off your website often matters more than what you change on it. Reddit, Quora, YouTube, LinkedIn, and Instagram all feed the same pool AI engines pull citations from. Tap through each platform to see exactly how much it matters, and to which engine:

13. SEO Metrics vs AEO Metrics

You cannot manage what you do not measure — and AEO needs a different scorecard from classic SEO. Here is the paired comparison:

14. Your Toolbox

A short, real list of tools worth bookmarking for this work — no fluff, no obscure names, just the ones that actually help:

15. Myths & Mistakes

A handful of traps that quietly waste people's AI SEO effort:

• "Add schema and citations will follow." The evidence is genuinely mixed — some studies show a real lift, a rigorous causal study by Ahrefs found citations barely moved. Add schema for AI-readability and correctness, not as a guaranteed citation lever.
• "llms.txt will fix my AI visibility." Google has explicitly said it does not use it. Add it, it costs ten minutes — but do not build a strategy around it.
• Keyword-stuffing for AI. AI models understand concepts and context, not repeated phrases. Write naturally, for the person asking the question.
• Faking statistics or outdated data. AI systems increasingly cross-check facts across multiple sources. Wrong or stale numbers get you quietly excluded from future citations, not just called out once.
• Ignoring classic SEO because "AI changed everything." It did not. 76% of AI Overview citations still come from pages already ranking in the top 10. Skipping SEO to chase AI tactics skips the foundation the tactics stand on.
• Treating every AI engine the same. A tactic that wins on Perplexity (original data, Reddit presence) is not the same tactic that wins on ChatGPT (Wikipedia presence, Bing indexing). Know which platform you are optimising for.

16. The Complete Action Checklist

Everything in this masterclass, condensed into one list you can actually work through: