⌘K
DARK
MENU
DARK

Last reviewed

Learn AI Search Optimisation (AEO and GEO): From Zero to Expert

AI search optimisation, also called AEO or GEO, is the work of getting a brand found, cited and recommended in AI answers from ChatGPT, Google AI Overviews and AI Mode, Perplexity, Gemini, Claude and Copilot. This learning path takes you from how AI answers are built to running a full AI visibility programme.

  • 182resources
  • 4stages
  • 6 to 9 monthsto job-ready
Start the roadmapSee the tools
Time to job-ready6 to 9 months, or 2 to 4 with SEO experience
DifficultyMedium: the field changes monthly
First job titleSEO Specialist with an AI search focus
You needSEO basics and accounts on four AI assistants
Learn firstHow AI answers find their sources
Cost to learnFree; tracking tools are mostly paid

What is AI search optimisation (AEO and GEO)?

AI search optimisation is the practice of making a brand visible inside AI-generated answers. You will see it called AEO (answer engine optimisation), GEO (generative engine optimisation), LLMO or AI SEO. The names differ; the goal is the same: when someone asks an AI assistant a question your business can answer, your brand is mentioned, your pages are cited and what the answer says about you is accurate.

AI assistants reach a brand by two routes. The first is retrieval: when a question needs current information, they search the web or their own index, pull passages and cite them. The second is model knowledge: what the model learnt about your brand during training, which only updates when a new model is released. Good AI search work covers both, and it sits on top of SEO rather than replacing it, because most retrieval still depends on pages being crawlable, indexed and trusted.

  1. 1AskSomeone asks a conversational question, often with context about their situation.
  2. 2Fan outThe system splits it into several related searches it runs in the background.
  3. 3RetrieveIt pulls passages from its index, from Google or Bing, or from its own crawler.
  4. 4ChooseIt favours relevant, trusted, clearly written passages, often from several sites.
  5. 5Answer and citeIt writes one answer, names brands and links some of the sources it used.

What does an AI search optimisation specialist do?

What a ai search specialist doesYour brand and content goes in. Track AI visibility, Map query fan-out, Make content answer-ready, Earn citations and mentions, Keep crawlers and entities clean, Measure AI traffic and impact. Cited and recommended in AI answers comes out, and results feed back into the next round.DATA AND RESULTS FEED THE NEXT ROUNDYOUR BRAND ANDCONTENTCITED ANDRECOMMENDED INAI ANSWERSAI SEARCH1Track AIvisibility2Map queryfan-out3Make contentanswer-ready4Earn citationsand mentions5Keep crawlersand entitiesclean6Measure AItraffic andimpact
The work flows clockwise from 1 to 6. Each step is explained below.
  1. 1Track AI visibilityRun fixed prompt sets across ChatGPT, Perplexity, Gemini, Claude, Copilot and Google AI Mode, and record mentions, citations and accuracy.
  2. 2Map query fan-outTurn each target question into the sub-questions AI systems search for, then find the gaps in your content.
  3. 3Make content answer-readyRestructure pages into self-contained, answer-first sections with definitions, data and comparison tables.
  4. 4Earn citations and mentionsGet the brand into the sources AI answers lean on: reviews, lists, Reddit, YouTube, media and industry sites.
  5. 5Keep crawlers and entities cleanMake sure AI crawlers can reach the site and that schema and profiles describe the brand the same way everywhere.
  6. 6Measure AI traffic and impactSeparate AI referrals in analytics, watch AI Overview performance and report what changed for the business.

How to learn AI search optimisation: a step-by-step roadmap

Work through the stages in order. If you already know SEO, move quickly through stage 1 but don’t skip it: the way AI answers are assembled changes what good optimisation looks like. The hours are a rough guide. Use the filter to show only videos, courses, templates or free resources.

Learning roadmap for AI Search Optimisation (AEO and GEO)4 stages and 182 resources from beginner to job-ready.YOUR ROADMAPAI Search Optimisation (AEOand GEO)4 stages, 182 resources, one project per stageBEGINNER10 TO 20Basics20 to 30 hours220 TO 50Core skills40 to 60 hours350 TO 80Practice80 to 120 hours480 TO 100MasteryOngoingJob-ready: SEOSpecialist with anAI search focus
Your route through this learning path. Select a stage to jump to it.
Show
1

0 to 20About 20 to 30 hours

Stage 1: How AI search works

Understand how large language models produce answers, how they find and cite sources, and how each AI platform differs.

How large language models work

You can't optimise for a system you don't understand. Get the basics of how models generate text before anything else.

Retrieval: how AI answers find sources

Most AI search answers use retrieval-augmented generation (RAG): search first, then write. This is where optimisation happens.

The AI search landscape

Each platform finds and shows sources differently. Learn how Google, OpenAI, Anthropic, Microsoft and Perplexity describe their own products.

Use the assistants yourself

Reading isn't enough. Ask the same questions in several assistants every week and notice which sources they choose.

Check yourself: you are ready for the next stage when you can

Project

Your first AI answer audit

Pick a business you know. Write ten questions its customers would ask, and run each one in four AI assistants.

Deliver
A filled-in AI Visibility Audit Template, plus one page on which brands and sources appear and why.
Done when
You can explain why each cited source was likely chosen.
2

20 to 50About 40 to 60 hours

Stage 2: Core AEO and GEO skills

Learn what makes content retrievable and citable, map query fan-out, and make a brand's entity clear to machines.

The research behind GEO

Start with the evidence. The original GEO paper tested which content changes improved visibility in generative answers.

Answer-ready content

AI systems lift passages, not pages. Content that answers first, defines clearly and backs claims with data is easier to retrieve and cite.

Query fan-out

AI Mode and other assistants break one question into many searches. If you only target the original question, you miss most of the retrieval.

Entities and structured data

AI systems need to know exactly who a brand is. Consistent entities, schema and profiles reduce confusion and wrong answers.

A structured course to tie it together

If you prefer one guided course, do one of these end to end alongside the topics above.

Check yourself: you are ready for the next stage when you can

Project

An answer-ready rewrite

Choose one important page on a real site. Map its fan-out queries, then rewrite it to answer them clearly with data and structure.

Deliver
The before and after versions, your fan-out map, and prompt results before and four weeks after.
Done when
You can show what changed in the page and in the AI answers, including if nothing changed.
3

50 to 80About 80 to 120 hours, mostly hands-on

Stage 3: AI visibility in practice

Run a measurable AI visibility programme for a real brand: tracking, content, citations, crawler access and reporting.

Measure AI visibility

Build a fixed prompt set, run it on a schedule and record mentions, citations, position and accuracy.

Track AI traffic in analytics

AI visits are easy to lose in Referral or Direct. Set up tracking before you need to report on it.

Earn citations and mentions

AI answers lean on what independent sources say about you. This is closer to digital PR than to on-page SEO.

Platform by platform

Google's AI features and ChatGPT search behave differently. Learn each platform's quirks.

AI crawler access

If AI crawlers are blocked by robots.txt or a firewall, nothing else matters. Check every bot, not just Googlebot.

Check yourself: you are ready for the next stage when you can

Project

A 60-day AI visibility programme

Pick a real brand (your employer's with permission, a client or a friend's business). Baseline its AI visibility, then improve content, crawler access and third-party presence for 60 days.

Deliver
Baseline and day-60 prompt results, GA4 AI traffic, the changes you made and what you would do next.
Done when
You can show a measured change, or explain clearly why there wasn't one.
4

80 to 100Ongoing, 6 months or more

Stage 4: Advanced AI search strategy

Understand the business impact of AI answers, set strategy across brand, PR and product data, and keep up with a field that changes monthly.

Impact on clicks and business

Leaders will ask what AI answers do to traffic. Know the main studies and their limits.

Strategy for brands and publishers

AI answers change the economics of content. Learn the frameworks and the debates.

AI agents and AI commerce

Assistants are starting to act for people: comparing products and completing tasks. Get ahead of it.

Advanced tracking platforms

At scale, manual prompt checks stop working. Learn what the main platforms measure and how they differ.

Keep up with a fast-moving field

Platforms change monthly. Pick two newsletters, one video channel and one community.

Check yourself: you are ready for the next stage when you can

Project

Publish original AI search research

Design a small study: for example, which brands AI assistants recommend in one industry and country, and which sources they cite.

Deliver
A published write-up with your method, data, limits and conclusions.
Done when
Someone else could repeat your study, and at least one other site references it.

AI search optimisation tools to learn

Learn firstGoogle Search ConsoleIndexing and performance in Google, including the pages AI Overviews and AI Mode can draw on.FreeLearn firstBing Webmaster ToolsBing indexing, which feeds Copilot and other assistants; IndexNow for fast updates.FreeGoogle AI ModeTest conversational queries and see which sources Google cites.FreeGeminiCheck what Google's assistant says about a brand and its competitors.Free planQforiaSimulates query fan-out to show the sub-questions behind a prompt.FreeHubSpot AI Search GraderFree one-off check of how AI assistants describe a brand.FreeAhrefs Brand RadarBrand and competitor mentions across AI answers at scale.PaidSemrush AI Brand PerformanceHow AI platforms talk about a brand, with share of voice.PaidProfoundEnterprise AI visibility and citation tracking.PaidPeec AIPrompt tracking and source analysis for marketing teams.PaidMoz AI VisibilityAffordable monitoring of ChatGPT, Perplexity and AI Overviews.PaidScrunchAI visibility plus how AI agents experience your site.Paid

Is AI search optimisation a good career?

It is one of the newest specialisms in marketing, so job titles are still settling and there is little long-term data. What is clear is that more searches now end in an AI answer, and companies want someone who understands how those answers are built. People who combine solid SEO with AI search skills are in a strong position.

Search is spreading across platforms

People now ask Google, ChatGPT, Perplexity and others. Brands need someone who understands how each one finds and cites sources.

Few people have real experience yet

Most practitioners started in the last two years. Hands-on projects and published results stand out quickly.

It builds on SEO

Crawling, indexing, content and links still matter. If you already know SEO, you are most of the way there.

It reaches beyond marketing

AI visibility touches PR, brand, product data and customer support content, which gives the role wide influence.

AEO and GEO career path and jobs

  1. Entry

    SEO Specialist with an AI search focus

    Runs prompt tracking, audits content for answer-readiness and reports AI referral traffic.

  2. Mid

    AI Search or GEO Specialist

    Owns the AI visibility programme: fan-out mapping, content changes, citations and measurement.

  3. Senior

    AEO or AI Visibility Manager

    Sets the AI search strategy across SEO, content, PR and product data, and reports to leadership.

  4. Specialist

    Content Strategist for AI search

    Plans and writes content built to be retrieved and cited, often with original data.

  5. Leadership

    Head of Organic and AI Search

    Owns organic growth across classic search and AI answers, with a budget and a team.

  6. Independent

    AI Search Consultant

    Audits brands' AI visibility and runs improvement programmes for several clients.

Common AI search optimisation mistakes

MistakeJudging visibility from one screenshot.

Do thisRun each prompt several times, logged out, and report rates over weeks.

MistakeTreating AEO as separate from SEO.

Do thisFix crawling, indexing and content quality first; AI search builds on them.

MistakeOnly publishing on your own site.

Do thisEarn mentions in the independent sources AI answers cite: reviews, lists, media and communities.

MistakeChasing tricks like hidden text or AI instructions in page code.

Do thisWrite clear, useful, well-sourced content. Tricks are spam and rarely last.

MistakeBlocking AI crawlers by accident.

Do thisCheck robots.txt, CDN and firewall settings for every AI crawler you want.

MistakePromising rankings in ChatGPT.

Do thisPromise a process and honest measurement. Nobody controls what an AI answer says.

How to get an AI search or GEO job

Few people have long AI search track records, so hiring managers look for evidence: prompt data, before-and-after results and clear thinking about what is and isn’t proven. A strong SEO foundation plus two or three published AI visibility projects is a competitive profile.

Job-ready checklist

  • You can explain how AI answers retrieve and cite sources
  • You can run a prompt-tracking study and report its limits
  • You can make a page answer-ready and map its fan-out queries
  • You can set up AI referral tracking in GA4
  • You can audit AI crawler access and entity consistency
  • You have at least one project with before-and-after AI visibility data

Portfolio pieces that get interviews

  1. An AI visibility audit of a real brand, with your prompt set and findings
  2. An answer-ready rewrite with fan-out map and before-and-after results
  3. A GA4 AI traffic report or Looker Studio dashboard you built
  4. A small original study of AI answers in one industry or country

AEO and GEO interview questions

How would you measure a brand's AI search visibility?

A strong answer covers: A fixed prompt set based on real customer questions, run on several assistants on a schedule, logged out, from the right country. Record mentions, citations, position and accuracy, report rates over time, and add AI referral traffic from analytics. Mention the limits of sampling.

What's the difference between SEO and GEO?

A strong answer covers: GEO builds on SEO: crawling, indexing and quality still matter. The differences are passage-level retrieval, query fan-out, the weight of third-party mentions, entity clarity and measuring mentions and accuracy rather than only rankings and clicks.

A competitor appears in ChatGPT answers and we don't. What do you do?

A strong answer covers: Check which sources ChatGPT cites for those prompts, compare our presence in them, check crawler access and Bing indexing, then plan content and citation work. Set expectations that results take weeks to months and vary by run.

Should we block AI crawlers?

A strong answer covers: Separate training crawlers from search and user-fetch crawlers. Blocking search crawlers removes you from those answers. Blocking training is a business choice that depends on whether the content is the product. Explain the trade-off rather than give a blanket answer.

How do you know which AI search advice to trust?

A strong answer covers: Prefer platform documentation, published research and studies with clear methods. Treat single-account anecdotes and tool vendor claims carefully, and test ideas on your own prompts before rolling them out.

AI search optimisation FAQs

What's the difference between AEO, GEO and SEO?

SEO is optimising for search engines in general. AEO (answer engine optimisation) and GEO (generative engine optimisation) both mean optimising to be mentioned and cited in AI-generated answers. They overlap heavily: AEO and GEO build on SEO foundations and add work on passages, entities, citations and AI-specific measurement.

Is GEO replacing SEO?

No. Google's AI features draw on its search index, and other assistants rely on search indexes and their own crawlers, so pages still need to be crawlable, indexed and trusted. GEO adds to SEO; it doesn't replace it. Most AI search specialists started as SEOs.

How do I get my brand mentioned in ChatGPT?

Make sure OpenAI's search crawler can reach your site and that you're indexed in Bing. Publish clear, answer-first content with original data, and earn mentions in the independent sources ChatGPT cites for your topic, such as reviews, lists and communities. Then measure over weeks, not days.

Does llms.txt help AI search visibility?

There's no evidence it does. llms.txt is a proposal, and no major AI search provider has confirmed using it for ranking or citations; Google has said it doesn't. It may help AI agents that read it describe a business accurately, so it's cheap to add, but it isn't a priority.

How do you measure AI search visibility?

Run a fixed set of real customer questions in several assistants on a schedule, logged out, and record mentions, citations and accuracy. Add AI referral traffic from analytics and Google Search Console data. Report trends and rates rather than single results, because answers vary between runs.

Can I learn AI search optimisation for free?

Yes. Platform documentation, the GEO research paper, practitioner guides, free courses and free tools cover every stage on this page. Paid tracking platforms help at scale, but you can run a meaningful study with free assistant accounts, a spreadsheet and Google Analytics.

More learning paths in the seo and ai search map

  • Technical SEOCrawling, indexing, site architecture, speed and rendering.
  • Content and on-page SEOKeyword research, search intent, briefs and on-page optimisation.
  • International SEOHreflang, market selection and localisation for search.
  • Local SEOGoogle Business Profiles, citations and reviews.
  • E-commerce SEOProduct and category pages, faceted navigation.
  • Programmatic SEOComing soonData-driven page templates at scale.
  • Enterprise SEOComing soonSEO across large sites, teams and governance.
  • Link building and digital PREarning links and mentions through stories and data.
  • Marketplace SEOComing soonRanking inside Amazon, Jumia and Etsy search.
  • App Store OptimisationApp store rankings and conversion.
  • Video SEOComing soonYouTube and video search.