AEO Course
AEO 101 to 105
A practical self-study course for people who already know the SEO basics and want to understand how AI answer systems choose, quote, cite, and absorb sources.
~20 min read, 5 modules
SEO and AEO overlap, but they are not the same game. Traditional SEO usually tries to win a ranked position in a list of links. Answer Engine Optimization tries to win a place inside the answer itself.
In classic SEO, the search engine's process runs roughly: crawl, index, rank, display. In AEO, that engine process is longer: crawl, index, retrieve, rerank, extract, attribute. The engine has to find your page, decide it fits the intent, pull a usable passage, and decide whether to name or cite you.
That is why a page can rank first in organic search and still be absent from the AI answer. A lower-ranking page with a clear definition, useful table, fresh facts, or original data can become the citation.
Mental model
SEO is trying to get your book placed on the front table at a bookstore. AEO is trying to get your paragraph quoted by the person writing the summary card.
Concrete example
A plumbing page titled "Water Heater Repair Denver" may rank well but get skipped if the page opens with sales copy. A competitor with pricing ranges, common repair tables, service-area details, and a last-reviewed date is easier for an answer engine to cite.
Why it matters
A page is not finished just because it targets a keyword. It needs to be easy for an answer engine to lift, verify, and attribute, even when the answer reduces clicks.
Key takeaways
- AEO means getting cited, mentioned, or absorbed into AI-generated answers.
- Ranking and citation are related, but they are different gates.
- The most useful content is extractable, attributable, and corroborated.
- Treat each answer engine as its own surface.
Quick check
Why can a top-ranking page still miss an AI answer citation?
It is easy to assume that "AI search" is one channel. It is not. Each engine has its own index, crawler rules, freshness bias, source mix, and trust model.
ChatGPT Search uses a hybrid retrieval system. OpenAI retrieval access matters, especially OAI-SearchBot. Perplexity is citation-forward and freshness-sensitive. Google AI Overviews and AI Mode build on Google systems but do not mirror organic rankings exactly.
Gemini leans heavily on entity clarity. Copilot starts with Bing visibility. Claude appears more selective and often favors stable, established, expert sources.
Mental model
Think of each engine as a different hiring manager filling the same job. One trusts recent work samples, one trusts elite references, one trusts an internal database, and one wants a clean portfolio.
Concrete example
A fresh guide with clean answer blocks may do well in Perplexity and ChatGPT. Claude may still prefer an older institutional source if the topic is sensitive or the newer page feels promotional.
Why it matters
Engine mechanics determine the work plan. You do not fix a Claude authority problem with the same first move as a Google local visibility problem.
Key takeaways
- ChatGPT Search needs OpenAI retrieval access, not only Google rankings.
- Perplexity rewards clear, fresh, extractable web sources.
- Google AI features share foundations with Search but are not identical to organic results.
- Gemini depends on entity clarity, Copilot on Bing visibility, and Claude on authority.
Quick check
What should you identify before trying to improve AEO visibility?
RECON means studying what the answer engine actually did: cited sources, answer shape, trusted third parties, and query language. Save raw responses across engines so you can compare behavior over time.
MIRROR means updating your content to match the useful structure the engine rewarded. Use answer-first blocks, question-style headings, tables, FAQs, and concise 40 to 80 word answers when they genuinely help.
SEED means earning presence in sources the engine already trusts. FOUNDATION means making the brand, product, location, service, people, proof, schema, and content base complete and consistent.
Mental model
AEO is like preparing a witness for court. The witness must be easy to find, consistent under questioning, supported by evidence, and quoted accurately in the transcript.
Concrete example
Instead of "We offer premium pool service in Scottsdale," write a direct service answer that names weekly tasks, frequency during peak heat, neighborhoods served, and equipment checks.
Why it matters
Most AEO improvements are disciplined content, entity, and source improvements aimed at retrieval and extraction. The hard part is applying the loop to real prompts and real competitors.
Key takeaways
- Use RECON to learn the engine's actual query language and cited sources.
- Use MIRROR to match answer structure, not just keywords.
- Use SEED to earn credible third-party corroboration.
- Use FOUNDATION to keep first-party entity information complete and consistent.
Quick check
Why is schema treated as hygiene rather than a citation hack?
Answer engines often trust third-party sources: publishers, review sites, comparison pages, directories, Reddit, YouTube, Wikipedia, Wikidata, and local platforms. The goal is to become corroborated across the web, not only to publish more pages on your domain.
White-hat AEO includes extractable content, accurate schema, entity completeness, original research, earned PR, legitimate third-party presence, and crawler access. Gray tactics need evidence and disclosure. Manipulative tactics should not be deployed on reputation-bearing brands.
Measurement needs discipline because AI answers vary. Run each query multiple times per engine, fix geography and language, store raw responses, and treat results as directional.
Mental model
AEO measurement is more like polling than checking a rank tracker. One answer is a sample. A query set across engines, repeated over time, starts to become evidence.
Concrete example
For "best AI receptionist for home services," measure ChatGPT, Perplexity, Google AI Mode, Copilot, and Claude separately. Track citations, absorbed positioning, and which review or comparison sites dominate.
Why it matters
Without measurement, AEO becomes storytelling. With measurement, you can see whether a change affected source selection, answer absorption, or gatekeeper visibility.
Key takeaways
- Third-party presence is often the gate to citation.
- Use white-hat tactics by default on real brands.
- Measure selection, absorption, and gatekeeper share.
- Repeat runs because AI answers vary.
Quick check
What is the difference between selection and absorption?
A brand is not simply visible or invisible in AI. It may be strong for alternatives, weak for pricing, cited in Perplexity, ignored by Claude, and mentioned without a link in ChatGPT.
The basic measurement unit is a query matrix. Use buyer-intent framings such as category terms, comparisons, alternatives, verticals, use cases, personas, and pricing or cost queries. Store raw outputs and track citations plus absorbed claims.
Apply the method by property type. Editorial sites need extractable pages and authority. Commerce needs product feeds and merchant infrastructure. Local services need entity truth, reviews, directories, profiles, and transaction readiness.
Mental model
AEO is not a one-time site audit. It is a lab notebook. You form a hypothesis, run repeated trials, record raw outputs, change one thing when possible, and update your beliefs when the engines change.
Concrete example
If a SaaS brand is absent for dental-office prompts, capture a baseline, add a real vertical page with direct answer blocks and proof, then re-run the same prompts over several weeks.
Why it matters
The winning move is usually the combination of crawl access, extractable content, entity consistency, third-party corroboration, and measurement.
Key takeaways
- Measure a matrix of query framings, not one prompt.
- Track citation selection and answer absorption separately.
- Keep commerce feeds separate from editorial citation work.
- Re-test every few weeks because platform behavior changes fast.
Quick check
Why should AEO experiments avoid changing too many variables at once?
Closing summary
AEO is the work of becoming the source an answer engine can confidently use.
The durable principles are simple: be crawlable, be clear, be quotable, be corroborated, be complete, and measure honestly. The unstable layer is platform behavior around those principles.
Use the current numbers as mid-2026 field notes. Use the method as the repeatable practice.
Want to talk it through?
A free 20-minute call, no pitch. We'll tell you what we'd do first.
Book a 20-minute callNote: the platform-specific numbers and behaviors in this course are mid-2026 field notes. Treat them as directional, not permanent.