ANSWER ENGINE OPTIMIZATION
Ranking first and being the answer are two different wins.
You can hold position one and never be named in the answer sitting above it. That gap is what answer engine optimization measures.
Answer engine optimization is the work of getting named inside an AI-generated answer, rather than ranked in a list of links beneath it.
The outcome is different from SEO’s outcome, so it needs measuring separately. Everything else about the term is contested, including who coined it.
How to measure it, by hand, for nothing
This is the method most published AEO advice describes, and it works. We are writing it out in full because you should be able to do it without us, and because you cannot judge a tool that automates a method you have not seen.
Write the questions a buyer asks
Keywords will not do it. Use the sentences somebody types when they are choosing: best X, cheapest X, which X should I use. Ten is enough to see a pattern.
Ask every engine, not one
ChatGPT and Google AI Overviews disagree constantly. A domain that looks dominant is often one engine's index talking, and you cannot see that from inside one engine.
Write down who got named
Every brand and every source URL in the answer, per question, per engine. The brands are the competition. The URLs are where to get in.
Repeat it on a schedule
One run is a snapshot. The number that matters is whether you appear in more answers this month than last, on the same questions.
It stops scaling at about ten questions across three engines. That is the reason a tool exists.
WHAT THE METHOD RETURNS
We ran it on ourselves first
9 buyer questions about AI visibility tools, asked across every engine we can reach. Our own domain was named in none of the answers, which is a result you can check rather than a claim you cannot.
- Engines asked
- 7
- Answers returned
- 62
- Distinct domains cited
- 133
- Times we were named
- 0
The 133 domains named instead are the worklist. One competitor, frase.io, appeared in 19 answers across 6 of the 7 engines. Roughly a quarter of the citation surface was not competitor pages at all, but roundups, YouTube and Reddit threads, which are reachable in a way a competitor’s homepage is not.
Engines asked: Google AI Overviews, ChatGPT, Perplexity, Gemini, Google AI Mode, Microsoft Copilot, Claude (API).
Before you measure, be readable
An engine cannot name a page it cannot fetch or cannot parse. These checks are free, need no account, and answer in one page load.
Is your site blocked?
Resolves your robots.txt for ten named AI crawlers, with real group precedence.
Will AI quote the page?
Twelve checks on whether a model can lift a clean passage out of it, and which passage.
What text survives?
Your page fetched with no JavaScript run, the way most AI crawlers read it.
Common questions
What is answer engine optimization?
Answer engine optimization is the work of getting named inside an AI-generated answer rather than ranked in a list of blue links. The outcome is different from SEO's outcome, so it has to be measured separately: you can hold position one and still be absent from the answer above it.
Is answer engine optimization different from SEO?
The work overlaps heavily. The measurement does not. SEO asks where you rank for a query. Answer engine optimization asks whether the answer names you, and if not, which sources it named instead. A page can rank first and go uncited.
How do you measure answer engine optimization?
Ask a fixed set of buyer questions across multiple answer engines on a schedule, and record which brands and which source URLs each answer names. Doing it by hand is a real method and it is what most published advice describes. It stops scaling at roughly ten questions across three engines.
Can you pay to appear in an AI answer?
Not through us and not through any structured data markup. Engines choose what to cite from what they read. The only levers are being readable to their crawlers and being present in the sources they already read.
Longer read: what the AEO checklist actually asks of you, which checks each item on the standard list against a primary source and throws most of them out.
Run it on your own domain.
The same 7 engines, your questions, and the list of who got named instead of you.