The AEO Action Loop
What is AEO?
Answer engine optimisation (AEO) is the practice of making a business the answer AI assistants give when a buyer asks for a recommendation. AI visibility is the outcome it is measured by.
Where search engine optimisation competes for a position in a list of links, AEO competes for a place in a synthesised answer that may name two or three companies and link to fewer. Being tenth is usually the same as being absent.
AEO, GEO, AI visibility: which word means what
These four are used interchangeably and describe different layers. Keeping them apart is what stops a team reporting a number nobody can act on.
| Term | Layer | What it refers to |
|---|---|---|
| AEO | The practice | The work of becoming the answer an assistant gives. |
| AI visibility | The outcome | Whether and how you appear in those answers, as a measurement. |
| AI search | The channel | Assistants that answer by retrieving live web content. |
| GEO, LLMO, AI SEO | Synonyms | Other names for the same practice, used interchangeably. |
AEO, GEO and AI visibility describe different layers›
What can honestly be measured
- Whether a named crawler may fetch an exact URL, and what the initial HTML says before JavaScript runs.
- What one answer engine returned for one question, at a stated time and API surface.
- Which brands that answer named, and whether each was mentioned, recommended, or cited.
- Which pages the answer used as sources.
- Whether a published change is publicly visible.
- Whether a later, comparable check of the same question came back different.
And what cannot
- A readiness score is not evidence that an assistant recommends you.
- A provider we did not check is not a negative result.
- One answer at one moment is a sample, not a trend.
- Mentioned, recommended, and cited are different outcomes.
- AI answers vary by provider, API surface, time, location, and personalisation.
- Published proves the change is live. Only a comparable later check can say whether it moved the answer.
A readable site can still be absent from every answer›
No tool or agency can guarantee a citation. There is no submission mechanism, and answers vary by provider, API surface, time, location, and personalisation.
What doing AEO actually looks like
It works as a loop rather than a checklist, because each stage produces the input to the next. Robot Visible is built around these five.
Measure the answer. Make the change. Prove what moved.
Stage 1 of 5 · Measure reality
Read the site the way a crawler does, then ask a real buyer question and record what came back.
- What you can point at
- Initial-HTML readiness, live buyer questions, provider and timestamp
- Where it comes from
- Site scan and live answer checks
The question you were measured on is the answer the next stage reads.
Stage 1 of 5 · See why others win
Read the exact answer, who it named, which role each brand held, and the pages it leaned on.
- What you can point at
- Exact answer, recommendation/mention/citation roles, source URLs, competitor pages
- Where it comes from
- AI Answers, source evidence, competitor benchmarking
The competitor and source that won the answer become the gap the next move has to close.
Stage 1 of 5 · Choose the next move
One ranked opportunity, tied to the gap that was actually observed rather than a generic checklist.
- What you can point at
- Ranked opportunity tied to the observed gap and its evidence
- Where it comes from
- Prioritised fixes and content opportunities
The chosen opportunity carries its evidence into the change you ship.
Stage 1 of 5 · Make the change
Paste-ready guidance, or a reviewable change you approve before anything is published.
- What you can point at
- Paste-ready guidance or an approved GitHub/Shopify change
- Where it comes from
- Fix plans, draft pull requests, Shopify publishing
The published change and its date become the baseline the next check is compared against.
Stage 1 of 5 · Prove what moved
Ask the same question again after the change is live, and record whether the answer improved, held flat, or declined.
- What you can point at
- Comparable before/after answer, publication verification, improved/flat/declined status
- Where it comes from
- Proof, monitoring, history, alerts
A flat or declined result is not a failure of the loop. It is the next thing to measure.
Buyer questionobserved answercompetitor/source advantageapproved actionpublished changecomparable result
Common questions
- How is this different from a normal SEO audit?
- Traditional SEO tools often focus on keywords, backlinks, and technical SEO. Robot Visible focuses on crawler policy, the initial HTML its scanner receives, structured data, supporting pages, clear answers, and classic search basics. Its AI interpretation is a grounded simulation; it does not claim to reproduce a provider's crawl.
- What is AEO, and is it different from AI visibility?
- AEO is answer engine optimisation: the practice of becoming the answer an AI assistant gives when a buyer asks for a recommendation. AI visibility is the outcome that practice is measured by, and AI search is the channel it happens in. GEO, LLMO, and AI SEO are commonly used synonyms for the same practice. A tool can report your AI visibility without doing any AEO, which is the distinction worth checking when comparing vendors.
- How do you decide which companies are my competitors?
- Suggested competitors are inferred from the public AI answer and the pages it cited, and they are labelled as suggestions rather than findings. A company is only suggested when it is named in the answer itself, so a publisher or directory that was merely cited is not proposed as a business rival. Suggestions can be incomplete or wrong, and they establish no relationship between you and the company named. Inside your workspace you choose which competitors are actually tracked; a suggested set from a free scan is not carried across for you.
- Why did the same question give a different answer this week?
- AI answers vary by provider, API surface, time, location, personalisation, and model version. Two checks of the same question can legitimately differ without anything changing on your site. That is why a single answer is treated as a sample rather than a trend, why a before-and-after comparison states the interval between the two checks, and why a provider that was not checked is recorded as not checked rather than as a negative result.
- Will fixing these issues guarantee more traffic or AI citations?
- No tool can guarantee rankings, citations, recommendations, or traffic. Robot Visible surfaces concrete, fixable gaps that commonly hold websites back from being discovered, cited, and recommended, and prioritises them by impact. Acting on them improves your odds, but results depend on your market, content, and competition.
What a first measurement usually looks like
We ran this on ourselves and lost. The exact question, the verbatim answer, the twenty sources it drew on, and the seven tools it named instead of us are all published, including everything the result does not license us to say.
Read next
3 min
AEO evidence: what can it prove?
For teams that understand the AEO category and need to distinguish measurable evidence from claims no tool can support.
3 min
AEO vs GEO vs AI visibility
For anyone comparing vendors or writing a brief who needs to know whether two documents using different words are describing the same work.
4 min
How to improve AI visibility: the AEO loop
For teams that have measured their AI visibility once and need a repeatable way to turn that measurement into changes and evidence.
The fastest way to understand AEO is to measure once
A free scan reads your site the way a crawler does, asks one real buyer question of a live answer engine, and shows you the answer in full. No sign-up.