SIGNAL
What the world tells you
Buyers are asking AI about your category right now.
Find out whether you are in the answer, why not, and what to publish about it. Then find out whether the fix worked, checked against evidence rather than asserted.
What AEO/GEO Monitor is
The AEO/GEO Monitor tracks whether your brand appears in the answers AI assistants give to the questions your buyers actually ask, and whether you appear in Google's AI Overviews for a given market. Where you do not appear it diagnoses why: missing entity information, content that is not structured so a model can quote it, a narrative that does not match how the category is described, or directory listings that are absent. It then produces content briefs, drafts, structured data and outreach to the publications AI assistants actually cite, and re-checks afterwards to confirm whether the fix worked. Answer Engine Optimization and Generative Engine Optimization are the measurement. Content engineering is the work.
The problem
The answer happens before anyone reaches your website.
A buyer used to search, see ten links, and choose one. Now they ask, and get a paragraph naming three brands. If you are not one of the three, you were never in the running, and nothing in your analytics will tell you it happened. There is no impression, no click and no bounce. The channel is invisible by construction, which is why most brands have no idea whether they are winning or losing in it.
What you get
Four things it does, in order
Whether you are named
Your category questions asked on a schedule across the major assistants and tracked over time, so you have a trend rather than somebody's screenshot.
On a schedule
Why you were not
Which sources the answer drew on, and what the brands named instead have published. Absence with a cause attached rather than an alert.
The cause, not the alarm
What to publish about it
A specific brief for the gap, structured to be extracted and attributed rather than merely read, and written against your own brand specification.
In your voice
Whether the fix worked
The same monitor re-checks the question afterwards, so the loop closes on evidence rather than on somebody saying it is done.
Checked, not asserted
In practice
Are you in the answer, or is your competitor?
Every tracked question is kept with its full answer and the sources behind it, run after run, so a change is visible as a change. Where you are absent, the diagnosis explains which of four things is missing, and the fix is checked afterwards against evidence rather than asserted.
Coverage
What it checks, and where
| Surface | What we read |
|---|---|
| ChatGPT | Answers to your tracked questions, and the sources cited |
| Perplexity | Answers and citations, which are unusually explicit here |
| Gemini | Answers to the same question set, for comparison |
| Google AI Overviews | Per market, because the country genuinely changes the result |
| Directory listings | Whether the places assistants draw entity facts from actually list you |
| Your own pages | Whether they are structured so a model can extract and attribute a passage |
A market is always required for the Google side. There is no default country, deliberately.
Where it connects
What it draws on and feeds
- Brand DNA supplies how your brand should be described, so the monitor can flag when AI describes you differently.
- Competitor Radar shares the competitor set, so you can see who is winning the answers you are losing.
- Content Studio takes the content brief and drafts the page, in your voice.
- Analytics shows what the traffic did once the fix landed, so the loop closes on a number.
Questions
AEO/GEO Monitor, answered
What is the difference between AEO and GEO?
In practice, very little. Answer Engine Optimization usually refers to getting extracted into direct answers and snippets; Generative Engine Optimization to earning citations inside AI-written responses. Both describe the same job: being the source a model quotes.
How is this different from SEO?
Search engines rank pages. Answer engines quote passages. Optimising for the first gets you a position; optimising for the second gets you named inside somebody else's sentence, which requires content structured to be extracted and attributed.
Can you guarantee we will appear in AI answers?
No, and anyone who does is selling something. What we can do is tell you where you stand, why, what to change, and whether the change worked.
How volatile is this channel?
Very. In August 2026 Reddit lost about 86% of its ChatGPT citations in four days because of a vendor-side change nobody announced. That volatility is the argument for monitoring rather than treating it as a project.
How many questions should we track?
Start with the ten to fifteen questions a buyer genuinely asks before choosing in your category. More than that and you are paying to watch noise.
Do you publish the content for us?
We produce the brief, the draft and the structured data. You publish, because it goes on your domain and should read like your brand.
Find out what AI is telling your buyers.
Tell us five questions someone might ask before buying in your category. We will run them before the call and show you the answers.