AI visibility
Why you are not cited, said page by page
A model cites what it can extract, attribute and verify. The audit checks those three fronts on a specific page and returns changes you can apply today.
- Extractability: whether the answer sits in a paragraph that can be lifted whole
- Entity: whether the model knows who signs the page and can attribute it without hesitating
- Evidence: whether claims carry a source and a date, or go bare
The problem
- Your page is the best on its topic and the model cites a blog with half the content.
- You know something is off, but not whether it is the content, the markup, or that nobody else mentions you.
- The best-practice lists you have read do not say what to fix first on your page.
A model does not reward quality: it rewards what it can copy without getting it wrong.
How it works
The audit works on the public HTML of a URL, or on text you paste. It does not guess: it reads what is there.
- 1
It reads the page the way a model would
Without running your JavaScript, because the models' crawlers do not run it either. If your content only appears after hydration, the audit says so on the first line instead of scoring text the model never sees.
- 2
It separates failures by front
A list of thirty mixed warnings helps nobody. Each finding is attributed to a front, and within each front they are ordered by effort against effect.
- 3
It returns text, not advice
Every problem arrives with the current fragment and a ready-to-paste replacement. Saying improve clarity fixes nothing; showing the rewritten paragraph does.
What you get
One report per page, ordered by what moves the needle most.
- The diagnosis by front, including what is already right, so you know what not to touch.
- The proposed rewrites, with the original beside them so you can judge the change.
- The limitations found, said out loud: if the page is thin or depends on JavaScript, the report declares it instead of glossing over it.
Measure before you pay anything
The scanner runs 10 real responses in about 60 seconds. No signup, no card. If the number it returns tells you nothing, do not buy.
Measure my brandWho it is for
- Anyone who already knows they are not cited and needs to know where to start.
- Content teams that want a concrete list rather than a course.
- Agencies auditing other people's sites who need a repeatable format.
It is not for you if...
It is not for you if you do not yet know whether you are cited. Auditing a page before measuring is fixing something that may not be broken.
An honest comparison
| By hand | With Citenza | |
|---|---|---|
| What it costs you | Reading the page and trusting your own judgement. | The same examination for every page on the site. |
| The method | A different criterion for every person who reviews it. | Separate fronts and findings ordered by effort against effect. |
| Repeating it next month | No way of knowing whether the change worked. | You audit again and compare the two reports. |
Availability by plan
The audit starts at Starter. Each run covers one page, and the estimated cost is shown before launching. The plan's spending cap still rules: if the audit does not fit in what is left, it does not run and you are told why.
Honest questions
- Does auditing a page guarantee I get cited?
- No, and distrust anyone who promises it. The audit removes the obstacles we know get in the way: content the crawler cannot see, unsourced claims, authorship that cannot be attributed. Removing them improves your odds. Guaranteeing a citation would be selling something nobody controls.
- Does it work with a site built in React or Vue?
- It works, and that is usually where it gives its most useful diagnosis. If your content is painted in the browser and does not travel in the HTML, the audit tells you so and everything else becomes secondary: no markup or wording fixes a page the crawler receives empty.
- How many pages should I audit?
- Start with the ones that should already show up and do not: the ones answering the questions where your competitors do appear. Auditing the whole site before knowing which questions matter is expensive and disorderly. Measure first, then audit what the measurement points at.
- Can I audit a competitor's page?
- You can, because only public HTML is read, the same any visitor sees. It is in fact one of the most useful uses: if a model cites a competitor and not you, comparing the two audits usually explains why in two minutes.
Before rewriting, measure
First find out which questions you are missing from. Auditing the right page beats auditing ten.