How it works

What actually happens when you run an audit

When someone asks ChatGPT or Perplexity a question, the engine answers using pages it can read and quote. If your page does not answer that question plainly, it quotes somebody else's. An audit tells you which questions you are losing, and why.

We fetch the page

You paste a URL and we download that one page — the same way a crawler would. We strip out navigation, headers, footers, cookie banners and scripts, and keep the main content, because that is the part an engine would actually quote. Nothing is installed and no account is needed.

We read one page at a time, not your whole site. Audit the page you care about ranking.

We work out what the page is about

Before we can ask useful questions we need to know the subject. We read the content and the domain and reduce it to a short description of the business — “HVAC installation company”, “SaaS project management tool”. Everything downstream depends on this, so it is worth a glance when your report loads: if the niche is wrong, the questions will be too.

We generate 25 questions people ask AI about that topic

These are not keywords and they are not taken from your page. They are the kinds of questions a real person types into an assistant about your subject — a mix of what something is, how it compares, what it costs and what goes wrong.

We detect the language your page is written in and write the questions in that language, phrased the way a native speaker would ask them — not translated from English.

We check which ones your page answers

Each question is graded against your actual page content — not by keyword matching, but by reading it. Every question lands in one of three buckets:

Covered: The page states an answer the reader can act on without going anywhere else.
Partial: The page raises the topic but stops short — the reader would still have to infer it, search further, or contact you.
Missing: The page does not address it, or mentions it only in passing.

Covered counts once, Partial counts half, Missing counts nothing. That ratio is your Question Score, and it is the single largest part of your overall score at 30%.

We run 6 technical checks on the page

The questions measure whether your content answers what people ask. These six measure whether an engine can find, parse and trust the page in the first place. Unlike the question score, every one of these is a verifiable fact about the page — we read your robots.txt, parse your JSON-LD and look at your markup.

AI crawler access20%

Whether your robots.txt actually lets GPTBot, ClaudeBot, PerplexityBot and Google-Extended read the page. Block them and nothing else on this list matters — the page cannot be cited if it cannot be read.

Structured data15%

JSON-LD markup that tells an engine what the page is: an article, a product, an organisation, a set of questions and answers. Without it the engine has to infer everything from prose.

Answer structure15%

Whether the page is laid out as questions and direct answers. A heading that asks a question, followed by a paragraph that answers it, is far easier to quote than the same fact buried mid-paragraph.

Citations and statistics10%

Concrete numbers and links to outside sources. Pages that back claims with data get cited more than pages that assert them.

Freshness5%

A machine-readable publish or update date. Static pages like an About or Pricing page rarely carry one and that is fine — it matters most for guides, blog posts and anything where recency changes the answer.

E-E-A-T signals5%

Evidence of who wrote this and who stands behind it: an author byline, a link to an author or about page, an Organization schema.

llms.txt is checked too, but deliberately kept out of the score and shown as a bonus. No AI provider has confirmed it affects citation ranking, so scoring you on it would be inventing a standard that does not exist yet.

You get fixes, not just a number

Every gap comes back with something you can paste: the answer text for a missing question, an FAQ block, or the JSON-LD snippet for a schema you are lacking — written in your page's own language. Each fix shows the effort involved and what it is worth, so you can start with the cheap ones.

What the score means

The overall score is the seven signals — question coverage plus the six technical checks — weighted and combined into one number out of 100.

71–100

Well optimised

An engine can read the page and it answers most of what people ask.

41–70

Needs GEO optimisation

The page is readable but leaves real questions unanswered — usually the quickest wins.

0–40

Needs significant work

Either engines are blocked outright, or the content answers very little of what is asked.

One thing worth being clear about: this measures how well your page is set up to be cited. It is not a measurement of whether an engine cited you today — that changes constantly and no tool can promise it. What we can tell you is whether the page gives an engine a reason to.

Want the technical detail?

The methodology page has the exact formula behind every signal, the thresholds, and what we deliberately do not measure.

Read the methodology See an example report Audit a page