GenGrowth

GEO tools

AI Visibility Check

Ask a frozen question set on the engines you select, several times each. Compare mentions and citations with independently read page evidence. Historical single-engine reports keep their original scope; no report proves why a model chose its answer.

Sign in to run a visibility check

A run costs real provider calls and points at a frozen knowledge base, both of which belong to an account.

Measurement scope and limitations

Choose ChatGPT, Perplexity, or both. Results come from the recorded API surfaces, not consumer chat sessions; each report identifies its engines and observed models.

  • Observed gaps are not causal explanations of a model's answer or proof of future growth. Any site inventory is bounded and explicitly scoped; incomplete page evidence cannot establish absence.
  • It is a sample. The same question can cite four sources one minute and none the next, which is why each is asked several times and why every number carries its interval.
  • Questions written the way a buyer would ask usually get answered from memory. Their mentions are worth counting; their citations are reported without a denominator.
  • The questions are in English, because that is the language the wording was measured in. A market other than the calibrated one is reported as such.

Questions about this check

Why ask the same question several times?
Because one answer is not a measurement. Asked four times, the same question produced four completely different citation lists and zero overlap. A run that asked once would report a coin toss as a finding, which is why every number here carries the sample count it came from.
Why do some questions have no citation rate?
The model decides whether to search. Wording measured to make it search carries a citation rate; wording a buyer would actually type usually gets answered from memory, and a citation rate over answers that never looked anything up would be a number about nothing.
It says I was not mentioned. What do I fix?
The current report can offer evidence-conditioned content, page-citability, third-party-placement or comparison actions when independent reads support them. A content gap applies only to the audited inventory. Missing or incomplete evidence remains unattributed. Historical v1 reports contain observations only.

Where to go next

This run observes. The other two tools are where something changes.