What AI visibility means
A new place where customers decide.
When someone asks ChatGPT, Claude, Gemini, Perplexity or Google's AI answers who to call, where to eat or which firm to trust, the engine names a handful of businesses. Being one of them depends on two different things, and we measure them separately because they behave differently.
Readability
Whether AI crawlers can reach your site, whether your pages state plainly who you are, what you do and where, and whether structured data names the business. This is checkable, and it is either in place or it is not.
Recommendation
Whether the engines actually name your business when real customer questions are asked, and which pages they read before answering. A site can pass every technical check and still be named zero times. The engines cite pages that answer the question people ask.
What Janua assesses
Two levels of assessment.
Website Foundation Check
Run it yourself at visibility.janualabs.com. It reads your site live the way an AI crawler would: the requirements engines need before they can read a site, marked pass or fail, plus separate Classic Search and AI Readability grades, with every check shown and how to fix anything not in place. It does not test whether AI assistants name or recommend your business.
AI visibility audit
We put a fixed set of real customer questions to ChatGPT, Claude, Gemini and Perplexity with live web search on, and record every answer: whether you were named, which businesses were named instead, and which pages each engine read. Every result in the report traces to the stored answer behind it. Scoped and quoted up front.
What Janua improves
The foundations engines rely on.
- Access. AI crawlers allowed, pages reachable, nothing important hidden behind scripts the engines do not run.
- Identity. Structured data and on-page facts that state your name, services, locations and contact details the same way everywhere.
- Pages that answer the question. Documents written for the questions customers actually ask, which is what engines cite.
- Evidence and authority. Specific, checkable facts on your own site, and accurate listings in the directories and sources the engines read.
- Reviews and reputation signals. Honest work on the signals engines weigh, never review gating or invented feedback.
This is the human-led part of AI Visibility, delivered and quoted up front. How the fix works →
How competitor context is used
Your competition, as the engines see it.
The competitors in a Janua report are the businesses the engines actually named in place of yours, taken from their stored answers. We never build a competitor list by assumption. Seeing who is named, how often and on which questions shows where the gap is, and which of their pages the engines are reading.
The process
Assess. Remediate. Validate. Monitor.
- 01
Assess
The free check, then the audit: what engines can read, and what they say today.
- 02
Remediate
Close the gaps the audit found, in order of impact, with every change recorded.
- 03
Validate
Re-run the same check and the same questions, so the before and after use one method.
- 04
Monitor
The same questions to the same engines each month, reported against the previous month.
What we will not claim
A transparent limitation.
No one can guarantee a specific AI answer, ranking, citation, or recommendation. Janua's work improves the measurable foundations that help AI systems and customers find, understand, and accurately represent your business.
Engine answers vary from one run to the next. Our reports show that variation as variation, not as progress, and every figure traces to the answers behind it.
Start with what the engines can read today.
The free check takes under a minute and shows every check it ran.