Free Audit
Free AI Visibility Audit
Ask ChatGPT about your business right now. Are you confident in what it says back? Most brands have no idea — and what these models “know” about you is quietly shaping whether you get recommended at all.
This free audit checks how ChatGPT, Perplexity, and Google’s AI Overviews currently describe your brand, and flags the gaps that are costing you visibility.
Most businesses have no idea whether AI systems can describe them, and there is no dashboard that will tell you. Search Console does not report it. Analytics does not report it. The first sign is usually a customer mentioning they asked ChatGPT for a recommendation and got a competitor.
An AI visibility audit answers a narrow question with evidence: when someone asks an AI assistant about what you do, are you named — and if not, what specifically is missing?
What the audit covers
1. Entity recognition
Whether Google and the major AI systems recognise your business or your name as a distinct thing rather than a string of words. This includes checking for a Knowledge Graph identifier, a knowledge panel, a Wikidata item, and whether anything at all distinguishes you from others with a similar name.
2. Live assistant testing
A fixed set of prompts — the ones a real buyer would actually type — run across ChatGPT, Gemini, Perplexity and Claude. Not “what is my brand”, which tells you nothing, but the commercial questions: who does this, who is good at this, who serves this area. The output records whether you appear, how you are described when you do, and which competitors are named in your place.
3. The corroboration gap
An inventory of what independent sources currently say about you, and which of the records that matter you do not yet have. This is usually where the answer is found. Most businesses invisible to AI are invisible because nothing they do not own says they exist.
4. Structured data and entity home
Whether your site states, in a machine-readable form, who you are and how everything you own connects. Missing markup, contradictory markup, and identity split across several pages are all common and all cheap to fix.
5. Name and claim consistency
How many different versions of your name, address, description and core claims are in circulation. Inconsistency dilutes confidence faster than almost anything else, and it stays invisible until somebody counts.
What you get
- A written report of what was tested and what came back, including the raw assistant responses
- A named list of the competitors being recommended instead of you, and what they have that you do not
- A prioritised list of fixes, ordered by effect rather than by effort
- A baseline you can re-run later, so the next audit measures movement rather than opinion
Who it is for
It is most useful if you sell something people research before buying, if your competitors are starting to appear in AI answers, or if you have invested in SEO and cannot work out why it is no longer turning into enquiries.
It is not useful if you have no website and no third-party presence at all. In that case the answer is already known and the audit would only confirm it.
What it will not do
It will not produce a knowledge panel. It will not get you cited next week. Both are outcomes of the corroboration work the audit identifies, and that work takes months, not days.
What the audit does is stop you guessing about which months of work are the right ones.
Common questions
How is this different from an SEO audit?
A traditional SEO audit measures whether pages can rank. This measures whether an entity can be identified and cited. The overlap is real but partial — sites that rank well can still be invisible to AI systems, and increasingly are.
Why test more than one assistant?
They draw on different sources and different retrieval methods. Appearing in one and not the others is diagnostic in itself, and points at which part of the corroboration layer is thin.
How long does it take?
The testing and analysis take a few days. Re-running it later takes less, because the prompt set is fixed.
Request an audit
Ryan Gooding runs AI visibility audits for Australian businesses and personal brands, and writes about the underlying method in Branding: is bullshit. To request one, get in touch with your website and a one-line description of what you sell.
To work through it yourself first, the guide to getting a Google Knowledge Panel sets out the same sequence in full, and the Claim, Frame, Prove method explains how to test any claim you make.
Try it yourself, free
Before you book anything, run the free AI search visibility audit. Enter your domain and up to three keywords and it will tell you where you rank in Google and whether Gemini mentions you at all. No signup.