AI Visibility Audit
Find out exactly how ChatGPT, Perplexity, Gemini, Claude and Google AI Overviews describe your brand — before your next buyer does.
A transcript, a score, and a plan.
Not a dashboard login and a PDF of best practices. The actual answers, and what to do about them.
The raw transcript
Every prompt we ran and every answer we got back, verbatim, across five engines. You see precisely what a buyer sees — including the answers you won't enjoy reading.
Visibility scoring
Mention rate, citation rate and sentiment per engine, with competitive share: who gets recommended when you don't.
Accuracy review
A catalogue of what the engines get wrong about you — stale pricing, wrong positioning, outdated leadership, invented limitations — and the source feeding each error.
Prioritised remediation
A ranked list of fixes ordered by expected lift against effort, specific enough for your team to execute without us.
A page from a real audit
Anonymised, but this is the actual format: the prompt, the verbatim answer, who got cited, and the traced cause.
How we run it
Prompt design
We build 60+ prompts from your actual buying journey — category questions, comparisons, objections and branded queries — not vanity searches for your own name.
Multi-engine execution
Every prompt runs across all five engines, repeated to account for non-determinism, captured with citations and links intact.
Scoring & diagnosis
We score each response and trace every inaccuracy back to the source that produced it, so fixes target causes rather than symptoms.
Typical findings
- Brand absent from 60–80% of category prompts
- Pricing quoted from a page retired 18 months ago
- A competitor named as market leader in your own category
- Capabilities you've shipped described as unavailable
- Conflicting company facts across directories
- Citations pointing at reviews rather than your site