What is an LLM visibility audit?
The LLM Visibility Audit is our productised entry point to AI search. We build a structured set of prompts that mirror how your buyers research: brand queries, product questions, category comparisons, “best X for Y” recommendations and problem-led questions where you should be the answer.
We run them across the major AI platforms, record every response and analyse what comes back: whether you appear, how you're described, who appears instead, and which sources are shaping the answer.
Why it matters now
AI is already describing your brand to prospects. Do you know what it's saying?
AI assistants are forming opinions about your brand in conversations you never see. They may be recommending a competitor, quoting outdated information, misdescribing your product or leaving you out of your own category entirely.
Unlike a search ranking, you can't simply check a position, and most businesses have no idea what AI is telling their prospects. The audit makes it visible, and gives you a clear, prioritised plan to improve it before the narrative sets.
By the numbers
- More than 2 billion people a month use Google's AI Overviews, according to Google's July 2025 results.
- In March 2025, 18% of Google searches produced an AI summary, and users clicked a link inside it on only 1% of visits, according to Pew Research Center. Most people take the answer as given.
- 75% of knowledge workers use AI at work, according to Microsoft and LinkedIn's 2024 Work Trend Index. Your buyers are already asking AI about your category.
If the answer they get is wrong or leaves you out, you won't see it in your analytics. The audit makes it visible.
What we deliver
Prompt set design
A tailored set of buyer-intent prompts covering your brand, products, competitors and the key questions in your category.
Cross-platform testing
Every prompt run systematically across ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews, with responses captured and logged.
Visibility and share of voice
How often you're mentioned, cited and recommended, benchmarked against your main competitors.
Accuracy and sentiment review
Where AI gets your brand right, where it gets it wrong, and the tone it takes when it describes you.
Source analysis
The websites, publications and data sources AI systems draw on in your category, and where you're missing from them.
Prioritised action plan
A ranked set of recommendations to improve your AI presence, with the effort and expected impact of each.
Everything is delivered as a detailed written report, followed by a walkthrough session so your team understands the findings and owns the next steps.
Who it's for
- Brands that want to know what prospects hear when they ask AI about them
- Leadership teams building the business case for investment in AI search
- Companies that have spotted inaccurate or damaging AI answers about themselves
- Marketing teams who want a clear baseline before starting AEO and GEO work
How we work
Scope
We agree the brands, products, competitors and questions the audit will cover.
Test
We run every prompt across each platform, repeatedly, to capture patterns rather than one-offs.
Analyse
We score visibility, accuracy and sentiment, and trace the sources behind each answer.
Report
We deliver the report and action plan, and walk your team through it.
It works best alongside AEO, GEO & AI Search Optimisation, AI Content Strategy and Search Engine Optimisation.
Frequently asked questions
Which AI platforms does the audit cover?
ChatGPT, Claude, Gemini, Perplexity and Google AI Overviews as standard. We can add other platforms, such as Microsoft Copilot, where they matter in your market.
AI answers change all the time. Is a one-off audit useful?
Yes. Because individual responses vary, we test each prompt multiple times and analyse patterns rather than single answers. That gives you a reliable baseline, and many businesses repeat the audit periodically to track progress.
What happens after the audit?
You'll have a prioritised action plan your team can implement independently. If you'd like support, the findings feed directly into our AEO, GEO and content strategy work.