Free AI Visibility Audit
Get Your Free AI Visibility Audit
Your free AI visibility audit shows how ChatGPT, Perplexity, Google AI Overviews and Copilot describe your SaaS today. We test the buying questions your category actually asks, benchmark you against the competitors being cited instead, and deliver a prioritised action plan — completely free, no strings attached.
Four quick questions, so the audit is actually about you
About 4 minutes left
The whole buying conversation, not a handful of prompts
We build a prompt set of 30 to 50 questions in your category and freeze it before we run anything. It covers four stages: the open category question a buyer starts with, the head-to-head comparisons, the qualifying questions about pricing, integrations and company-size fit, and the follow-up turns where a buyer narrows.
That last group matters more than any of the others. Visibility routinely collapses by the third turn of a conversation, and a test that runs one prompt and stops will never see it. Most audits you will be offered do exactly that.
Every prompt is run five times across a two-week window, because model outputs are non-deterministic and a single run proves nothing. We record which model version produced each answer.
ChatGPT
51.3% of US assistant usage
Google Gemini
27.7%, and a different product from AI Overviews
Claude
10.3%, over-indexed with technical buyers
Perplexity
Visible source lists, so citation share is observable
Microsoft Copilot
Grounded in the Bing index
Grok
Where your category has real conversation on X
Google AI Overviews
The generated summary above the results
Google AI Mode
Conversational search, where fan-out changes retrieval
Six numbers, the raw responses, and an ordered roadmap
Every metric is defined in our published measurement standard, with its formula and what it excludes, so you can check the working or hand the definitions to another vendor and compare.
Prompt coverage
The declared question set, so you can see the denominator.
Citation rate
Answers linking your domain, per platform, never blended.
Brand mention rate
Answers naming you without a link. Reported separately, never merged into citation rate.
Share of voice
Against a competitor set declared before measurement.
Brand accuracy
What the assistants get wrong about you, and where it came from.
Source concentration
Which domains own citations in your category. Usually the most useful page.
Alongside the numbers you get the raw model responses, so you can read what was actually said rather than take our summary of it, and a roadmap ordered by damage done rather than by effort.
Ten working days, and here is why it is not faster
- 1Days 1 to 2: we build the prompt set and competitor set with you, and freeze both.
- 2Days 3 to 12: the prompts run five times each across the window, on all eight surfaces.
- 3Days 13 to 14: we score the responses, trace the accuracy failures to their sources, and write the roadmap.
- 4Then a 45-minute call to walk you through it. No obligation, and no pitch deck.
A same-day audit is a single run of a handful of prompts. That is a screenshot, not a measurement, and we would rather take two weeks and give you something you can act on.
What free actually means here
- No cost, no card, no trial that converts into anything.
- You keep the audit whether or not we work together, including the prompt set, so you can re-run it yourself.
- One free audit per company. The work is real and it is done by people.
- We will tell you if the answer is that you do not need us. That happens, and saying so is cheaper for us than a bad engagement.