AI visibility baseline

Find out what AI tells buyers about you.

We run your real buyer questions through ChatGPT, Perplexity, Gemini, and Grok, then show you exactly who gets cited — you, a competitor, or the wrong business entirely. You see every word the engines said.

Start your baseline

Direct answer

An AI visibility baseline measures whether answer engines cite your business by name, with a link, when buyers ask about your category. It scoresbranded visibility (people who already know you) separately fromdiscovery visibility (people shopping your category), because those two numbers are usually very far apart — and the gap between them is the problem worth fixing.

What you get

A number you can act on, and the receipts behind it.

Four engines, scored separately

ChatGPT, Perplexity, Gemini, and Grok each get their own score. They disagree more than you would expect — one business we measured scored 65 on Perplexity and 31 on ChatGPT for the same questions.

Branded vs. discovery split

Most businesses score well when named and near zero when not. Scoring them together hides the only number that affects new customers.

Wrong-domain and entity checks

We flag when engines cite a domain that is not yours, or confuse you with a similarly named business, an old location, or a closed listing.

Full verbatim evidence

Every answer from every engine, in full, in a separate appendix. No screenshots, no cherry-picking — you can audit our reading of it.

Start here

Tell us about the business.

Takes about 10 minutes. Optional uploads make it sharper, but nothing is required — and we never ask for logins, passwords, or account access.

Helpful to have on hand (all optional)

  • A Search Console query export, last 90 days
  • A Keyword Planner or existing keyword list
  • Questions your customers actually ask you
  • Names of 3–5 competitors
  • Any business you get confused with

No search data? No website yet? Still fine — we substitute analyst research.

What happens next

You approve the questions before anything runs.

1

You submit

The intake above. We read it and come back with questions if anything is unclear.

2

We write the prompts

An analyst drafts the exact buyer questions, grounded in your search data if you shared it.

3

You approve them

You see every prompt before it runs. Anything about complaints, pricing, or competitors is opt-in.

4

You get the report

One business day from approval — scores, findings, and the full verbatim answers.

The one-business-day clock starts at prompt approval, not at form submission. An incomplete intake pauses it.

AI visibility baseline — questions

What is an AI visibility baseline?
An AI visibility baseline is a point-in-time measurement of whether ChatGPT, Perplexity, Gemini, and Grok cite your business by name and link when buyers ask about your category. It separates branded prompts (people who already know your name) from discovery prompts (people shopping your category who have never heard of you), because those two scores almost never match.
What do I get back?
A summary report with your visibility score per engine, split into branded and discovery, plus four evidence dimensions: visibility, evidence quality, entity accuracy, and the sources engines cite instead of you. Every claim is backed by a separate appendix containing the full verbatim answer each engine gave, so you can read exactly what a prospect sees.
Do I need Search Console access?
No. Search data is optional. If you export your Search Console queries or a Keyword Planner list, we ground the discovery prompts in what people actually search for. If you do not have that data — or do not have a website yet — we use your customer questions plus analyst research instead. We never ask for logins, passwords, API keys, or account access.
How long does it take?
One business day, starting from the moment you approve the prompt list — not from when you submit the form. We review your intake, write the prompts, and send them to you for approval first. If the intake is missing something we need, we come back with questions and the clock pauses until it is answered.
Why do you show me the prompts before running them?
Because the prompts are the measurement. A baseline run on the wrong questions produces a number that means nothing — we have seen a healthy site score 17 out of 100 purely because the prompt set was invented rather than grounded in real demand. You approve the exact wording before anything runs, including any question about complaints, pricing, or competitors.
Do you ask questions about complaints or competitors?
Only if you approve them. Negative-framed prompts like "complaints about X" are some of the most diagnostic questions we run — being cited on them usually means your review posture is strong — but they are opt-in on the intake form, and a no simply removes them from the set.
What if my business does not have a website?
You can still run a baseline. Without a domain, the thing being measured is your entity footprint: what AI engines believe about your business name, location, and services based on directories, reviews, maps, and third-party mentions. That is often where the real problem lives, and it is a common pre-launch check before a new site goes live.
Is this the same as an SEO audit?
No. An SEO audit measures where you rank in a list of blue links. This measures whether AI engines name you inside a synthesized answer, which is a different retrieval path with different failure modes — wrong domain cited, wrong business entity conflated with yours, or a competitor named in your place. The two findings often disagree.