Home/Strategy/AI visibility report
Get named when buyers ask AI
With our AI visibility report, we ask ChatGPT, Perplexity and Gemini your buyers' questions and find where they leave you out. Our experienced founders and engineers read every answer, compare who gets named and rank the gaps to decide which pages or information to change first.
What is an AI visibility report?
An AI visibility report records how assistants answer the questions your buyers ask without naming your company. It shows whether someone describing a need hears about you, and which competitors they hear about instead. You can see the names a buyer would consider before deciding what to spend on getting yours into those answers.
You can ask an assistant a few questions, use a tracking tool to count mentions, or have us follow each answer. We trace where your company drops out and name the first change to make. The question to ask of any check is whether it tells you what to change.
Each AI answer that skips you costs a shortlist place
When a buyer asks an assistant which providers to consider, the companies in its answer become the names they investigate.
- Competitors take the mention. In our October 2026 check, a software startup appeared in 1 of 90 answers; its main competitor appeared in 43.
- Familiar names keep their lead. Companies an assistant already knows can keep appearing while buyers pass over yours.
- Effort goes to the wrong fix. One company had completed its AI-ready checklist, but Google had indexed only two of its pages.
Finding where each answer drops your company shows which gap to address.
We ask your buyers' questions and read every answer
We put the questions, answers, competing companies and site checks into a report, so each proposed fix has evidence behind it.
Test the buyer's need
We write 30 questions in buyers' words without your company name, to test whether the assistants introduce it.
Keep each answer
We ask ChatGPT, Perplexity and Gemini and save their responses separately so you can inspect what each engine said.
Record competitors and citations
We list named companies beside cited pages, distinguishing a source the assistant used from a provider it recommended.
Inspect what assistants read
We check schema, llms.txt and the markdown copy of your site for gaps in the information available to assistants.
We rerun the same questions after every change
We compare later answers with the first report. Repeating the same questions shows which mentions, descriptions and citations changed after the work.
First
Choose the question set
We discuss the buyer needs and expected competitors with you, then agree which questions will represent those buying situations.
Then
Save a dated baseline
We preserve the exact question wording, answers and date so later checks can use the same starting point.
Next
Select the first gap
We walk through the report on the weekly call and agree which finding to address first.
After changes
Compare the next answers
We repeat the original questions after each change goes live, keeping additional questions separate from the comparison.
Example: spotting where ChatGPT passed over a startup
ChatGPT found one startup's website during a search, read it and still left the company out of its answer. The missing recommendation pointed to the case the page made for the company. That finding directed the work toward stronger evidence on the page ChatGPT had already read.
We fix the exact step where AI drops your company
If an assistant reads your page and passes over the company, we work on the evidence that page gives it. Other gaps need different work.
- Answer the missing search. When your site is absent from search results, we write a page that answers the query the assistant used.
- Strengthen the page's evidence. When an assistant reads a page but leaves you out, we add customer evidence and numbers to its HTML and markdown versions.
- Build familiarity beyond the site. When assistants never name you from memory, we work on third-party mentions and the founder's LinkedIn content.
It fits B2B startups whose buyers research before they talk
Pre-seed to Series A
We check how assistants describe the market for B2B startups with a product, clients and no experienced marketing lead.
AI-enabled service companies
We examine the answers buyers read while deciding whether to trust the technology and the people delivering the service.
Companies entering new markets
We test questions from prospective buyers in the US or Europe who have yet to hear about your company.
Talk to Tin
Let's discuss the questions buyers ask in your market and the competitors you expect them to hear about. Our growth experiments involve no long-term commitment.
FAQ
Should we also test questions that name our company?
Yes, those questions show how an assistant describes you when asked directly. The questions without your name test whether it introduces you to a buyer describing a need. Keeping the two sets separate lets us distinguish what the assistant knows about you from whether it recommends you unprompted.
What if ChatGPT, Perplexity and Gemini disagree?
We examine their answers separately. A combined mention count can hide an engine that never names your company. Keeping the question beside each engine's response lets us see where the omission happens and which sources that engine used to answer it.
Can you see why an assistant left us out?
We can see the answer and the search results and cited pages available with it. The model's internal reasoning stays hidden. We separate what we observed from what we think explains it, then test that explanation by changing the relevant page or information and asking again.
Can our own team run the check with Tin's open-source system?
Yes. The AI visibility audit in Tin asks five buyer questions and saves AI_VISIBILITY.md in the project. Your team can run it on a self-hosted Tin system. When we do the work for you, we test 30 questions on three engines and make the fixes those answers point to.
Talk about what AI tells your buyers today
Meet the founders and engineers running the experiments to discuss the gap you suspect in the answers buyers receive. We can talk through missing mentions, competitors appearing instead, and which change would be worth testing first.