AI visibility check
When buyers ask AI what to use, find out if it names you.
Said by AI asks the questions your customers ask (“what’s the best … for …”) and records which brands the answers recommend. You see how often you are named, who gets named instead, and what to change.
“What is the best project management software for a small agency?”
1. Asana: flexible boards and timelines, strong for client work
2. ClickUp: lots of features in one place, steeper learning curve
3. monday.com: visual workflows, popular with creative teams
4. Trello: simple Kanban, good for small teams
An illustrative question from an audit, with a fictional brand. The full audit repeats this 60 times.
What the audit does, step by step
- 01
Write the questions
From your category, it writes 20 questions the way buyers ask them: best, cheapest, for startups, alternatives, most trusted.
- 02
Ask them three ways
Each question goes to OpenAI's GPT-4o mini as a direct answer, a ranked list and a research-style list: 60 answers in total.
- 03
Read every answer
A second pass reads each answer: were you named, in what position, and which other brands were recommended.
What it does not do: it doesn't scrape live ChatGPT, Gemini or Perplexity sessions, and it can't see personalised answers. Read the full method
The $49 audit
Everything in the full report
Visibility score
The share of 60 answers that named your brand.
Result by answer style
How often you appear in direct answers, ranked lists and research-style lists.
Competitors named instead
Up to 10 brands the answers recommend, with how often each came up.
Where you were mentioned
The question and the sentence around each mention.
Action plan
Prioritised next steps based on your score and the competitors that beat you.
One-off vs subscription
| Said by AI | Tracking tools | |
|---|---|---|
| Price | $49 once | From $29/month |
| Answers read | 60 | Depends on plan |
| Live platforms | No (model API) | Yes, several |
| Tracking over time | Re-run when you need it | Yes |
Best for a baseline before a launch, a content push or a client proposal. For weekly tracking, use a subscription.
Start with the free checkQuestions
Which AI does the check ask?+
OpenAI's GPT-4o mini, the same model family behind ChatGPT, through the API. Each question is asked in three answer styles: a direct answer, a ranked list and a research-style list. It reflects what the model knows about your category. It is not a scrape of live ChatGPT search results, and it does not query Gemini, Claude or Perplexity.
What exactly is in the $49 audit?+
20 buyer-style questions about your category, each asked in three styles (60 answers). You get the share of answers that named you, a count per answer style, up to 10 competitors named instead of you with how often, quotes of where you were mentioned, and a prioritised action plan. It arrives as an HTML email report, usually within a few minutes.
Why do results change if I run it again?+
AI answers are not deterministic. The same question can produce a different list a minute later. That is why the audit samples 60 answers instead of one, and why a score should be read as a range rather than an exact number.
How is this different from Otterly.ai or Peec AI?+
Those are subscriptions that track prompts over time across several live AI platforms. Otterly's Lite plan is $29/month for 15 prompts. Said by AI is a single $49 baseline read with an action plan. If you need ongoing tracking across platforms, a subscription is the better tool.
What's the refund policy?+
Full refund within 24 hours, no questions asked. Reply to your order email.
Who built this?+
Doruk, an indie maker in Antalya, Turkey. No team and no investors. Reply to any order email and it reaches me directly.