Enter your category, location and buyer type and this tool generates five prompts written the way real buyers ask ChatGPT, with a scoring sheet for recording whether your brand is named, where it appears and who else is named. Fifteen answers will show you the pattern; a full audit gives you a number you can act on.
The tool below generates five prompts that match how buyers in your category ask ChatGPT for a recommendation, so you can test them in a few minutes and see whether your brand comes up, where it sits, and who ChatGPT names instead. No account or email required.
How to use the generator
Fill in three fields. Category is what you sell, in a buyer’s words (“payroll software”, “immigration lawyer”). Location is where your buyers are, because ChatGPT’s answers change with place. Buyer type is who is asking (“a 15-person startup”, “a first-time home buyer”), because the brands recommended to a solo founder and to an enterprise team differ.
The generator returns five prompts spanning the ways one intent gets phrased: a direct ask, a comparison, a situational description, a budget-constrained version and a trust question. They are a starter prompt variation set, not a full prompt cluster, which in our audits runs to 20 to 40 prompts.
How to run the test
Open ChatGPT in a logged-out session on your phone, in the country you entered. Paste the first prompt and read the answer. Open a fresh chat and ask it again, and a third time. Move to the next prompt. You will finish with 15 answers.
For each answer, record four things in the scoring sheet:
- Mentioned: is your brand named anywhere in the answer, yes or no.
- Position: if named, is it first, second, third or lower in the order ChatGPT lists brands.
- Linked: does a link to your own domain appear among the sources.
- Competitors: every other brand named, in order.
These are the same four fields we classify in a full audit, described on the full method write-up, so your results are comparable to a paid baseline at a much smaller sample.
How to read what you find
Fifteen answers show a pattern, not a percentage.
| What you see across 15 answers | What it most likely means |
|---|---|
| Named in nearly all, usually top three | You are an established entity in this category; protect the position |
| Named in some, position varies widely | Visible but inconsistent; the model is unsure how you rank against peers |
| Named only for the trust prompt | ChatGPT knows you exist but does not place you among the recommendations |
| Never named, same three competitors every time | A consensus has formed without you; see why ChatGPT does not mention your brand |
| Different brands in every answer | A fragmented category with no settled answer, which is the easiest kind to enter |
Read the Named column (0 to 3 per prompt) first, then Best position: being named in all three runs but never above position four is a different problem from being named once in first place. The Linked column tells you whether ChatGPT is reading your own site or only third parties; a brand that is named but never linked is being described from other people’s pages. The Competitors column is often the most useful: a competitor named in 12 of 15 answers has a mention rate you cannot yet compute precisely, but it is clearly the brand ChatGPT trusts, and the set of names that recur is your real competitive field in this channel. The guide to measuring ChatGPT visibility shows how to turn these observations into tracked metrics.
When to get the full audit
The free test answers one question: does ChatGPT currently tell buyers about you. It cannot give a mention rate within a defensible band, cover every phrasing, or be repeated identically next month to show movement.
A full baseline audit covers 20 to 40 prompts with about 100 real answers each, collected from genuine mobile connections in your target country, classified by hand and re-run at day 30 and monthly. If your 15 answers show you absent or inconsistent in a category that matters to revenue, see what ChatGPT says about your brand across the whole cluster. To expand the test yourself first, follow the step-by-step DIY audit guide.
Frequently asked questions
How do I use the ChatGPT prompt generator?
Enter the category you sell in, the location your buyers are in and the type of buyer you serve, then press generate. You get five prompts phrased the way a buyer types them. Open ChatGPT in a logged-out session on your phone, paste each prompt, and record whether your brand is named, its position, whether your domain is linked and which competitors appear. Repeat each prompt three times.
Why ask each prompt three times instead of once?
Because ChatGPT does not answer the same way twice. The brands it names, and their order, change between runs as the model samples and as its web search returns different pages. One answer tells you almost nothing. Three answers per prompt, fifteen in total, is the minimum to see whether a brand is consistently present, consistently absent or appearing sporadically.
Should I be logged in to ChatGPT when I test?
No. A logged-in account carries chat memory and personalization that can skew the answer toward brands you have discussed before, including your own. Use a logged-out session, ideally on a mobile connection in the country your buyers are in, so that what you see is closer to what a new buyer sees. Our full audits hold these conditions constant for the same reason.
What does it mean if ChatGPT never names my brand?
It usually means the model has too little consistent information about you to rank you alongside competitors, or that the pages it retrieves for buying questions in your category do not mention you. Neither is permanent. The guide on why ChatGPT does not mention your brand walks through the common causes, most of which relate to entity consistency and third-party coverage.
When should I get a full audit instead of using the free tool?
When you need a number you can defend to a board or a budget owner, when you sell in more than one country, or when you want to track movement month to month. The free tool gives you five prompts and fifteen answers. A full audit covers a cluster of 20 to 40 prompts with about 100 real answers each per market, classified by hand and re-run monthly.
Find out what ChatGPT says about your brand before you spend anything.
Apply in about two minutes. We check your keywords, then run your real buyer prompts live on a 30-minute call. If we cannot create the result for your keywords, we tell you on the call, not after an invoice.
See if my keywords qualify →