OpenAI
ChatGPT Visibility Tracker
ChatGPT is the most used AI assistant, and with web search enabled it answers buying questions with named products and cited sources. Track how it treats your brand, every day.
How ChatGPT answers and cites
For current or factual questions, ChatGPT runs a web search, reads the results and writes an answer that lists sources. Which brands it names depends on which pages that search retrieves, so visibility can change from day to day as pages and rankings change.
Two OpenAI crawlers matter: OAI-SearchBot indexes pages so they can be cited in search-enabled answers, and ChatGPT-User fetches a page when a user asks about it. GPTBot, the training crawler, does not affect citations.
Answers are not deterministic. The same prompt can name three brands one day and four the next, which is why a single screenshot is not a measurement.
How LeapScope tracks ChatGPT
Each tracked prompt is run on the ChatGPT answer surface once a day through a scraping provider, so results reflect what a user sees rather than a bare model API call. Every run stores the full answer text and its source list.
The answer is scanned for your brand and your competitors, the position of each mention is recorded, sentiment toward your brand is judged on the sentences that mention it, and cited URLs are normalized to real pages.
What is recorded for every prompt
- Mention
- Whether the brand appears in the answer
- Position
- First, early or later in the answer
- Sentiment
- Positive, neutral or negative portrayal
- Citations
- Every source URL, resolved to the real page
- Competitors
- The same for each named competitor
- History
- One data point per prompt per day
Limitations
- ChatGPT Shopping results (product cards) are a separate surface and are tracked only on plans that include it.
- Answers depend on the account region; prompts are run from the countries you select.
- Answers without web search (pure model memory) are not tracked by default, because they cannot cite pages and change only when models are retrained.
Other engines
FAQ
How do I know if ChatGPT recommends my brand?
Run the prompts your buyers ask, daily, and record each answer. A ChatGPT visibility tracker does this automatically and reports the share of answers that mention your brand, where you appear in them, and what sources they cite.
Why does ChatGPT give different answers to the same question?
Generation is probabilistic and the web search behind a grounded answer can return different pages on different days. Daily tracking turns that variation into a rate you can act on instead of a coin flip.
Does blocking GPTBot remove my site from ChatGPT answers?
No. GPTBot gathers training data. Citations in search-enabled answers depend on OAI-SearchBot and ChatGPT-User. Check all three in your robots.txt.
Can I see the exact sources ChatGPT cited?
Yes. Every run stores the source list, so the citation report shows which domains and pages ChatGPT relies on for your prompts and which cite competitors but not you.
Which countries can I track ChatGPT in?
Prompts can be run from the markets you sell in; the countries available per plan are listed on the pricing page.
Does LeapScope track ChatGPT through the API or the app?
The answer surface users see, with web search, not a bare API call. The API without search cannot cite pages and does not reflect what buyers read.
Track ChatGPT and the other engines from one workspace
LeapScope runs your prompts on every supported engine daily and reports mentions, citations and share of voice side by side.