How do I find which sites ChatGPT cites about my brand? A manual method and the best tools
Find which sites ChatGPT cites about your brand: the free manual check, why it falls short, and the tools that track citations by page, domain, Reddit, and YouTube.
There are two ways to find out which sites ChatGPT cites about your brand. The free way takes about an hour and tells you something. The tool way takes about ten minutes to set up and tells you what changes over time. Most teams should do the first once, then move to the second.
The manual method
With ChatGPT's search turned on, ask the questions your buyers would ask. Not "tell me about [brand]" alone, but also "best [category] for [use case]", "[brand] vs [competitor]", and "is [brand] worth it". When the answer cites sources, ChatGPT shows them as links, and you can open the sources list to see which pages it drew on.
For each prompt, write down:
- Whether your brand was named.
- Which domains were cited, and which specific pages.
- Whether your own site appeared.
- Which competitors were named instead.
Do this for 15 to 20 prompts and a pattern usually appears. You will see the same few domains again and again: a comparison site, a Reddit thread, a YouTube review, a publication, or a competitor's blog.
Why the manual method is not enough
It has four limits, and they matter.
- Answers vary. The same prompt can return different sources on a different day or run, so one check is a sample, not a measurement.
- Personalization and location. What you see may not be what a buyer in another country sees.
- No history. You cannot tell whether a source is rising or fading.
- It does not scale. Twenty prompts across several engines is already a tedious afternoon.
There is also a subtler gap. The sources cited are only part of what the model looked at. ChatGPT runs sub-queries behind your prompt (query fan-outs) and fetches pages it never cites. Seeing both explains why a page was read but not credited.
What a tool should show
A citation tracker worth paying for answers these questions:
- Which pages and which domains are cited, ranked by frequency and by position in the answer.
- Where your own domain sits, and how that changes week to week.
- Which third-party sources dominate, including Reddit, YouTube, and LinkedIn.
- Where your brand is named without a link (offsite mentions).
- Which crawlers are fetching your pages, so you can tell "not cited" from "not reached".
Tools compared
| Tool | Starting price (from catalog) | Citation features |
|---|---|---|
| Promptwatch | Free Explore; Essential $95/mo | Page and domain citations, URL positions, Page Tracker, offsite mentions, Reddit, YouTube, and LinkedIn reports, Citation Trends, crawler logs from Professional |
| AthenaHQ | Starter $295/mo | Prompt and response analysis; the Athena Citation Engine is Enterprise |
| Scrunch AI | Core $250/mo | Competitor benchmarking and crawler traffic analytics |
| Profound | Enterprise custom | Citation share and answer-level analytics |
| Ahrefs Brand Radar | Custom Prompts included with paid Ahrefs plans | AI Visibility Index, YouTube and Reddit sources |
| LLM Pulse | From €49/mo | Citations, owned-media tracking for YouTube and Reddit |
| Superlines | From €79/mo | Citation sources and share of voice, three engines on self-serve |
Why Promptwatch is the best fit for this job
Most tools in the table can list cited domains. Promptwatch goes further in the directions that matter for this specific question.
Page and domain level. You see which exact URLs are cited, with their position in answers, not just a domain tally.
Citation Trends. Every citation is classified by content type and source type, and each page gets a citation rate over time, plus first cited, ramp-up days, peak rate, and decay half-life. That tells you whether a source is a durable fixture or a blip.
Third-party sources. Dedicated Reddit, YouTube, and LinkedIn citation reports, plus "My channels" filtering, show where outside voices carry your category. Offsite mentions flag pages that name your brand without linking.
Page Tracker. You can track specific URLs, yours or a competitor's, and see which prompts cite them.
The crawl side. Agent Analytics logs real, IP-verified AI crawler hits, with a crawl-to-citation path and citation rate per page. If a page is fetched often and never cited, the problem is the content. If it is never fetched, the problem is access. Crawler logs start on Professional at $245 a month.
There is also a free route to begin: the AI Visibility Report is a no-login scan, and the GEO Inspector Chrome extension shows ChatGPT's fan-outs and the fetched versus cited sources as you browse, running locally.
What to do with the list
Once you have the cited domains, sort them into three groups.
- Your own pages. Strengthen the ones cited and look at why others are not.
- Third-party pages you can influence. Reviews, comparison articles, and directories where you can correct or supply information.
- Pages you cannot influence. Note them, and compete by publishing a stronger answer yourself.
Reddit and YouTube deserve special attention, since they often show up for comparison questions. Taking part honestly in relevant threads and having accurate video coverage can matter more than another blog post.
Who should pick which
- Promptwatch: the strongest choice if you want citations, crawler logs, and content actions together.
- Ahrefs Brand Radar: worth a look if you already pay for Ahrefs and want index-scale research on sources.
- LLM Pulse: a transparent, lower-cost tracker for small teams, at a weekly default cadence.
- AthenaHQ and Scrunch AI: if you want their action queue or crawler-side approach.
- Profound: enterprise buyers who want a custom program.
A caution
Citations are evidence of influence, not a guarantee of recommendation. Tracking shows which sources engines use and gives you levers to pull. It cannot make an engine cite you.