By Steve Merrill, Founder of WRKNG Digital | November 11, 2026
The fastest way to check if ChatGPT and Perplexity know your brand exists is to ask them directly, in a blank session, with no chat history to bias the answer. Here are 8 free tests. No tools, no sign-ups, no budget required.
I spent years staring at ecommerce data trying to figure out why one product shows up and an identical one doesn't. Brand visibility inside AI answers works the same way. You can't guess your way to an answer. You have to go look.
1. Ask ChatGPT Directly in a Fresh, Logged-Out Chat
Open a new chat, logged out or in an incognito window, so memory and history can't help it. Ask "What is [your brand]?" or "Who makes [your product category]?" If it describes you accurately, you exist in its world. If it hallucinates or goes quiet, that's your gap.
2. Ask Perplexity and Read the Numbered Source List
Perplexity shows every source it pulled from, right next to the answer. Perplexity's own FAQ confirms every claim in its answer traces back to a numbered citation, so ask the buying question your customer would ask, like "best [category] for [use case]," and check those citations for your domain. A brand mention with no linked citation is weaker than it looks.
3. Check Google AI Overviews for Your Core Keyword
Search your main keyword logged out and look for the AI Overview box. Click "show more" and Google's own AI features documentation confirms the sources listed there are the ones driving the summary. If a competitor is named and you aren't, that's your answer.
4. Test Gemini With Grounding Turned On
Gemini cites sources when search grounding is active, similar to Perplexity. Ask it the exact same comparison question you asked ChatGPT and compare which domains it names out loud.
5. Run the Comparison and Alternative Prompt Test
Ask "[your brand] vs [competitor]" and "best alternative to [competitor]." This tells you whether AI treats you as a known player in your category or just a name it half-remembers from training data.
6. Ask Claude With Web Search Enabled
Claude's web search mode cites live sources the same way Perplexity does. Anthropic's own announcement of the feature describes it pulling and citing current pages in real time, so ask a buying question in your category and check whether your site shows up anywhere in the cited list, not just the prose answer.
7. Search Your Brand Name Plus Reddit
AI models lean heavily on Reddit threads for grounding real-world answers. Search "[your brand] reddit" yourself. Nothing there means one less path for AI to find you and cite you when someone asks a peer-recommendation question.
8. Check robots.txt and llms.txt for AI Crawler Access
Visit yourdomain.com/robots.txt and confirm you aren't blocking GPTBot or PerplexityBot. Then check yourdomain.com/llms.txt. If it doesn't exist, you can build one from the open spec at llmstxt.org, which gives AI crawlers a direct map to your most citable pages.
How We Chose This List
Every method here costs zero dollars and takes under five minutes. No scanning tool, no API key, no paid subscription. Just the same tools your customers already use to ask questions about your category.
Run All 8 Before You Spend a Dollar on AEO
Most brands guess at their AI visibility. Guessing is free and useless. These 8 tests take thirty minutes total and tell you exactly where you stand across ChatGPT, Perplexity, Google, Gemini, and Claude, today, for nothing.
→ Find out if your business is visible to AI assistants. Start here.
Frequently Asked Questions
Does ChatGPT update in real time, or only from training data?
By default ChatGPT answers from training data, which can be months old. Ask it to search the web, or test it inside a session with browsing enabled, to see your current standing.
Why does Perplexity mention my brand but not link to my site?
A name mention can come from training data alone. A linked citation means Perplexity actually pulled live content from your domain, which is the stronger signal to chase.
What is llms.txt and do I actually need one?
It's a plain-text file that points AI crawlers to your most important pages, similar to a sitemap built for language models instead of search engines. You don't need one to get cited, but it removes guesswork for the crawler.
Can blocking GPTBot hurt my AI visibility?
Yes. If GPTBot or PerplexityBot is disallowed in your robots.txt, those models can't read your site directly, and you drop out of contention for citations regardless of how good your content is.
How often should I run these 8 checks?
Monthly is enough for most brands. AI answers shift as models update and as competitors publish new content, so a single check goes stale fast.

