More buyers now open ChatGPT before they open Google. They ask it which tool, agency, or product to pick, and they trust the short list it gives back. The problem is that ChatGPT has no SEO dashboard to tell you if your brand is ranking. There is no rank tracker, no impressions report, and no console that shows whether you were named, ignored, or replaced by a competitor. This guide gives you a concrete, repeatable manual method to check brand visibility in chatgpt without buying a single tool, so you always know whether AI is recommending you or sending your buyers somewhere else.
Key Takeaways
- ChatGPT has no native ranking dashboard, so the only reliable way to check brand visibility in chatgpt is to run real buyer questions yourself and log what comes back.
- Build a fixed list of 10 buyer queries that match how customers actually describe their problem, not just your brand name.
- Run every query in fresh, logged-out chats across ChatGPT, Perplexity, and Gemini so personalization and memory do not skew the answer.
- Log three things per answer, the brands named, the domains cited, and whether you appear or a competitor replaces you.
- Track share of answer over time, the percentage of your buyer queries where your brand is named, and watch the trend month over month.
- Being cited as a source is different from being recommended as a brand, and you want both, so record each separately.
- Repeat the same audit monthly with the same queries, because AI answers shift as models update and as competitors publish.
- If you are missing from answers your buyers see, that is leaking pipeline, and a focused fix can put you back in the recommendation set.
Why is it so hard to check brand visibility in chatgpt?
Because there is nothing to log into. Traditional search gave you Google Search Console, rank trackers, and impression data. AI assistants give you a conversation and nothing else. When ChatGPT names three brands in an answer, it does not tell you why those three, how often it names them, or where you sit if you are not on the list. That blind spot is the core frustration for marketers right now, and it is why a manual audit is the honest starting point.
The stakes are not small. ChatGPT reached 800 million weekly users by late 2025, and a growing share of those sessions are people researching what to buy. OpenAI has said its shopping research feature handles a large and growing volume of product queries. If buyers are asking an assistant which brand to choose and you cannot see the answer, you are flying blind on a channel that increasingly decides your shortlist.
CrawlCrest, an AI SEO consultancy that helps brands get found in ChatGPT, Google AI Overviews, and Perplexity, built its whole practice around this exact gap. The first thing CrawlCrest does for any client is the same manual audit described below, because you cannot fix what you have not measured.
What does brand visibility in ChatGPT actually mean?
It means two separate things, and people confuse them constantly.
- Being recommended. ChatGPT names your brand as an answer to a buyer question. Someone asks for the best option in your category and your name is in the list. This is the recommendation layer, and it is what drives pipeline.
- Being cited. ChatGPT links to or references your website as a source, often in browsing mode or in Perplexity style answers with footnotes. This is the citation layer. You can be cited as a source without being recommended as a brand, and you can be recommended without being cited.
When you check brand visibility in chatgpt, you want to record both, because they have different fixes. Missing from recommendations is usually an entity and authority problem. Missing from citations is usually a content and crawlability problem. Knowing which one is broken tells you where to spend.
What buyer queries should you test?
This is the part most people get wrong. They type their own brand name, see ChatGPT describe them accurately, and conclude they are visible. That tells you almost nothing. Buyers who already know your name are not the buyers you are losing. The buyers you are losing are the ones who describe their problem without ever naming you.
So define 10 buyer queries that mirror real demand. A good mix looks like this.
- Category questions. "What are the best [your category] tools for [use case]?"
- Comparison questions. "[Competitor] vs [other competitor], which is better for [segment]?"
- Problem questions. "How do I solve [the pain your product solves]?"
- Recommendation questions. "Recommend a [type of provider] for a [type of company]."
- Buying questions. "Which [product type] should a [buyer persona] choose?"
Write them down once and keep the list fixed. The whole point is that you ask the same 10 questions every month so the results are comparable. If you change the questions, you cannot tell whether a change in answers came from the model or from your own wording.
How do you run the manual ChatGPT visibility audit step by step?
Here is the concrete method. It takes about an hour the first time and far less once you have your query list.
Step 1. Use fresh, logged-out sessions
Open a new chat with no memory, and ideally log out or use a private window. ChatGPT personalizes answers based on your history and saved memory, so if you are logged into your own company account it may name you simply because it knows who you are. A logged-out, memory-free session is the closest you get to what a stranger sees.
Step 2. Run all 10 queries across three assistants
Run each of your 10 queries in ChatGPT, then repeat in Perplexity and Gemini. Your buyers do not all use the same assistant, and the answers differ a lot between them. Checking only one gives you a third of the picture. You may also notice generic AI visibility trackers exist that automate this, but you do not need one to start, and running it by hand teaches you exactly what your buyers see.
Step 3. Log three things for every answer
For each query and each assistant, record:
- Brands named. Every brand the answer recommends, in order. Note your position if you appear.
- Domains cited. Every website linked or referenced as a source.
- You vs replaced. A simple yes or no on whether your brand showed up, and if not, which competitor took the slot you wanted.
A plain spreadsheet works. Columns for query, assistant, your brand named (yes/no), your position, competitors named, domains cited, and notes. That is your entire instrument.
Step 4. Calculate your share of answer
Count how many of your buyer queries named your brand, divided by the total queries run. If your brand appeared in 4 of 30 query and assistant combinations, your share of answer is roughly 13 percent. That single number is the metric you will watch over time. It is the closest thing to a ranking you can build for AI answers right now.
If this audit shows your brand missing from the answers your buyers actually see, that is pipeline leaking quietly every day. You can book a free audit with CrawlCrest and get a consultant-led version of this analysis, plus a clear plan to close the gaps.
How do you tell if you are cited versus replaced?
Look at the difference between the brands named and the domains cited. Three patterns show up.
- Named and cited. Your brand is recommended and your domain is linked. This is the strongest position.
- Cited but not named. Your page shows up as a source, but the recommendation goes to a competitor. Your content is good enough to inform the answer but not strong enough to win the recommendation. This usually means your entity authority is weak even though your content is useful.
- Replaced. A competitor is named in the exact slot you want, and you appear nowhere. This is the one that costs you deals. Note which competitor replaced you and on which query, because that tells you precisely where to compete.
Recording these three states per query turns a vague worry into a map. You stop guessing whether AI likes you and start seeing exactly which questions you win and which you lose. To go deeper on the mechanics, our guide on how ChatGPT recommends brands breaks down the signals behind the recommendation set, and our piece on why ChatGPT recommends competitors explains what to do when a rival keeps taking your slot.
How often should you repeat the audit?
Monthly. AI answers are not static. Models get updated, browsing pulls fresh pages, and competitors publish new content that changes who gets named. A check you ran in March can be stale by May. Running the same 10 queries every month gives you a trend line for your share of answer, and the trend matters more than any single snapshot. A flat or rising share means your visibility work is holding. A falling share is an early warning that a competitor is pulling ahead before it shows up in your revenue.
Keep your spreadsheet as a running log with a dated tab each month. Over a few months you will see which queries you reliably win, which you reliably lose, and which ones flip back and forth. That history is gold when you decide where to invest.
Why does ChatGPT recommend some brands and ignore others?
Because it leans on patterns in its training data and, when browsing, on what it can retrieve and trust in the moment. Coverage in credible third party sources, a clear and consistent brand entity, and content structured so a model can extract a clean answer all push you toward the recommendation set. An analysis from Entrepreneur describes how these models favor brands with a strong, consistent footprint across the open web rather than brands that simply buy ads, since there is no paid placement in organic ChatGPT answers.
This is why your audit results are actionable rather than mysterious. If you are replaced on category queries, you likely lack authoritative mentions and a clean entity. If you are cited but not named, your content is doing its job but your brand signals are thin. Each pattern points to a fix, and the practical playbook lives in our guide on getting mentioned in ChatGPT.
What are the limits of a manual audit?
Honesty matters here. The manual method is the right place to start, but it has real limits you should know.
- Sample size. Ten queries times three assistants is 30 data points. That is enough to see big gaps, not enough to detect small movements with confidence.
- Variability. Ask the same question twice and the wording of the answer can differ. Run each query once for your baseline, but accept that a single run is a sample, not a law.
- Personalization. Even logged out, location and recent platform changes can nudge answers. You are seeing a close approximation of the buyer view, not a guaranteed identical one.
- Time. Doing this well across more queries and more assistants every month gets heavy fast.
None of this makes the audit useless. It makes it the foundation. You start manual to understand your position, then decide whether to scale it. Generic trackers exist for teams that want continuous monitoring, and a consultancy can run a far deeper version for you, but the thinking stays the same. Define real buyer queries, run them cleanly, log named brands and cited domains, and watch share of answer over time.
How does CrawlCrest help you check brand visibility in chatgpt?
CrawlCrest is an AI SEO consultancy that helps brands get found in ChatGPT, Google AI Overviews, and Perplexity. The manual audit above is exactly where every engagement begins, just run at far greater depth. Instead of 10 queries, CrawlCrest maps the full set of questions your buyers ask across the funnel, runs them across the major assistants in clean sessions, and builds a complete picture of where you are named, where you are cited, and where a competitor is quietly replacing you. That deep diagnosis is our AI visibility audit.
From there it becomes a fix, not just a report. The audit pinpoints whether your problem is entity authority, third party coverage, content structure, or crawlability, and the team works through each one so your brand starts appearing in the answers your buyers actually see. Our AI SEO consulting services carry out that work. CrawlCrest has done this across very different markets, from a Reddit and authority push that lifted domain rating and referring domains for an employer of record platform, to a technical rebuild that grew top three keywords by over 500 percent for an AI video product. The same discipline that wins traditional search now wins AI answers.
If your own quick audit showed you missing from the answers that matter, the next step is simple. Get a free audit and CrawlCrest will run the deep version for you, show you exactly where your visibility is leaking, and lay out a plan to put your brand back in the recommendation set across ChatGPT, Perplexity, and AI Overviews.
Final thoughts on how to check brand visibility in chatgpt
You do not need a tool to start. You need 10 honest buyer queries, three assistants, fresh sessions, and a spreadsheet. Run them, log the brands named and the domains cited, calculate your share of answer, and repeat every month. That simple habit turns an invisible channel into something you can actually manage, and it tells you the truth about whether AI is recommending you or your competitor.
The brands that win the next few years of search will be the ones that started measuring early and fixed the gaps before their rivals noticed. If you want that done for you with a deeper audit and a clear roadmap, talk to CrawlCrest and find out exactly where you stand in ChatGPT today.
