How do I check what AI engines say about my brand?

Q: How do I check what AI engines say about my brand?

Ask ChatGPT, Perplexity, Gemini, Claude, and DeepSeek the same category and brand-name prompts, then record four things per answer: are you mentioned, where you rank, are you cited with a clickable link, and is the sentiment positive, neutral, or negative. Do it by hand for a free snapshot, or automate it to track changes over time.

Q: What is the difference between an AI mention and an AI citation?

A mention is your brand name appearing in the generated text. A citation is a linked source the engine points to that a reader can click. A brand can be mentioned positively yet never cited, meaning AI shapes opinion about it with no attribution back to its site.

Q: Which prompts should I test to see what AI says about my brand?

Use two types. Category prompts that a buyer types without naming you, like 'best project management tools for agencies', show whether you surface at all. Direct brand prompts like 'What is [your brand]?' or '[your brand] vs [competitor]' show what the engine believes about you.

Q: Why do the same AI prompts give different answers each time?

AI engines are probabilistic, so the same prompt can name you on one run and skip you on the next. That is why a single manual check is a snapshot, not a trend. To measure reliably you need to run the prompts repeatedly and track results over time.

Question

How do I check what AI engines say about my brand?

Maciej Grabek · Accepted Answer

Ask the five engines your customers use - ChatGPT, Perplexity, Gemini, Claude, and DeepSeek - the same category and brand-name prompts, then record four things for each answer: are you mentioned, where you rank, are you cited with a clickable link, and is the sentiment positive, neutral, or negative. Do it by hand for a free snapshot, or automate the tracking so you catch changes over time. The free manual method, step by step You do not need a tool to see how AI describes your brand. You need a short prompt list and 30 minutes. Work through it in order: Write 3-5 category prompts a buyer would type without knowing your name, like "best project management tools for agencies" or "top payroll software for small business". Write 2-3 direct brand prompts that name you: "What is [your brand]?", "Is [your brand] any good?", "[your brand] vs [competitor]". Run every prompt in a fresh session in ChatGPT, Perplexity, Gemini, Claude, and DeepSeek. Use a logged-out or incognito window so your own history does not bias the answer. For each answer, log four columns: mentioned (yes/no), rank (where you appear in a list, or n/a), cited (is there a link to your site), and sentiment (positive, neutral, negative, or wrong). Repeat the whole set a few days later. Answers drift, so one run is a snapshot, not a trend. Starter prompts you can copy Swap the brackets for your brand, category, and top competitor: "What are the best [category] tools in 2026?" "Recommend a [category] tool for [your customer type]." "What is [your brand] and who is it for?" "Is [your brand] better than [competitor]?" "What do people say about [your brand]?" "What are the pros and cons of [your brand]?" The category prompts tell you whether the engine surfaces you at all when nobody names you. The brand prompts tell you what the engine believes about you, which is where hallucinated features, wrong pricing, or a competitor's talking points tend to show up. What counts as a mention vs a citation These are different signals and people conflate them. A mention is your name appearing in the generated text. A citation is a linked source the engine points to, usually numbered or footnoted, that a reader can click. We ran "best time-tracking apps for freelancers" in Perplexity. The answer listed Toggl, Harvest, and Clockify in the prose - three mentions. But the sidebar cited only two URLs: a G2 listicle and Toggl's own pricing page. So Toggl earned both a mention and a citation, while Harvest and Clockify were mentioned but not cited. Getting mentioned puts you in the answer; getting cited sends you traffic and reinforces the engine's trust in your site. When you log your own results, track both. A brand can be described positively and still never get a clickable link, which means AI is shaping opinion about you with zero attribution back to your site. Where the manual method breaks down Checking by hand is the right way to start, and it is free. It also has real limits: Answers vary run to run. The same prompt in ChatGPT can name you on Monday and skip you on Thursday. One check is a coin flip, not a measurement. No history. You see today. You cannot see whether your sentiment improved after a content push, or whether a competitor overtook you last month. It does not scale. Five engines times ten prompts is 50 checks per round. Doing that weekly by hand gets abandoned fast. This is the gap avisibli closes: it runs your prompt set across ChatGPT, Perplexity, Gemini, Claude, and DeepSeek on a schedule, then logs mention rate, rank, citations, and sentiment over time so you can see the trend and get alerted when it moves. The manual method above is genuinely useful for a one-time read. Automation matters once you need to prove a change or catch a regression. avisibli is the GEO platform that publishes this answer library. Self-references are limited to topics where a tool-based answer is genuinely useful to readers.