2026-08-03 · Jack Stovell
How to check what AI says about your brand (free, in ten minutes)
You can check what AI says about your brand in about ten minutes, free: ask each major assistant the questions your customers actually ask — once with web browsing on, once off — and record what it says and which sources it cites. Do that monthly and you have a working tracking system. Paid trackers automate this at scale; for a single brand, the manual loop plus a free technical checker covers most of it.
How do you see what ChatGPT says about your company?
Ask it directly, using the questions a real customer would type — not your brand slogan. Prompts to run, verbatim, with your details swapped in:
- What is [brand]?
- What does [brand] do, and who is it for?
- Is [brand] legitimate? What do people say about it?
- What are the best [category] options for [customer type] in [market]?
- How does [brand] compare with the alternatives?
Run each prompt in ChatGPT, Claude, Perplexity, and Google's AI results if you have them — and run each twice, with browsing on and off. Browsing-off shows what the model absorbed in training; browsing-on shows the retrieval path, which is the one you can change this quarter. The category question is the one that counts: not being named in "best [category]" answers is where business is actually lost, and why ChatGPT doesn't recommend your business digs into that specific failure.
How do you track brand mentions in AI search results?
Tracking is the same exercise with a spreadsheet and a calendar. One row per run: date, assistant, browsing on or off, the prompt, whether you were mentioned, whether the answer was accurate, and every source cited. Repeat monthly. Three things make the log useful rather than noise:
- Assistants are not deterministic. The same prompt can produce different answers on different runs, so a single answer is a sample, not a verdict. Run the prompts that matter two or three times.
- The citations column is the actionable one. Those pages are currently speaking for you — or for whoever is being recommended instead. They tell you exactly which pages to improve or outdo.
- Watch for drift, not just presence. Answers change as models and indexes update; an answer that was accurate at the start of the quarter can quietly go stale.
What does a free checker automate?
A checker automates the technical half — whether AI systems can read your site at all — which no amount of interviewing the assistants will show you. Our free AI visibility checker tests your robots.txt AI stance, detects JavaScript shells that serve crawlers an empty page, checks for identity markup (JSON-LD), llms.txt and a sitemap, and runs a live model probe: does an AI model recognise the name at all? It is rate-limited, and free.
The two halves diagnose different failures. If the interviews show you absent and the checker shows blockers, fix the blockers first — then work on the content. Get your business cited by AI covers that improvement side.
Do you need a paid tool to track AI mentions?
For a single brand: probably not. Paid AI-visibility trackers exist and they make sense at scale — an agency reporting across many clients, or a portfolio tracking many prompts, models and markets with trend lines — where running the manual loop by hand stops being realistic. For one brand with one market, a monthly manual loop plus a free checker captures most of the value for none of the spend.
One thing no tool changes, at any price: tracking measures the answer, it doesn't move it. If what AI says about your brand is wrong or missing, the fix lives on your pages — open, crawlable, direct, and consistent — not in the dashboard that told you so.
Check your AI visibility — free
Related guides
- How to get your business cited by AI assistants
- Why doesn't ChatGPT recommend your business?
- Generative engine optimisation (GEO), explained