What does Claude say about my brand?

The direct way to find out what Claude says about your brand is to ask it the questions a buyer would ask, not your brand name directly. Asking Claude to describe your brand by name gets you a summary of what it knows. Asking the buyer question, the best options in your category, tells you whether it thinks of you unprompted, which is the version that actually matters.

Two different questions, two different answers

"What do you know about [brand]?" asks Claude to describe you directly. It will usually produce something from training data, and may or may not search live depending on the question and the surface it runs on.

"What are the best options for [category]?" is the real test. This is what buyers actually ask, and it is the question where your brand either shows up on its own merit or does not show up at all.

What to look for in the answer

• Are you named at all, and how early in the list.
• What Claude says about you, if anything, and whether it is accurate.
• Which competitors get named instead or alongside you.
• Whether the answer cites a source, which means Claude searched, and which domain.

Why the answer can change

Claude decides for itself whether to search, so the same question can get a training-based answer one week and a search-based one the next, with no signal from outside that the mode changed. Training-based answers move slowly, on model release cycles; search-based ones can shift as fast as any live search result.

Get the full picture, not just Claude

Buyers research across more than one engine. ASRM's free scan asks your category's real buyer questions across ChatGPT, Claude, Perplexity, and Gemini, and shows your score, what the engines say about your brand, the competitors taking your place, and the sources they cite. About a minute, no signup.