Are fixed prompts enough to measure AI search visibility?
Many brands measure AI search visibility by running a predefined list of prompts and counting how frequently they appear in the answers.
For example, a SaaS company might track prompts such as "best CRM software," "top CRM platforms," and "CRM tools for sales teams."
These prompts provide a useful starting point, but buyers often ask more specific questions based on their requirements.
Someone might need a CRM for a 10-person team, while another is looking for software that integrates with Salesforce or supports complex enterprise workflows. Each scenario can produce a different set of recommendations.
Tracking prompts across buyer segments, use cases, and purchasing stages can reveal where a brand appears consistently and which competitors are recommended for particular requirements.
How are you selecting the prompts used to measure your brand's AI visibility? Are they based on actual buyer questions or a fixed keyword list?
Replies
Fixed lists work as a baseline but they drift from how people really ask. We built ours from real buyer questions (support chats and Search Console queries), then froze them and re-ran each one several times, because a single run tells you almost nothing. I work on PromoPilot, so Search Console was the natural source for us. Do you refresh the prompt list over time, or keep it frozen so the trend stays comparable?
@ksanyokm Hi
We keep the prompt set frozen for a while. It is the baseline against we measure the performance. But once we see a consistent up-trend, we revise the prompt set and start anew.
I started with a fixed prompt, which worked one 1 LLM but I found that another needed more specifics of the scenario before it revealed EM as the obvious choice and I found that another takes the words in the prompt more literally that others. So 1 would identify my page if I said 'app' but another would only classify actual playstore listings as apps. It's been an interesting exercise
@chantal_r that's a nice way to audit.
Suggest you run a fixed set of multiple prompts via multiple engines and audit.
Language is worth tracking as its own segment. For the small businesses I work with in Israel, buyers ask the same question in Hebrew or in English, and each version can cite different sources and recommend different vendors. For local markets, I'd build the prompt set in every language customers actually search in.