GLOSSARY
Prompt set
- Prompt set
- A prompt set is the fixed basket of buyer-intent questions you measure AI visibility against — the same prompts, run on the same schedule, across the same engines. Holding it fixed is what turns 'are we visible in AI' from an anecdote into a trend line of mention rate and citation share.
A good prompt set is built from how your buyers actually ask, not how you'd describe yourself. For a personal-injury firm that means prompts like 'should I get a lawyer for a car accident that wasn't my fault' alongside 'best injury lawyer near me' — problem-first phrasing, brand names of engines' users' vocabulary, constraints included. Typical engagements track 25–50 prompts spanning the funnel: problem questions, comparison questions, 'best X for Y' selection questions, and branded prompts about you and named competitors.
Fixedness is the methodological spine. Because engine outputs vary run to run, the only way to see real movement is to hold everything else constant: same prompts, same cadence, same engines, logged the same way. Each run produces mention rate and citation share per engine; month-over-month deltas on a fixed basket are evidence, while a shifting basket produces numbers that can't be compared to anything — including, conveniently for some vendors, last month's.
Example: a dental practice tracks 30 prompts across six engines. Baseline: named in 4. After a quarter of answer-page and directory work: named in 11, with the report showing exactly which prompts moved and which didn't. That prompt-level honesty — including the 19 that didn't move — is what separates measurement from marketing. Building the set is step one of every engagement; see how we measure.