Your prompt library is the set of buyer-style questions every benchmark asks. It’s the measuring stick — so it’s built to be realistic, stable, and yours to curate.

Grounded in real demand

Prompts aren’t synthetic templates. Generation starts from real Google search queries in your category, with monthly search volume attached, and turns them into the questions a buyer would ask an AI assistant: “best X for Y”, “compare A vs B”, “does A support Z”. Each prompt records why it was generated — the keyword, feature, or persona behind it — so you can always trace a question back to the demand it represents.

The mix leans deliberately competitive: most prompts put you in an evaluation context against named competitors, because that’s where deals are won and lost. The rest are open landscape questions that measure organic visibility — whether AI brings you up unprompted.

Persistent, not regenerated

The library persists across runs. Every benchmark snapshots the currently active prompts and asks the same questions as the last one, which is what makes scores comparable over time — movement in your numbers reflects changes in AI’s answers, not changes in the questionnaire.

You can add, edit, archive, and re-generate prompts between runs. Archived prompts stop being benchmarked but keep their history; new prompts start contributing from the next run.

Per-subject libraries

Each subject — your brand and each product — has its own library, generated from its own category and keywords. A benchmark for a product runs that product’s prompts against that product’s competitors.

Curating well

• Volume beats cleverness. A prompt tied to a query real buyers search monthly is worth ten hypothetical ones.
• Keep the set stable. Resist rewording prompts between runs; every edit resets that prompt’s comparability.
• Prune what you’d never act on. If a prompt’s outcome wouldn’t change what you write or fix, archive it — it’s diluting the questions that matter.