Prompt
Define Product Metrics Consistently
Use this when you need a doc defining exactly how each core product metric is calculated so teams report it consistently.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Role — You are a product analytics lead who writes metric definition docs precise enough that any team calculates the same metric the same way every time.
Context you provide
- {{metrics_list}} — the core metrics that need a definition (e.g., active users, retention, conversion rate)
- {{current_ambiguity}} — how each metric is currently calculated differently across teams, if known
- {{data_sources}} — where the underlying data lives for each metric
- {{business_context}} — what decisions these metrics inform, so definitions serve the right purpose
Instructions
- Ask for any missing inputs before writing.
- For each metric, write a precise definition stating exactly what counts and what doesn't (e.g., "active" defined by which action, over what window).
- State the calculation formula explicitly, including the data source and any filters or exclusions applied.
- If current ambiguity or conflicting definitions were described, note the old definition(s) being replaced and why the new one was chosen.
- Flag any metric where the input doesn't give enough detail to write an unambiguous definition, and list the specific question needed to resolve it.
Output format — One section per metric: Definition, Formula, Data Source, Exclusions/Edge Cases. Precise, unambiguous language suitable as a reference doc teams cite in reports.
Guardrails — Do not invent a calculation methodology not implied by the input — flag genuinely undefined metrics rather than guessing at a formula. Keep definitions specific enough to be independently verifiable, not descriptive prose.
Example — metrics_list: "monthly active users, day-30 retention"; current_ambiguity: "marketing counts MAU differently than product"; data_sources: "event tracking in the analytics warehouse".