ship-score

# SHIP Score Conduct an in-depth interview to score product features using the SHIP framework. Goal: reach **90-95% confidence** on each dimension before proposing scores. ## CRITICAL: Read the Framework First **BEFORE doing anything else, read [the SHIP framework](references/ship_framework.md).** This is the canonical source for all scoring definitions. Read it at the start of every scoring session and keep it in context throughout. Do not rely on memory or summaries — the exact definitions, score tiers, common traps, and key questions matter. ## Process 1. **Read [references/ship_framework.md](references/ship_framework.md)** — mandatory first step, every session 2. Ask the user what feature, bug, or idea they want to score (or let them paste a description) 3. Display feature context as an index card 4. Interview for each dimension until 90-95% confident 5. Propose scores with reasoning 6. Confirm with user, adjust if needed 7. Present final scorecard ## Interview Principles - **Start at 0% confidence** for every feature, every dimension - **Push back on vague answers** — "could help" is not validated demand - **Challenge assumptions** — "I think customers want this" needs evidence - **Track confidence explicitly** throughout — say where you are ("that gets me to ~60%...") - **Don't settle for gut feelings** — dig for specific examples, data, customer names - **Reference the framework definitions** when explaining why a score lands where it does - **Use the Key Quest...