north-star-designerlisted

# North Star Designer A product in, a metric tree out — rooted in a number that leads value, not one that reports it after the fact. ## Verification gates (defined first; output is blocked until all pass) - **G1 — Leading, with the check stated:** the proposed NSM passes the causality test, written out in the output: *when this number moves this week, revenue/retention moves later — and the reverse is not true*, plus what a team can do this week to move it. Revenue, churn, NPS, and their composites are lagging — proposing one as NSM fails the gate. - **G2 — Tree is actionable:** every input metric maps to the team or lever that moves it. An input metric no one owns fails. - **G3 — Gaming named:** the output names at least one way the NSM can be gamed and pairs a guardrail metric against it. No invented correlations or benchmark values — the tree builds on stated facts, and assumed links are labeled assumptions. ## Steps 1. **Locate the value moment.** What event means the user got what they came for (meeting successfully scheduled, document shipped, ticket resolved)? The NSM lives at or next to this moment — value delivered, countable, weekly-moveable. 2. **Run the causality test on candidates.** For each candidate: does moving it move revenue later (leading), or does revenue move it (lagging)? Write the test result for the winner — this sentence is the gate's evidence. If the user currently tracks a lagging NSM, reject it explicitly with the classification and reason. 3