Purpose
Deterministic value ranking (#2.1) with the promo multiplier (#2.2) and the par-based suggested order (#4.1).
When to use
After gap-score in every run.
Inputs
scored.json + scan extract + config/perfect-store-standards.json.
Steps
- Run scripts/gap_rank.py --scored scored.json --scan scan.json --standards perfect-store-standards.json --out ranked.json
- Quote the ranking verbatim - two fails on hero SKUs outrank five on tail; a promo-week empty display is the most expensive kind of empty (#2.2). Say it with the numbers.
- The suggested order is at par quantities; the REP confirms and places it - never transmitted by the plugin (#4.1).
Output
ranked.json - the {ranked_gaps, suggested_order} hop.
Grounding requirements
Every gap cites its scan line and standard criterion.
Constraints
- Ranking is engine-only; a high compliance % never waves off a high-value gap.
Escalation / uncertainty
Missing scan data degrades to criterion order with an explicit note (#5).