Advanced RFM/RFE Segmentation, Jenks Natural Breaks & State Migration Analytics

This skill exists to stop: segmenting customers with arbitrary equal-split thresholds — producing non-homogeneous groups and mistargeted campaigns.

📁 Source note: [sage] = upstream Sage repo (github.com/xoai/sage, public) — optional deeper reading; this skill runs fully on the rules inlined here. A step marked MUST READ points at a file in your own project (e.g. an event registry) — if it is missing, stop and ask instead of improvising.

🤖 0. HOW TO USE (agent workflow)

A. Compute RFM/RFE with Jenks natural breaks — report before/after variance to prove thresholds aren't arbitrary. B. H1→H2 migration matrix: name the most valuable migration flows (who's falling, who's rising) + one action per flow. C. 5-second executive dashboard: one number, one action per tile; every number traceable to its source SQL. Standard output: segment → size → value → action → owner. A segment with no action gets cut from the report.

🔬 1. Variance Optimization: Jenks Natural Breaks Classification

The Problem with Equal Split (NTILE(5))

Financial and activity data can be highly skewed. Measure the actual concentration before choosing thresholds; no fixed customer share universally produces a fixed revenue share. Equal-count bins can group very different values together. Compare Jenks with that baseline on the supplied data.