R/F/M scoring 1–5 splits the base into lifecycle segments from Champions to Lost; recency and monetary diverge — 'frequent but cheap' and 'rare but large'.
Situation
Needed a simple, interpretable customer-value layer to complement cluster segmentation
Task
Build a value layer that stays simple and interpretable alongside cluster segmentation
Action
R/F/M quintile scoring → lifecycle tiers, plus a heatmap of mean R/F/M by segment
Result
Champions — 23.9% of the base with all three axes high (91/93/93); At Risk 11.5% and Lost 23.2% are a large reactivation reserve
Source: github.com/NikitaBoyarkin/volta-banking — 22 projects; figures recomputed from the repo's own datasets (data/*.csv) via its analysis scripts. Funnel counts from data/volta_funnel_data.csv (10,000 users); A/B, retention, segmentation, churn, RFM, CLV, attribution, anomalies, spend, support, NPS, JTBD, unit economics, premium, KYC deep-dive, referral, assisted CAC, FX sourcing, premium offers, anchor CAC and dormant win-back follow the published project narrative (README + part pages).
RFM profiles of lifecycle segments
Mean R/F/M scores (0–100) across the seven lifecycle segments. Recency and Monetary diverge: 'frequent but cheap' and 'rare but large' are different segments.
Key takeaways
R/F/M scoring splits the base into 7 segments from Champions to Lost.
Champions (23.9% of the base) hold all three axes high (R 91, F 93, M 93), while New has high R (100) but low F/M (~30–35).
At Risk (11.5% of the base) and Lost (23.2%) need reactivation, while Champions (23.9%) need upsell: one offer does not fit all.