{"id":1476,"slug":"informs-analytics-magazine-churn-binary-classification","title":"INFORMS Analytics Magazine — churn modelled as binary classification on RFM / engagement signals (logistic regression / decision trees / ensembles)","kind":"reference","scope":"business","status":"current","audiences":["kevin","smb-owner","candid-team"],"topics":["decision-linked-metrics","customer-retention"],"reference_body":"**Claim.** Customer churn is conventionally framed as a binary classification problem: at a given horizon, a customer either stays or leaves. The standard methods (logistic regression, decision trees, gradient-boosted ensembles) take RFM and engagement features as inputs and predict the probability of churn.\n\n**Quote.**\n> \"A customer either stays or leaves… This statistical method predicts the probability of a customer churning.\"\n\n**Source.** INFORMS *Analytics Magazine* (pubsonline.informs.org, accessed 2026-06-21).\n\n**Confidence.** Industry-consensus for the *capability*. The framing as binary classification is textbook; the methods are conventional.\n\n**Caveats.** Capability is not outcome. The model has to perform above base rates, on enough data to clear noise, and the firm has to *act on* the at-risk flags. None of those is automatic — and as [[express-analytics-volume-quality-undermine-models]] notes, data quality and availability \"can fundamentally undermine a model's reliability.\"\n\n**Implication / use.** Use to ground the retention-decision domain as a *capability* claim. Critically: do NOT pair with vendor-recycled magnitudes (5x-25x, 25-95pct) — see [[retention-magnitude-vendor-recycled-quarantine]].","rationale_body":null,"metadata":null,"links":{"outgoing":[{"slug":"research-brief-information-asymmetry-decision-edge-june-2026","title":"Research notes (capture-layer): the affirmative, inward decision-edge case for data intelligence — information asymmetry applied to pricing, demand, risk, retention, targeting (June 2026)","kind":"research-notes","scope":"business","link_type":"depends-on"}],"incoming":[{"slug":"rule-quarantine-recycled-retention-magnitudes","title":"Rule: quarantine the recycled retention magnitudes (5x-25x, 25-95pct, 80/20, 20-30pct AI churn) until primary sourced","kind":"rule","scope":"business","link_type":"depends-on"},{"slug":"express-analytics-volume-quality-undermine-models","title":"Express Analytics / INFORMS — \"data quality and availability can fundamentally undermine a model's reliability\"","kind":"reference","scope":"business","link_type":"relates-to"},{"slug":"retention-magnitude-vendor-recycled-quarantine","title":"Quarantine: \"5x-25x cheaper to retain,\" \"5pct retention → 25-95pct profit,\" \"80pct profits from 20pct customers,\" \"AI churn → 20-30pct retention improvement\" — vendor-recycled, untraced to primary","kind":"reference","scope":"business","link_type":"relates-to"}]},"created_at":"2026-06-21T01:14:48.774Z","updated_at":"2026-06-21T01:14:48.774Z"}