What problem does it solve? Product teams often track signups but lack a structured way to diagnose why users stop coming back. This Skill turns raw retention data (D1, D7, D30 rates), churn interviews, and analytics into a diagnosis of where users drop off, what behaviors separate retained from churned users, and which interventions to test. ## Core Features & Use Cases - Retention Curve Diagnosis: Classifies your curve as flattening, declining, or smiling, and pinpoints the biggest drop-off (D1→D7 activation problem vs D7→D30 habit problem). - Cohort & Segment Analysis: Compares retention across time-based, feature-based, and channel cohorts, with industry benchmarks for B2B SaaS, consumer, productivity, and marketplace products. - Churn Prevention & Win-Back: Provides early churn signal thresholds, a churn risk scoring model, and a win-back playbook with resurrection cohort tracking. - Use Case: A PM notices D30 retention fell from 32% to 24%. The Skill pulls context from metrics, research, and meeting notes, identifies that churned users never invited teammates, and produces a prioritized hypothesis list saved to outputs/analyses/. ## Quick Start Invoke /retention-analysis and provide your D1, D7, D14, and D30 retention rates along with your product's expected usage frequency and any known churn reasons.
