Model Online / On-Site
Duration
Modules 5 modules

Course Objective & Scope

Knowing who will leave is not enough; the real question is whom to contact, with which offer, to change the outcome. The programme covers churn definition in subscription and non-subscription businesses, features from RFM and usage behaviour, propensity models with gradient boosting, survival analysis to predict time to churn, and uplift modelling to find customers whose behaviour a campaign will actually change.

Model outputs are tied to business metrics through lift, campaign ROI and control-group design. Participants prepare a targeting list and test plan on a case adapted to their customer base.

Who Should Attend?

  • Customer analytics and CRM teams
  • Marketing analytics and campaign management specialists
  • Data scientists in telecom, banking, insurance and retail
  • Prerequisite: Python and classification modelling experience

Key Takeaways

  • Define churn and propensity targets that fit the business model
  • Engineer meaningful features from customer behaviour
  • Predict time to churn with survival models
  • Identify campaign-responsive customers with uplift modelling
  • Measure model impact with control-group campaign tests

Curriculum

Churn and Propensity Target Definition
  • Contractual and non-contractual churn definitions
  • Prediction horizon and label window
Customer Behaviour Features
  • RFM, usage trends, complaint and engagement signals
  • Customer lifecycle and value (CLV) metrics
Propensity Models
  • Churn, purchase and cross-sell scores with LightGBM
  • Lift, gain and decile analysis
Survival Analysis
  • Kaplan-Meier and Cox models (lifelines)
  • Timing interventions by predicted time to churn
Uplift Modelling and Campaign Testing
  • T-learner, X-learner and CausalML
  • Control-group design and campaign ROI measurement

Other Courses in This Category

  • Recommendation Systems

    Collaborative filtering, content-based and hybrid recommenders, matrix factorisation, implicit feedback, cold-start strategies and ranking metrics for building personalised product and content recommendation engines.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Predictive Maintenance with Machine Learning

    Failure probability and remaining useful life (RUL) prediction from sensor, telemetry and maintenance logs, with hands-on signal feature engineering, health indices, early-warning thresholds and maintenance cost optimisation.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Demand Forecasting for Supply Chain and Retail

    SKU-location demand forecasting covering promotions, price elasticity, stockout censoring adjustment, intermittent demand (Croston/ADIDA), hierarchical reconciliation and probabilistic forecasts tied to safety stock.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Optimization and Decision Systems

    Bridging predictive models and mathematical optimisation: LP/MILP formulation, capacity, scheduling, routing and budget/price allocation with OR-Tools and Pyomo, plus stochastic and contextual decision systems under uncertainty.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Model Development with XGBoost and LightGBM

    Mastering tabular gradient boosting with XGBoost, LightGBM and CatBoost: tree-growth strategies, native categorical and missing-value handling, ranking/quantile/custom objectives, monotonic constraints, Optuna and fast inference.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View
  • Automated Machine Learning (AutoML)

    Automated model selection, hyperparameter search and multi-layer stacking with AutoGluon, FLAML and Optuna, covering compute budget management, leakage checks and moving from rapid baselines to production.

    Machine Learning & Applied Data Science Corporate
    Live Course
    5 modules
    View

Get in touch