Analysis
Churn Prediction for a Stockholm D2C Cosmetics Brand
Overview
What this challenge is about.
You are a data science consultant hired by NordicGlow. Using the provided dataset (synthetic but realistic), you must preprocess the data, engineer features from transaction, clickstream, and support interactions, and train a classification model (e.g., logistic regression, random forest, or XGBoost) to predict churn. Success is defined by achieving an AUC-ROC above 0.80 on a held-out test set, and delivering a clear set of actionable insights (e.g., top 3 drivers of churn, recommended discount thresholds). You must also provide a short slide deck explaining your approach and results to the non-technical CEO.
The Brief
What you'll do, and what you'll demonstrate.
Predict which customers are likely to churn in the next 30 days and recommend cost-effective retention strategies.
Earning criteria — what you'll demonstrate
- Apply data preprocessing techniques to real-world messy data
- Engineer features from multiple data sources (transactional, behavioral, support)
- Train and evaluate a classification model using appropriate metrics (AUC-ROC, precision-recall)
- Interpret model coefficients or feature importances to derive business insights
- Communicate technical results to a non-technical audience
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
Career mappings coming soon.