Predictive analytics · Python

Turning churn predictions into business decisions.

Can a churn model produce a useful retention strategy instead of another probability score?

Role
Analysis, feature engineering, and model evaluation
Format
Independent project
Tools
Python · Pandas · scikit-learn · LightGBM · PyCaret
  • 10,000Customers analysed
  • ~20%Churn rate
  • 6Models compared
  • 0.63Best F1 score

The problem

A churn score only matters if it changes an action.

I analyzed 10,000 retail-banking customers with a churn rate of roughly 20%. The project connected model performance to specific retention actions instead of stopping at a ranked model table.

The business needed enough recall to find at-risk customers, but low precision would waste outreach. No single metric could decide that trade-off without the campaign context.

The approach · 4 steps

Compare models, then connect signals to retention.

  1. Explored churn patterns across age, geography, balance, activity, and product use.
  2. Compared six classification models with five-fold cross-validation.
  3. Created behavioural features such as product usage, engagement, and balance-to-salary ratio.
  4. Used PyCaret as a second check on the manually built pipeline.

The result

The output was a ranked intervention plan.

  • 0.63Random Forest F1
  • 0.68Recall on the test set
  • 6Models compared

Age, number of products, activity, and balance were the strongest signals. The practical output was a set of targeted retention ideas and a plan for ranking customers by intervention priority.

What I would do next

  • Tune the threshold around campaign cost and retention value.
  • Add recent activity trends instead of relying on a static snapshot.
  • Track whether outreach changes churn, not only whether the model predicts it.