A compact visual aid for explaining data science under interview pressure.

This site was built as a backup visual aid for the Bertelsmann Future Leaders Program assessment center on May 28-29, 2026. It is designed to run locally from a static build, with no runtime network calls.

The goal is not to teach every detail of data science. The goal is to make the shape of a responsible data project visible in a few minutes.

CRISP-DM stands for Cross-Industry Standard Process for Data Mining. It was originally published in 1996 through the IBM/SPSS lineage and remains one of the most widely recognized process frameworks for analytics and data science work.

The Bertelsmann mini-cases are illustrative scenarios using publicly known business units. They are not claims about real internal projects.

The churn scorer is a hand-crafted formula for explanation, not a trained model. It is deliberately labeled that way because the point is to show modeling intuition, not pretend to have trained on private data.

Built with Astro, Svelte, Tailwind CSS, D3 shape utilities, and static bundled data.