Skip to content
OrionStats
Log inStart learning
← Project Lab
PROJECT 31Intermediate

Python to Power BI

Business scenario

Churn scores are produced in a Python notebook and pasted into Excel for the BI team.

Problem statement

Create a reproducible Python step that produces a clean churn-risk table and connect it to a Power BI report that refreshes without manual work.

Learning objectives

  • Write a robust Python transformation
  • Choose an integration method (file, database, Power BI Python source)
  • Document refresh

Instructions

Python computes per-subscriber features and a simple risk score (e.g. logistic regression). Output to a database table or Parquet/CSV in a shared location. Power BI reads it; explain gateway needs if refreshed in the Service.

Tasks

  1. Python script with logging and error handling.
  2. Output table spec.
  3. Power BI model + report.
  4. Refresh documentation.

Deliverables

  • Python script
  • PBIX
  • README

Evaluation rubric

CriterionWeightExcellentProficientDevelopingInsufficient
Python quality25%Robust, logged, reproducible.Correct and complete with minor gaps in rigour or presentation.Partly correct; important steps missing, unvalidated or unclear.Missing, incorrect, or not supported by the data.
Integration design25%Appropriate method, refresh explained.Correct and complete with minor gaps in rigour or presentation.Partly correct; important steps missing, unvalidated or unclear.Missing, incorrect, or not supported by the data.
Report20%Correct and clear.Correct and complete with minor gaps in rigour or presentation.Partly correct; important steps missing, unvalidated or unclear.Missing, incorrect, or not supported by the data.
Documentation15%Complete.Correct and complete with minor gaps in rigour or presentation.Partly correct; important steps missing, unvalidated or unclear.Missing, incorrect, or not supported by the data.
Presentation15%Concise, well-labelled outputs a manager can act on without explanation.Correct and complete with minor gaps in rigour or presentation.Partly correct; important steps missing, unvalidated or unclear.Missing, incorrect, or not supported by the data.