Cleaning Data

Frameworks/Thinking

C.L.E.A.N

• Conceptualize: Understand the problem.

  • understand what data is for

  • helps you prioritize what to fix, what to optimize it for

  • Identify the grain, measures, and dimensions

    • what each row

  • identify critical vs non-critical columns

  • understand definitions

• Locate Solvable Issues: Find fixable problems.

• Evaluate Unsolvable Issues: Identify issues you can’t fix.

• Augment & Improve: Clean and enhance the data.

• Note & Document: Record what you did and learned.