⚙️Data Transformation
Also known as Transformation, Data Transform
Raw data is changed into the shape a report needs.
Think of it like
It is like making soup from a bag of shopping. You wash, cut and mix until it is ready to eat.
Example
Raw orders hold prices as text and dates in three formats. A transformation fixes both, then adds up sales by day.
Why it matters
Raw data is rarely ready to use. Most of a data team's time goes into this step.
Definition
Common mistake
Key takeaways
- 1Raw in, shaped out. Fix, join, filter and sum.
- 2The T in ETL. Often SQL in the warehouse.
- 3Small tested steps. Catch a bad change early.