- Data cleaning means correcting or removing incorrect, incomplete, or inconsistent data.
- We handled null values by filling defaults or excluding invalid rows.
- Duplicate records were detected using unique business keys and removed.
- Text fields were trimmed to remove extra spaces and special characters.
- Date and numeric fields were converted into proper data types.
- In my project, invalid order statuses were mapped to standard categories.
- Outlier values like negative quantity were flagged for review.
- We also standardized naming conventions across multiple sources.
- Cleaning was done in Power Query before loading the model.
- This ensured accurate analysis and trustworthy dashboards.
What is data cleaning conceptually?
Updated on February 9, 2026
< 1 min read
