Garbage in, garbage out. Before running any analysis, clean your data. Here's the checklist:
Structure Checks
1. Remove duplicate rows. 2. Check for consistent column headers (no 'Name' vs 'name' vs 'NAME'). 3. Verify data types (numbers stored as text?). 4. Check for merged cells if from Excel.
Missing Data
5. Count missing values per column. 6. Decide: delete rows, fill with mean/median, or flag as unknown. 7. Check if missing data is random or systematic (a whole month missing = data collection issue).
Value Checks
8. Look for outliers (a salary of $1M in a dataset of $50K-$100K). 9. Check for impossible values (negative ages, dates in the future). 10. Standardize formats (dates: 2026-03-16 vs 03/16/26 vs March 16).
Text Data
11. Trim whitespace. 12. Standardize capitalization. 13. Check for encoding issues (é instead of é). 14. Normalize abbreviations (US vs U.S. vs United States).
Final
15. Spot-check 10 random rows manually. Trust but verify.
Need to quickly inspect your data? Our CSV viewer lets you browse and search without Excel. For format conversion, try CSV to JSON or JSON to CSV.