Sort the data and look for values far from the rest
Compare each value to the median and quartiles
Use the IQR rule: values below Q1 – 1.5×IQR or above Q3 + 1.5×IQR
Check z-scores: values with large absolute z-scores may be outliers
Plot the data with a box plot
Plot the data with a scatter plot or line chart
Look for sudden jumps or isolated points
Compare values against known business or domain limits
Check for values that are impossible or highly unlikely
Review data entry errors, duplicates, and missing context
Compare groups separately instead of only using the full dataset
Use robust statistics when the data is skewed
Flag values that differ greatly from nearby observations
Validate suspicious points with source records