CSV Column Profiler
About CSV Column Profiler
CSV Column Profiler summarizes the quality and shape of every field in a table: inferred data type, missing values, uniqueness, extrema and the most frequent values.
How it works
Practical type inference
Each non-empty value is classified as integer, number, boolean, ISO-style date or string. Columns with incompatible evidence are marked mixed instead of forcing a misleading type.
Completeness and cardinality
Missing and unique counts include percentages, making sparse identifiers, low-cardinality categories and accidental blanks easy to spot.
Ranges and common values
Numeric and date columns receive typed min/max values. Frequency counts expose dominant values and are capped by the selected Top setting.
Frequently asked questions
Does profiling change the CSV?
No. It only reads the table and produces a JSON report.
How are blank values counted?
Empty strings are missing; optionally, whitespace-only cells are treated as missing too.
Is the data uploaded?
No. Profiling runs entirely in your browser.
Sources & last verified
The behaviour of this tool and the claims on this page were checked against the specifications below on . Everything runs in your browser, so the output you see is exactly what the engine produced.
Found a wrong result or an outdated claim? Report an error and it will be re-verified.