Focused purpose
What this tool does
Profile CSV or TSV data locally with row, column, type, missing, distinct, duplicate and structural metrics. Review invalid examples without an account.
- ✓ Row and column counts
- ✓ Probable types and confidence
- ✓ Missing and distinct values
- ✓ Duplicate and structural issue detection
Workflow
How to use it
- 01
Open the file and confirm how it is parsed.
- 02
Review overview metrics, then inspect each column.
- 03
Export the report or open a focused cleaning tool.
Available controls
Available options
- Minimum and maximum text length.
- Representative invalid examples.
- Exact and key-based duplicate counts.
- Read-only profile before suggested fixes.
Small quality audit
Example before and after
A short file hides a missing email, a duplicated identifier and a type mismatch.
Before
id,email,age 001,ada@example.com,36 002,,unknown 002,ada@example.com,36
After
Rows: 3 Columns: 3 Missing email: 1 Duplicate id: 002 Invalid age example: unknown
Expected result: The profile identifies each issue without changing any row.
Practical scenarios
Common use cases
Audit a new data source.
Estimate cleanup work before migration.
Document a dataset for a teammate.
HONEST LIMITS
Important limitations
- iProbable types are evidence, not a database schema.
- iDistinct counts may use bounded sampling for very large files.
FAQ
Questions about this tool
Does profiling change my CSV?
No. Profiling is read-only until you explicitly add a suggested operation.
How is a probable column type chosen?
The profiler compares non-empty values against text, number, boolean and date patterns and reports confidence.
Can I see invalid values?
Yes. Bounded representative examples are shown without sending them anywhere.