How to use this office tool
- Choose a CSV or TSV file and confirm the header row.
- Run the inference and review type, confidence, blank count, and examples.
- Download the XLSX profile before changing any source schema.
Profile a CSV or TSV file and infer whether each column mainly contains text, numbers, dates, booleans, blanks, or mixed values—all inside your browser.
This tool samples every populated value, classifies it with conservative rules, and creates an XLSX report with inferred types, confidence, null counts, and examples.
Type inference is a recommendation. Identifiers, leading-zero codes, and ambiguous dates should stay as text unless you confirm otherwise.
Files are processed locally. XLSX and XLS workbooks are read with SheetJS; CSV and TSV files are parsed with delimiter-aware handling. Advanced Excel features may not be preserved in generated files.
| Format | Support |
|---|---|
| XLSX / XLS | Read workbook values and formulas |
| CSV / TSV | Read delimited rows |
| Output | New downloadable file or report |
Each nonblank value is tested against conservative boolean, number, and ISO-style date patterns, then the dominant type and confidence are reported.
Values with meaningful leading zeros are classified as text so codes such as 00123 are not silently shortened.
It profiles all parsed rows up to the page safety limit rather than relying only on the visible preview.
A mixed column contains substantial values from more than one type, so automatic conversion could be risky.
Yes. TSV input is parsed with a tab delimiter and receives the same column profiling report.
| Stage | Action |
|---|---|
| Input | One CSV or TSV file |
| Privacy | Local browser processing |
| Result | New downloadable output |