Free tool · nothing uploaded
Drop a spreadsheet export and find out what is actually wrong with it before it breaks an import: ragged rows, duplicate records, numbers stored as text, two date formats in one column, and text mangled by the wrong encoding.
Your file never leaves this tab. Everything below is computed by JavaScript on your own machine. There is no server to upload to, so nothing is stored, logged, or sent anywhere.
…or drag one anywhere onto this box. .csv, .tsv
and .txt up to about 50 MB.
Nothing to hand?
Inferred type, how empty it is, and how varied the values are.
If this does not look like your data, the delimiter or quoting is the first thing to check.
The cleaned file is written as UTF-8 with a BOM
so Excel opens accented characters correctly, using , as the delimiter and
quoting only where needed. Duplicate column names get a numeric suffix rather than
being dropped. A percent sign is removed but the value is not divided
by 100 — 5% becomes 5, not 0.05, because which
one you want depends on the column. Keep your original: this tool cannot know which of
these values were deliberate.
It will not guess a date order that genuinely cannot be resolved.
If a column contains only days 1–12, 03/04/2026 is either 3 April or
4 March and no amount of parsing settles it — that column gets flagged and left
alone rather than silently converted. Half of all "the dates are wrong" bugs start
with a tool that guessed here.
It reads .csv, .tsv and delimited
.txt. It does not open .xlsx — export to CSV first, or
see the PDF→Excel demo for the workbook side of this.