Clean a messy CSV, in your browser.
Duplicate rows, stray spaces, dates in four formats, European decimal commas, headers nobody can use. Drop the file in and take the clean one back.
Your file never leaves your computer · No sign-up, no email, no limit on size · Works offline once this page has loaded
What is in it
What to fix
Why numbers are the part that goes wrong
Everything else on this page is tidying. This one changes what your totals say.
1.234 is one thousand two hundred and thirty-four in
Portugal, Spain, Germany, Italy and Brazil, and one point two three four
in the United States and the UK. A spreadsheet that reads it the wrong
way is out by a factor of a thousand, in one cell, with no error
anywhere.
So this does not guess per value. It reads the whole column first, looks for a value that settles it — one carrying both a dot and a comma, or the same separator twice — and applies that answer to every value in that column. When the column never says, it leaves the numbers alone and tells you how many. The long version, with the measurements.
When this page is not enough
It handles the common shape of the problem. What it cannot do is anything that needs a judgement about your data: merging “Ana Silva” with “Ana Sliva”, splitting one column into three, matching rows against a second file, pulling a date out of a free-text note, or any of it on a schedule rather than once.
That is the paid version
49€ one-time
Send the file, say what “clean” means for it, and get back a script written for that file — yours to keep and re-run on next month’s export. Working code within 48 business hours, or a full refund.
Buy — Stripe checkout what you get
If this free page already solved it, keep it and pay nothing. It is not a trial and there is nothing to cancel.
The
command-line version is free too, and open source — it also reads
.xlsx directly.