Fix garbled CSV text without guessing the original words

Why does this happen?

A CSV may not identify its encoding. If the original bytes remain, reading them as the correct encoding can reveal the intended text. Simply re-saving already damaged text as UTF-8 cannot restore missing characters.

Follow these steps

1. Keep a copy of the original file before opening and saving it in another program.
2. Choose the file. The tool checks its format first, then UTF-8, UTF-16 with BOM and Korean CP949. Review the suggested reading in the preview. If it is uncertain or looks wrong, open “Text or columns look wrong?” and select the encoding yourself.
3. Review the table and export Excel text cells or UTF-8 CSV with BOM.

Fictional example

Original bytes C3 A9 → UTF-8 → é

What this can and cannot do

More than one encoding may decode the same bytes. Successful decoding is not proof of the intended language. This tool cannot recover characters overwritten with question marks.

Check names, headings and representative values in the downloaded copy against the source.