SPARKPOND / GUIDE

How to Fix UTF-8 and Windows-1252 CSV Encoding

Garbled text appears when bytes are decoded with a different character encoding from the one used to create the file.

Problem symptoms

  • é appears as é.
  • Curly quotes appear as ’.
  • Black diamonds or replacement characters appear.

Why it happens

  • UTF-8 bytes were read as Windows-1252.
  • A legacy file lacks a BOM or encoding declaration.

Example before and after

Before

André,Montréal

After

André,Montréal
Download example ↓

Procedure

  1. Keep the original bytes unchanged.
  2. Compare UTF-8, Windows-1252 and Latin-1 previews.
  3. Choose the preview with valid language and fewer replacement characters.
  4. Export UTF-8, adding BOM only when the destination requires it.
Fix Garbled CSV Characters and Encoding →

Common mistakes

  • Repeatedly opening and saving the garbled file.
  • Assuming detection is certain without a BOM.

Other possible solutions

  • Re-export from the source system with an explicit UTF-8 option.

Important limitations

  • Characters already replaced by question marks cannot be reconstructed.

Related guides

Why a CSV Opens in One Column and How to Fix It

A CSV usually opens in one column because the file delimiter differs from the delimiter expected by the spreadsheet application.

CSV Comma vs Semicolon: Which Delimiter Should You Use?

Both formats are valid delimited text; compatibility with the destination matters more than the extension.