Practical guide

CSV encoding: UTF-8, BOM and Excel

Prevent broken characters across spreadsheet apps.

What matters

UTF-8 is the safest general exchange encoding. A UTF-8 BOM can help some spreadsheet versions recognize it, while CP949 may appear in older Korean workflows. Test a small round trip before replacing a production export.

A safe workflow

Keep the source unchanged.
Test on a representative copy.
Review the output for meaning, not only file size or a green status.
Record the settings that worked.

Common mistake

Automation can make a wrong assumption repeat faster. Confirm encoding, units, schema meaning or visual quality before applying a result broadly.

Why text becomes garbled

A CSV file stores bytes, while a character encoding explains how those bytes become letters. UTF-8 is widely compatible, but older exports may use a regional encoding. Opening bytes with the wrong assumption can turn names and symbols into replacement characters.

Safer import steps

Keep the untouched source.
Use the spreadsheet application's Import command instead of double-clicking the file.
Select UTF-8 first and preview international text.
Confirm comma, semicolon or tab as the delimiter.
Mark identifier columns as text when leading zeros matter.

BOM and spreadsheet behavior

Some Windows spreadsheet versions use a UTF-8 byte-order mark as a hint. It is not part of the data, but it can improve automatic detection. Other systems handle plain UTF-8 correctly. Always verify with the receiving application.