How it is calculated
The tool inspects the raw bytes in order:
- Byte-order mark —
EF BB BFmeans UTF-8 with BOM;FF FE/FE FFmean UTF-16. With a BOM present, Excel picks the right encoding on its own. - Strict UTF-8 decode — without a BOM, every byte sequence is checked against UTF-8 rules. If nothing violates them, the file is UTF-8 without BOM: the classic "exported from a web app, broken in Excel" case.
- Legacy fallbacks — if it is not UTF-8, the tool tries CP949 (Korean Windows) and finally Windows-1252.
- Rewrite — the text is re-encoded as UTF-8 and a BOM is prepended, so Excel, Google Sheets and web importers all read the same characters.
The delimiter (comma, tab, semicolon) is auto-detected by PapaParse from the first rows. Cell values are never altered — numbers, dates and quotes stay as they were.
Examples
| Situation | Detection | Download |
|---|---|---|
Shopify order export shows José in Excel | UTF-8 (no BOM) | CSV (UTF-8 BOM) → opens correctly |
| CSV saved from Excel breaks accents when uploaded to a web app | Windows-1252 | CSV (UTF-8 BOM) → works in both |
Order ID 202609060001234 displays as 2.02609E+14 | any | .xlsx (all cells stored as text) |
Common mistakes
- Re-saving in Excel undoes the fix if you choose "CSV (Comma delimited)". Pick "CSV UTF-8 (Comma delimited)" instead.
- This does not repair characters already lost. If a previous save replaced letters with
?, that information is gone — re-export the original file and convert that. - All data in one column is a delimiter/locale issue (semicolon vs comma), not encoding. The preview will show a single column in that case.
- Files containing customer data are processed locally, but where you store or send the converted file remains your responsibility under privacy law.
FAQ
Is my file uploaded anywhere?
No. The file is read and converted entirely in your browser tab. Open the Network tab in developer tools and you will see no upload request. Close the page and the data is gone.
Why does Excel show é or ’ instead of é or ’?
Excel on Windows opens a CSV without a byte-order mark (BOM) using the system code page (Windows-1252 in most English locales), while most web exports are UTF-8 without BOM. The bytes are fine; the two programs simply disagree on how to read them.
Should I download the CSV (UTF-8 BOM) or the .xlsx?
Choose CSV if you will import the file into another system (shipping carrier, accounting software). Choose .xlsx if a person will open it in Excel — it also stores every cell as text, so order numbers do not turn into 1.23E+11 and leading zeros in ZIP codes and phone numbers survive.
Can it convert UTF-8 back to a legacy code page?
Not yet. Save as .xlsx here, then in Excel use Save As → "CSV (Comma delimited)", which writes the legacy encoding your locale uses.
What is the size limit?
50 MB. Everything runs in browser memory, so split larger exports first. There is no row limit; the preview shows the first 20 rows.