Editing a CSV sounds trivial — it is just text — but it is also the step where most files get corrupted. A dropped comma, a spreadsheet auto-format, or an accidental re-sort can break an import. Here is how to edit CSV files safely at every scale.
Before you start: keep the original
Always keep an untouched copy of the original file. If an edit goes wrong, you can go back. This single habit prevents most CSV disasters.
Editing a few cells: spreadsheet
For small changes, open the CSV in Excel or Google Sheets, click the cell, type the new value, and save. It is fast and visual. The risks are the same as when creating:
- Leading zeros disappear from SKUs (
00123→123). - Dates and numbers get reformatted to the spreadsheet's locale.
- Saving may change the encoding — choose CSV UTF-8 when saving.
Editing many rows: find and replace
For uniform changes (rename a category, fix a typo across 500 rows), find-and-replace in a text editor works well. Two cautions:
- Replace whole values, not substrings — replacing "old" inside a product name will also hit words containing "old".
- If your search term contains a comma or quote, you are editing structure, not just data. Stop and use a proper tool instead.
Editing with rules: describe the change
The most error-prone bulk edits are the ones with logic: "increase all prices by 10%", "set inventory to 0 for everything in the Sale category", "add 'Organic' to every product name in the Beverages category". Doing this by hand across hundreds of rows is slow and inconsistent.
This is where CSV Pilot helps: upload the file and describe the change in the chat — for example "make all products 15% cheaper" — and the agent applies it to the matching products. Numeric changes are computed exactly in code, not guessed, and every result is shown in a preview grid before you export.
Editing the structure itself
Sometimes the change is structural: adding a column, renaming a header, splitting one field into two. Rules of thumb:
- Renaming headers can break imports that map by column name — check the target format first.
- Adding a column means adding a value (even if empty) to every row.
- Re-sorting rows is usually safe for product data, but re-sorting columns is not — keep the header order aligned with the values.
After editing: validate
Before importing an edited file, check:
- Every row still has the same number of values as the header.
- Numbers are plain (no currency symbols or thousand separators).
- No cells accidentally start with
=,+,-or@(formula injection risk). - The file is still UTF-8 encoded.
A structured editor with validation — like CSV Pilot's preview — flags these problems automatically instead of leaving you to spot them in the import error log.