Removing rows from the cleaned CSV
Some rows can't be cleaned automatically. The data is too damaged or too ambiguous for CSVwise's auto-fixes to handle safely. For those rows you have two options: fix the source data yourself, or remove the row so it never reaches the cleaned CSV.
When to remove a row
Remove a row when:
- The product is no longer relevant (discontinued, never going to be imported).
- The row is corrupted (cut off mid-field from a bad spreadsheet save).
- The row is missing data in a way that can't be reconstructed (no Title and no Handle, or a price that's clearly garbage like
"see vendor"). - You'd rather get the rest of the file cleaned now and handle the removed rows separately later.
When NOT to remove:
- The row has a typo you can fix manually. Open the source CSV in Excel or Numbers, correct the cell, re-upload. That's faster than figuring out why CSVwise can't auto-fix it.
- You're tempted to remove because there are "too many errors." That's often a column-matching issue (Stage 2) where one mis-matched column produces hundreds of false-positive errors. Go back to Stage 2 and re-check your column matches first.
How to remove a row
On the Fix issues stage, every error card that applies to specific rows carries a Remove all N affected rows from the cleaned CSV button. Store-match cards carry Remove these N rows from the cleaned CSV.
To clear everything CSVwise can't fix in one go, use Remove all N error rows from the cleaned CSV, the action shown above the error cards in the Errors section. Rows that match products already in your store are deliberately left out of that set.
You can also remove a single row from the live preview. Each preview row leads with its source row number (#142) and a Remove button. Once removed, the same button switches to Restore.
If a removal would mark most of your file for removal, CSVwise asks you to confirm first.
Removed rows stay in the preview, struck through, so you can see exactly what's being left out.
Restoring rows
Click Restore on the row in the live preview to bring a single row back.
To undo everything at once, use the banner at the top of the problem list. As soon as any row is marked for removal it reads "N rows marked for removal" and offers Restore all rows.
What ends up in the download
The cleaned CSV contains only the rows you kept. Removed rows are simply not written to the output file. There's no marker, no blank line, no metadata showing they were ever there.
This is on purpose. Shopify's importer doesn't have a concept of a removed row. The cleanest output is a CSV that contains only what you intend to import.
Confirming before download
Before the download runs, CSVwise shows a confirmation: "Remove N rows and download?" It spells out how many rows are being excluded and how many will be cleaned and exported. Your source file is never modified, so you can always re-upload and start over.
CSVwise also records the number of rows you removed as part of the cleanup, for its own stats. It stores the count only, never the contents of the removed rows.
Common mistakes
- Removing rows to dodge a warning that doesn't actually need action. Warnings never block your download. Removal should be saved for actual unrecoverable damage.
- Removing the wrong row. If your CSV came from a system where row order changed during export, the row numbers in CSVwise might not match the row numbers in your source spreadsheet. Use the preview to verify the actual data of the row before removing it.