Pre-upload checklist

Amazon flat-file upload checklist before resubmission

Review workbook structure, seller SKUs, variation relationships, required values, and the prior processing report before an Amazon flat-file resubmission.

Reviewed by the SKUFix product teamLast reviewed 2026-07-23

Preserve the evidence before editing

Save the exact workbook Amazon processed and its report as a matched pair. Make repairs in a separate copy so every change can be compared with the source and reversed if the next submission produces different evidence.

Confirm that the file is an unencrypted XLSX workbook for the intended marketplace and product type. Do not remove instruction, definition, valid-value, or hidden sheets simply to make the file smaller.

Check the table and seller SKU identity

Verify the likely data sheet, header row, required headers, and duplicate columns. Every sellable row should have the intended seller SKU, and duplicate or whitespace-only variants should be reviewed before submission.

  • One clear data table and header row are detectable.
  • Seller SKUs are present, unique where required, and intentionally cased.
  • Required values are not blank.
  • Values do not contain accidental leading or trailing whitespace.

Validate parent-child variation families together

Every child should reference an existing parent in the same submission when the template requires it. Parent and child rows should agree on relationship fields and the category's allowed variation theme. Child variation values should be distinct within a family.

Do not select a theme because it looks similar to one used in another category. Use the exact values exposed by the current product-type template.

  • Every referenced parent SKU exists.
  • No row points to itself and no relationship cycle exists.
  • Themes and relationship types are consistent within the family.
  • Child variation values are complete and not duplicated.

Review allowed values and the previous failure

Use the template's definitions and valid-values sheets for enumerated attributes. Map every prior report finding to its strongest workbook evidence and resolve the shared cause before changing rows one by one.

After repairs, reopen the output, verify each intended cell, and compare the source and output for unrelated edits. Keep a human-readable change list with the file you plan to upload.

Questions sellers ask

Will a pre-upload check guarantee Amazon accepts the file?

No. It can catch deterministic workbook problems, but Amazon-side catalog, eligibility, brand, identifier, and policy decisions can still affect acceptance.

Should I copy rows into a fresh template?

Use a current template when the old workbook lacks required structure or allowed values. Copy deliberately and verify the mapping instead of pasting all columns without review.

Can SKUFix check a workbook without changing it?

Yes. The free diagnostic reads the workbook, reports structural and row-level findings, and leaves the original unchanged.

Sources checked

These primary references provide error and processing context. SKUFix independently defines its workbook checks and repair boundaries.