Amazon flat files

Duplicate SKUs: the flat file error you never get

Most flat file problems announce themselves with a code in the processing report. This one doesn't. The upload succeeds, the report is clean, and the wrong price goes live.

What happens when a SKU appears twice

A flat file is processed row by row. If the same seller SKU appears on more than one row, each row is applied in turn — so the last one wins. Not the correct one. Not the newest. The one that happens to sit lower in the spreadsheet.

There's no rejection because, from Amazon's side, nothing is wrong: you submitted two updates for one SKU and both were valid. The processing report says the rows succeeded, because they did.

Row	SKU	Price	Quantity	Result
18	TEE-BLACK-M	24.99	40	Applied, then overwritten
112	TEE-BLACK-M	9.99	0	This is what goes live

You find out from a sales report, a margin that looks wrong, or a customer buying at a price you retired six months ago.

Where the duplicates come from

Almost never from typing the same SKU twice on purpose. The usual sources:

Merged exports. Two category files, or last month's file plus this month's additions, pasted into one sheet.
Re-exported data with an overlapping date range, so a subset of SKUs appears in both halves.
A supplier feed appended to your own rows, where the supplier uses your SKU convention.
Copy-paste while building variations — duplicating a child row to make a sibling and forgetting to change the SKU.
Case and whitespace. TEE-BLACK-M and tee-black-m may resolve to the same SKU while looking different to you in the sheet.

How to find them before uploading

In a spreadsheet

Add a helper column next to your SKU column. In Excel or Google Sheets, assuming SKUs are in column A starting at row 2:

=COUNTIF(A:A, A2)

Fill it down and filter for values greater than 1. Anything that shows up is a SKU submitted more than once.

To catch the whitespace and case variants that a plain COUNTIF misses, compare normalised values instead:

=COUNTIF(B:B, B2) where column B is =LOWER(TRIM(A2))

What to do with what you find

Don't just delete the extra rows — decide which one is correct first. The whole risk here is that the row you keep is the stale one. Check the price and quantity on each copy against what you actually intend to sell today, keep that row, and delete the rest.

Why this one is worth checking every time

Errors with codes get fixed, because Amazon tells you about them. A rejected upload is annoying but it's visible: you read the report, correct the file, upload again.

A duplicate SKU is the opposite. It costs nothing at upload time and shows up later as a margin problem, a stockout you can't explain, or a listing that quietly went to zero quantity. The gap between the mistake and the symptom is where the money goes.

Amazon Flat File Preflight

Reads your flat file and reports every SKU that appears more than once, with the row numbers, so you can decide which copy is the right one. It also checks parent–child relationships, variation theme mismatches, and prices or quantities that look wrong.

It runs in your browser: the file is never uploaded anywhere. Free for the first 30 rows, no account. One payment for larger files, no subscription.

See how it works

Amazon's processing behaviour and category requirements change without notice. Always read the processing report for your specific upload. This guide describes what happens when one SKU appears on several rows of a single file; it does not cover Product ID conflicts with existing ASINs (errors 8541 and 8542), which depend on Amazon's catalog rather than on your file, and nothing here guarantees Amazon will accept an upload.