How to upload a bulk sheet to Amazon

The other half of the workflow

You've downloaded a sheet, made your changes, and saved the file. Now you need to upload it back to Amazon for the changes to actually apply. This article walks through the upload process: what Amazon does when you upload, how to read the result report, what to do if upload fails, and how to verify your changes went through.

If you haven't made any changes yet, the upload step doesn't apply — there's nothing to upload. Make your edits first (see the relevant operations article for whichever type of change you're making), then come back here.

The upload process

From the Bulk Operations page in Seller Central:

1. Find the Spreadsheets for upload section. It's usually below the download section, or on a separate tab.
2. Click Choose file (or drag and drop your file into the upload area).
3. Select your modified .xlsx file.
4. Amazon does a quick client-side check (file format, size). If that passes, click Upload.
5. Wait. The processing takes anywhere from a minute to about thirty minutes depending on file size and current Amazon load.
6. When done, a result file is available for download. Click it.

The result file is the same structure as your uploaded sheet, with two extra columns at the end: Status and Reason. Status tells you per-row what happened — Success, Error, Skipped. Reason explains errors.

What Amazon does during processing

Amazon's bulk-upload validator runs three passes:

1. Format validation. Is each row in the right shape? Are required columns populated? Are numeric cells actually numeric? This catches typos and structural mistakes.
2. Cross-row validation. Do the IDs link correctly? If a Keyword row references an Ad Group ID, does that Ad Group ID exist either in the file or in your existing account? This catches broken references.
3. Application. For rows that passed validation, Amazon applies the change — creating the campaign, updating the bid, archiving the keyword, etc. Each row succeeds or fails independently at this stage; one failure doesn't block other valid rows.

The result is per-row: each row in your file gets a status, and the changes that succeeded are live in your account. Failed rows are documented in the result file for you to fix and re-upload.

Reading the result file

Open the result file Amazon gives you. The original columns are unchanged. Two new columns appear (typically at the right):

• Status: SUCCESS means the row was applied. ERROR means it failed validation or application. SKIPPED means Amazon recognized the row but didn't act on it (usually because Operation was blank — Amazon treats it as "no change requested").
• Reason: For SUCCESS rows, blank. For ERROR rows, a human-readable error message ("Bid below minimum," "Invalid match type," "Campaign with this ID not found," etc.).

Filter the Status column to ERROR. The visible rows are the ones you need to fix. The Reason column tells you what was wrong with each.

Common error reasons (covered in detail in the upload errors article):

• Required field missing
• Invalid enum value (State, Match Type, Bidding Strategy, etc.)
• Bid below the $0.02 minimum
• Date format wrong
• Currency symbol or comma in a numeric field
• ID reference doesn't exist
• Duplicate row

What to do about errors

The good news: errors don't affect other rows. If your upload had 500 changes and 12 failed, the other 488 succeeded — they're live in your account. You only need to deal with the 12.

The workflow:

1. Open the result file.
2. Filter to Status = ERROR.
3. For each failed row, read the Reason and fix the underlying cell. Maybe a Bid was 0.01 (below minimum); change it to 0.02. Maybe a State was "Enabled" with a capital E; change to "enabled".
4. Save the fixed file (or create a new file containing just the fixed rows).
5. Upload again.
6. Amazon processes only the rows you've now corrected — the successful ones from the first upload are still good, you don't need to re-include them.

For very small fix-up uploads, you can extract just the failed rows into a fresh small file with the same column structure and upload that. Amazon doesn't require the file to contain every row in your account; it just processes what's there.

Verifying changes after a successful upload

The result file says SUCCESS for the rows you cared about. Two ways to verify the changes are actually live:

• Spot-check in the UI. Open Seller Central's campaign manager. Navigate to one of the campaigns or keywords you changed. Confirm the new value is shown. The UI updates within 5–30 minutes of a successful bulk upload.
• Download a fresh bulk sheet a few hours later and verify the new values appear in the downloaded data. This is the strongest verification, especially for changes made in batches across many campaigns.

Don't skip verification on consequential changes (campaign creates, major bid adjustments). Bulk uploads are correct the vast majority of the time, but the cost of a silent error on a live ad account is high. Five minutes of verification is cheap insurance.

What if the whole upload fails?

Occasionally an upload fails entirely — no result file, just an error message. Common causes:

• File format wrong. The file isn't .xlsx, or it's a corrupted .xlsx (sometimes happens if you save through unusual tools). Re-save in Excel or Google Sheets and try again.
• File too large. Very large bulk sheets (over 100MB) can fail. Split into multiple smaller uploads.
• Worksheet name was changed. Amazon expects the worksheet to be named exactly "Sponsored Products Campaigns" (or the equivalent for other ad products). If you renamed it, validation fails. Restore the original name.
• Required header row missing or modified. Don't delete or rename the header row in the worksheet. Amazon uses it to parse the columns.

If you get a whole-upload failure, the safest move is to re-download a fresh bulk sheet, copy your changes into the fresh file (with all original structure intact), and re-upload that.

Timing and how often to upload

Practical questions:

How long does a typical upload take to process? Small uploads (under 50 rows) usually process in under 2 minutes. Medium uploads (50–500 rows) take 5–15 minutes. Large uploads (1000+ rows) can take 30+ minutes. Peak times (US business hours, Monday-Wednesday) are slower than off-peak.

How quickly do successful changes go live? Within 5–30 minutes of the result file being generated. Bid changes typically apply faster than campaign creates. New campaigns may take longer to start serving impressions because of Amazon's internal indexing.

Can I upload while a previous upload is still processing? Yes, you can queue multiple uploads. They process in order. The UI shows the status of each.

Are there daily upload limits? No published per-day limit. Practical guideline: avoid more than 10 uploads per day per account. Beyond that you risk hitting rate limits or getting Amazon support attention you don't want.

What's a reasonable upload cadence? Most accounts upload weekly during a regular optimization cycle. High-velocity accounts (launches, sales events) might upload daily. Stable accounts often go 30+ days between uploads. The right cadence is driven by the volume of changes you're making, not the calendar.

Safe upload practices

A few habits that prevent expensive mistakes:

• Always keep a backup of the original download. Before you edit, save a copy with today's date. If the upload causes a problem, the backup is your rollback path (re-upload it with Operation set to Update everywhere).
• Review the file before upload. Filter to Operation = Update (or Create, or Archive). Look at the visible rows. Are they the rows you intended to change? Is the count reasonable?
• Spot-check sort orders. Sort by Bid descending — are the highest bids reasonable? Sort by Bid ascending — any below $0.02? These two sorts catch most numerical typos.
• Upload during off-peak hours when possible. Errors are easier to fix when Amazon's support and your colleagues aren't both inundated. Saturday morning is a common bulk-upload window for serious sellers.
• Be more careful with Archive operations than with Create or Update. Archive is permanent for most practical purposes — un-archiving requires creating a new entity, you can't bring back the old one with its history. Triple-check rows where Operation is Archive before uploading.

What happens if I upload the same file twice?

It depends on the Operation:

• Create: The second upload fails for the entities that were created the first time (Amazon will recognize duplicate creation attempts and error them).
• Update: The second upload sets the same values again. No harm done (the values are already what you're "changing" them to).
• Archive: Same — archiving an already-archived entity is a no-op.

So a duplicate upload is mostly harmless for Update and Archive, but produces errors for Create. If you accidentally upload twice, just look at the result file for the second upload and ignore the Create errors.

Next steps

Now that you can complete a full download-edit-upload cycle, the rest of this guide focuses on what to actually do in the middle step:

• Common upload errors and how to fix them — the next article, a reference for what to do when things go wrong.
• Adjusting bids in bulk — the most common ongoing operation.
• Anatomy of a bulk sheet — if you haven't read it, it's the foundation for everything else.

And once your upload cadence is regular, BulkSheet Pro automates the edit step — read your downloaded sheet, apply your decisions, produce an upload-ready file. Try it free →