Overview

Error 90003 is a feed processing error that Amazon sellers encounter when their product feed files contain characters that cannot be recognized or mapped by the specified character encoding set. This error typically appears during inventory uploads, listing updates, or other bulk feed submissions through Seller Central or third-party tools. Understanding how to diagnose and resolve this encoding mismatch is essential for sellers who rely on feed-based workflows to manage their catalogs efficiently.

Key Points / What Sellers Need to Know

• Error message format - The full error reads: "Too many unmappable bytes in line {line number} for the given character set {character set name}," where the line number and character set are specific to your file and submission.
• Root cause - Your input file contains characters that do not belong to or cannot be translated within the character encoding format you specified when submitting the feed.
• Common triggers - Special characters, accented letters, non-Latin scripts, curly quotes, em dashes, and symbols copied from word processors or websites are frequent culprits.
• Resolution approach - You need to review your input file, identify the problematic characters on the flagged line, correct or remove them, and resubmit the feed.

How Character Encoding Works in Feeds

When you submit a product feed to Amazon, you specify a character encoding standard such as UTF-8, ISO-8859-1, or Windows-1252. This encoding tells Amazon how to interpret the raw bytes in your file as readable text. If your file contains bytes that fall outside the valid range for that encoding, Amazon cannot process those lines and returns Error 90003. UTF-8 is the most widely compatible encoding and supports virtually all international characters, making it the recommended choice for most sellers. Problems most often arise when a file is saved in one encoding but submitted under a different one, or when text is copied from sources that introduce invisible or incompatible characters.

Common Causes and How to Identify Them

The most frequent cause of this error is a mismatch between the actual encoding of your file and the encoding declared in your feed submission. For example, a file saved with Windows-1252 encoding but submitted as UTF-8 can produce unmappable bytes. Another common scenario involves copying product titles or descriptions from websites, PDFs, or word processing applications, which often embed special characters like smart quotes, trademark symbols, or non-breaking spaces that may not be supported by your declared encoding. To identify the problematic characters, open your file in a text editor that displays encoding information, navigate to the line number referenced in the error, and look for unusual or non-standard characters.