How to use the Amazon Backend Search Terms Cleaner.
Clean search terms using the current rules supplied for the marketplace. Check byte limits and duplication without inventing keyword popularity or relevance.
Make the workflow fit your task.
Specify the marketplace rules, encoding and deduplication unit before cleaning. Separate candidate phrases into tokens only when that rule is intended, preserve first-occurrence order and count the final encoded string. Report removed duplicates and unresolved relevance questions separately.
- What you provide
- Supplied keyword list and current platform rules.
- What you get
- Deduplicated search-term draft with byte count.
See the input and the result.
Illustrative input and output · a teaching example, not a live WebAct run
Example input
Candidate phrases: bamboo organizer, desk organizer, bamboo. Example rules: lowercase; split commas and spaces into words; remove repeated words in first-occurrence order; join with one space; count the final string in UTF-8. Illustrative maximum: 40 bytes, not a statement of Amazon's current limit.
Completed example
Cleaned string: bamboo organizer desk Unique words: bamboo, organizer, desk Removed repeats: organizer once; bamboo once UTF-8 byte count: 21 Example limit check: 21 ≤ 40, leaving 19 bytes under the supplied rule. Counting check: café desk has 9 characters but 10 UTF-8 bytes. Confirm the current marketplace requirements and product relevance before publishing search terms.
Load this input into the prompt, then copy it to WebAct to try the task. Your result may differ from the illustration.
Decisions and troubleshooting.
Why can byte count differ from character count?
UTF-8 uses more than one byte for many non-ASCII characters. Count the encoded final string, including the separators required by your chosen rule.
Can this cleaner prove a term is popular or permitted?
No. Deduplication and byte counting do not establish demand or policy compliance. Review terms against current marketplace rules and actual product relevance.
Try it with your own source.
Replace the example with your material in the task prompt. Keep the requirements you need, then copy the task into WebAct.
Customize and copy the task ↑