Separate what is visible from what is known
A photo may help identify a pattern, neckline or sleeve shape. It does not reliably establish fibre percentages, garment measurements or care instructions. Use product labels and supplier records for those facts. Review AI suggestions as drafts and correct them before exporting.
- Confirm fabric composition from product records.
- Check neck, sleeve, fit and pattern against the garment.
- Keep unsupported certifications and performance claims out of the copy.
Use the category’s vocabulary
A marketplace field may require a specific allowed value rather than a free-form phrase. Pick the category and article type first, then review the supported values. Similar words are not always interchangeable in a workbook. If none of the available choices describes the item accurately, check the category and source details rather than forcing a convenient answer.
Treat size measurements as product facts
A size label alone does not explain how a garment fits. Confirm the measurement unit and whether the data represents body measurements or garment measurements. Keep the reference consistent across the size range and ensure the uploaded size information belongs to this product, not a visually similar style.
Review the sellable rows
Shared style copy can appear across several size rows, but each row still needs the correct SKU and variation details. Check for duplicate identifiers, missing sizes and mismatched colours. One product with S, M, L and XL represents four listing rows in the EcommerceListing allowance model.