Review CSV data from eight Chinese marketplaces before catalogue import
Review row structure, variants and source-specific fields when preparing Chinese marketplace data for catalogue import.
A supplier file may open neatly in a spreadsheet and still need work before catalogue import. Start by agreeing what each row represents and how its fields map to the destination.
This review covers data projects using 1688, Alibaba, Taobao, JD, Pinduoduo, Weidian, Goofish and Yiwugo. Use the eight fictional files in the PandaBuyNow sample pack to discuss the required structure; actual field availability must be checked against the chosen source and agreed sample.
In a Reddit discussion, a store operator asked how others clean supplier CSV files before Shopify import. Replies described column mapping and validation, including promotional suggestions for tools. The useful question is what a repeatable review should check, rather than which advertised tool to trust.
Write down what one row means
PandaBuyNow’s website examples use one CSV row per fictional source record. Nested variants and price tiers stay as JSON text inside cells. A destination that expects separate variant rows needs a different mapping. Being a CSV file does not establish compatibility.
Keep an untouched copy of the delivered file. Make the proposed import file separately, with a short record of which source column supplies each destination field.
In those fictional files, 1688 and Alibaba keep quantity-price tiers, Taobao and Weidian keep option relationships, and JD keeps displayed availability separate. Pinduoduo retains offer conditions; Goofish retains the stated condition; Yiwugo retains published shop-location context. Map each source on its own terms rather than forcing every value into a generic price column.
Test identifiers and option relationships
Use an original fictional test case: a product has two variants, with review codes 00042-A and 00042-B. Preserve each code as text. Check that both options remain attached to the intended product after conversion.
If you turn one product row into two variant rows, compare product counts and variant counts separately. Two rows can be a correct conversion of one product; they should not automatically be treated as duplicate products.
Check non-Latin titles, commas inside descriptions and nested values. Reopen the saved file to see whether those values survived the export. A readable first row does not test every case.
Read the destination’s update rules
Shopify’s product CSV documentation distinguishes blank optional columns from columns omitted during an overwrite. It also documents dependencies between variant-related fields. Confirm the current rules for the exact import you intend to run.
A missing source value is therefore a review decision. Decide whether it should remain unavailable, prompt a supplier question or be left out of an update file. Avoid treating every blank as an instruction to erase existing catalogue information.
Review a small conversion before the full file
Shopify recommends keeping a product-data backup; its guidance for Partners performing large imports calls for testing a small subset in a development store. Use an appropriate test environment for your destination and review the resulting records.
Check the chosen options, identifier text and any image associations. Record rejected rows and resolve their cause before expanding the import. Keep the source export and transformation notes with the review.
For a PandaBuyNow project, share the destination requirements and representative links. Discuss row structure and sample acceptance before agreeing delivery. The website sample pack illustrates data structure; it is not a ready-made Shopify import or an automatic store integration.