Excel/CSV imports, refresh, and freshness

Choose an import path

Use a report preset when the export type is supported. For another file, choose generic import and map columns manually. Availability of Wildberries, Ozon, and Yandex Market presets follows the current interface.

• CSV and Excel (.xlsx) are accepted within published limits.
• The file needs a header row and stable product identifiers.
• Keep the untouched source export as the reconciliation record.

Map and normalize

1. 1

   Select the Excel sheet or CSV delimiter and encoding, then set timezone and currency.

2. 2

   Map required fields: date, seller account, SKU or article, operation, quantity, and amount.

3. 3

   For a wide report, enable column-to-row transformation (unpivot) and select identifier columns.

4. 4

   Review the preview before running and save a mapping profile only after validation.

Review failures

• File error means format, size, encoding, or sheet could not be read.
• Schema error means a required column is missing or a value has the wrong type.
• Rejected rows mean processing continued but individual records need correction.
• Duplicates require checking the account, report, operation, product, and date key before retry.

States and freshness

1. 1

   The file enters the queue and receives an import identifier.

2. 2

   Source rows are validated, normalized, and linked to provenance.

3. 3

   Accepted data is published to analytics datasets; rejected rows remain in the error report.

4. 4

   After completion, verify period, row count, control totals, and the new freshness watermark.

Decision check

Reconcile revenue, returns, fees, logistics, and sale quantity against the source export. Do not use the result for a financial decision if period, currency, timezone, or row count does not match.

First marketplace import

Start sync queues an initial import for every active marketplace connection with stored credentials. The request being accepted does not mean that data is already available. Review import jobs and dataset freshness.

The onboarding step completes only after successful import jobs and a successful refresh from raw data to analytics. A dry run or failed job does not complete the step. Resolve credential or permission errors, then retry; an active job is reused rather than duplicated.