How should I build a product synchronization workflow?

Match existing products before creating them, update only the fields your source owns, and keep the Endless product ID with the external identifier.

Before implementation, agree which system creates products, owns each field, manages inventory, and handles archived or discontinued products. Decide how identifiers, attributes, media, prices, and composition will map. Separate ownership prevents later updates from overwriting another system’s work.

A practical sequence is:

1. Review the relevant operations and inputs in the API reference.
2. Match the product by SKU or a stable external identifier, rather than its name.
3. Create required definitions and related products first.
4. Create or update the product and save its returned Endless ID.
5. Apply related attributes, media, prices, composition, and inventory through their documented operations.
6. Record results and retry only unfinished steps.

Begin with a small set covering variants, composed products, packaging, and inventory. Compare results before loading the full catalog. For ongoing syncs, process changed records, checkpoint completed work, and reconcile periodically.

A network failure can occur after a product was saved. Look it up before trying to create it again. Restore and update an archived product when its SKU is still the identifier you need.

See product integration behavior, API connection, and errors and retries.