The workflow at a glance
- Review accounts and sources
- Map products and variants
- Submit pilot data
- Monitor final processing
What we need before implementation
- Merchant Center account and access
- Cloud project and developer link when required
- Consistent product IDs and pages
- Agreed price and stock owner
At a glance
A Google Merchant Center connection supplies structured Eshop catalogue data for product presence on Google. It is not a generic marketplace order-import workflow. Implementation concentrates on stable IDs, accurate prices, availability and the final processed product. A working connector does not guarantee exposure, ranking or approval of every item; it provides a controlled way to prepare and maintain the underlying data.
Choose a feed or API for the actual workload
The choice follows catalogue size, change frequency and shop infrastructure. A scheduled file may serve a simpler workflow, while Merchant API supports programmatic management. An API is not selected solely because it appears more modern. We assess who maintains each field and how staff will identify failed updates. The operating plan matters as much as the initial method used to deliver the catalogue.
Design data sources before updating products
Official API documentation distinguishes submitted data from the final processed product. API updates target appropriate API data sources and do not arbitrarily overwrite products owned by a file source. Sources are planned to avoid conflicting updates. Language, feed label and offer ID form part of product identification, so changes require a controlled transition rather than an unexplained shift that creates a different record.
Match the catalogue to the destination page
Title, image, price and availability must agree with the landing page. Specifications come from verified product information and variants need explicit mapping. We check the exact product opened by the URL. A family page showing a different default option or starting price can produce an inconsistency even when the submitted file is syntactically valid and all the expected fields exist.
Example: frequent ERP price changes
An electronics retailer updates ERP prices several times during the day. The bridge identifies actual changes, preserves IDs and sends current data to the selected source. After submission it checks processed state and exposes pending issues. Staff can correct identifiers or images through an item-level report. This scenario describes organized data management, not a claim that every update is instantly visible or that uploading a catalogue automatically increases sales. A pilot should include both successful submissions and deliberately incomplete products.
Check processing, not only request success
Acceptance includes price and availability changes, removal of discontinued items and retries after temporary failure. A submission response is not the final success criterion: processed products and issue indicators are monitored. Ownership of the Merchant Center account, Cloud project and access credentials should be documented. Regular refreshes remain an operational task after handover, with alerts when the expected update has not completed or a source stops supplying current data.
Acceptance checks before launch
- Landing-page consistency
- Processed-product diagnostics
- No conflicting source updates
- Visible refresh failures
FROM READING TO ACTION
A clear plan for your platform.
Tell us where your platform struggles and what needs to improve. Our team can connect the diagnosis with code, performance and infrastructure work.
Discuss your platform