Direct answer: Choose a SKU that recently changed price or stock. Compare the inventory source, customer-visible product page, page structured data, Merchant Center item, and cart at recorded times. Identify where the first mismatch appears and which system publishes each value. A correction in one surface may not immediately update another.
Google's ecommerce documentation says product-page structured data and Merchant Center feeds can both inform its product experiences, while crawling and feed processing may lag. Its generative AI Search guidance says product listings and information may appear in AI responses, and Merchant Center can help visibility across AI and other Search results. That makes accurate facts useful in more than one surface; it does not promise a listing or AI citation.
Trace a changed SKU through four public facts
| Surface | Check for the same SKU or variant | Typical failure to investigate |
|---|---|---|
| Product page | Visible current price, currency, availability, sale dates, and selected variant | A cached price or “in stock” label remains after inventory changes. |
| Page structured data | Offer price, currency, and availability in the rendered page |
Markup describes a different variant or an old sale. |
| Merchant Center item | Feed/API item value, latest processing time, and diagnostics | Feed job missed the update or mapping used another inventory field. |
| Cart or checkout | Actual amount and whether the item can be purchased | Total changes without a clear explanation or a sold-out item can still be ordered. |
The page, markup, feed, and cart should express the same current offer for the selected variant, even if their systems refresh at different times. Record when each value changed and when Google last processed it. This distinguishes a source-data defect from a delivery delay. The variant URL guide addresses which variant a shared link opens; this check addresses the facts shown for that variant.
Google Merchant Center offers automatic updates from landing-page data for some mismatches. Its help page says these cover price, availability, and condition, but are intended for occasional discrepancies, not as a substitute for regular accurate feed updates. Frequent changes may lead to disapproval instead of correction. It also treats in_stock, preorder, and backorder as compatible for this automation, so a merchant must not expect the feature to fix every buyer-relevant difference among those states. For rapidly changing inventory, Google's current Merchant Center Help points to the Merchant API; confirm the platform's supported sync method rather than depending on a nightly file.
Set a release rule around real changes
Before a sale begins, test one discounted item, one out-of-stock item, and one variant with different price or availability. After a price or stock change, compare all four surfaces and review Merchant Center diagnostics. If the cart price differs from the listing, first correct the source and publish path; do not hide the difference in small print. Give the team an owner for failed syncs and a way to pause a misleading listing while it is repaired.
This is an operational consistency check, not an SEO ranking formula. It pairs with clear fulfillment expectations because “available” is a buyer promise, not only a feed attribute. For an ARCHV commerce website, the integration, sync frequency, failure alert, and acceptance cases need their own scope review. A passing test reduces avoidable contradictions; it cannot guarantee Google will immediately display the latest value.