Map the catalog's full data path
Identify the ecommerce system, feed or partner integration, Commerce Manager catalog, data source, dataset or pixel, product sets and active campaigns. Record the catalog owner, feed schedule, last successful update and whether inventory is supplied through a scheduled file, partner sync or API. A campaign screenshot alone cannot tell you which system owns a broken field.
Choose a sample of products: best sellers, out-of-stock items, discounted variants, newly added items and recently discontinued products. Follow each through the source record, uploaded feed, catalog status, product set and landing page. This reveals whether a problem affects one SKU, an entire category or the whole pipeline.
Read errors, warnings and policy decisions separately
In Commerce Manager, review product-level status and issues alongside the data source's latest upload report. A feed error can prevent ingestion; an item issue can restrict a product's eligibility for ads or shops; a warning may identify a quality problem without blocking every use. Meta distinguishes commerce and advertising policy contexts, so a product may be eligible for one surface but not another.
Export the issue report when available. Group by error type, affected product count, estimated business importance and the system that must be fixed. Do not begin by editing hundreds of items directly inside Commerce Manager if the next feed sync will overwrite those edits. Correct the authoritative source and rerun the pipeline.
Verify required product attributes and destinations
Check stable item IDs, descriptive titles, accurate availability, condition, images, links, category or brand fields where required, and price/currency consistency between feed and destination. Test product URLs in a clean browser session and on mobile. Confirm that a variant's feed link leads to the same colour, size and availability the ad promises.
Pay special attention to image access and redirects. A product image may render for staff but be blocked for Meta's fetcher. An expired CDN URL, anti-bot rule or temporary signed image link can create intermittent ingestion failures. A landing page that silently switches variants can make a technically approved catalog ad misleading to the shopper.
Audit product sets before judging campaign performance
Review the filters defining each product set: category, custom labels, availability, brand, collection or price band. Count eligible items before and after the filter and compare with the expected merchandising inventory. A set that appears populated in the ecommerce system may contain very few approved, in-stock items in Meta after policy and data checks.
Keep evergreen sets distinct from seasonal promotions when the inventory and creative promise differ. For example, a holiday gift set that still includes discontinued products may cause the campaign to serve an irrelevant destination or have little delivery. Validate sets after promotions end and whenever feed taxonomy changes. Avoid creating dozens of tiny sets that no one maintains.
Reconcile product identifiers with website events
For dynamic retargeting and product-level optimization, compare catalog item IDs with the identifiers passed in website events such as product views, add-to-cart and purchases. Check whether the integration sends parent IDs, variant IDs or both, and whether content type and currency are consistent. A mismatch can reduce event-to-catalog coverage even when the feed has no obvious errors.
Test with a small set of synthetic journeys and use the available event diagnostics. Do not assume that an event count equal to a purchase count proves correct item matching. Verify a specific test item's ID through the browser event, the server event if used, and the catalog record without exposing customer data.
Create a monitoring and escalation routine
Track last successful feed ingestion, eligible-item count, rejection count, out-of-stock accuracy, broken product URLs, image failures and event-to-catalog match quality. Assign a response owner for each failure class. Critical problems such as an empty catalog or invalid destination links should trigger prompt investigation; cosmetic warnings can enter a scheduled maintenance queue.
Before changing campaign budgets, record a baseline and distinguish feed fixes from creative or audience tests. A recovery in return on ad spend after restoring eligible inventory is not necessarily evidence that the audience strategy improved. Document when product-set filters, feed mapping and website events were changed.
Practical launch and recovery checklist
Before launch, confirm the catalog and dataset are linked, the correct set contains sufficient approved products, landing pages match item details, and event IDs resolve to catalog items. After launch, inspect diagnostics again rather than relying on initial approval. Use the same checks after major ecommerce platform updates or product imports.
Meta's Commerce Manager help documentation covers data feed errors, product status and catalog eligibility. Use those primary sources to distinguish a technical data defect from a policy rejection. This article extends the Meta placement and creative QA guides by focusing on whether the right products are available to be advertised at all.