top of page

Meta Catalog Launch Checks: Reconcile the Items You Plan to Advertise

Writer: DIGITALTHROB
DIGITALTHROB
Oct 29
4 min read

Reconcile the items you intend to advertise before approving a Meta catalog launch. A completed upload does not establish that every approved item arrived, carries the current offer or leads to the right product option.

Start with a named launch range. Match its item identifiers to the resulting catalog records, inspect the commercial details, then follow the customer destinations. The outcome should be an explainable list of ready items and specific exceptions.

A turquoise catalog tray holds colorful product tiles beside a coral inspection lens and one gold approval seal, with the DIGITALTHROB wordmark.

Define the range you are approving

Record the catalog, proposed product group, source-export date and approved item identifiers. Product names alone are insufficient when colours, sizes or packs share a name but differ in price, availability or customer choice.

Ask the campaign owner to show the saved selection and compare it with that list. A campaign name containing “selected range” does not establish which items the configuration can use. Resolve that scope before interpreting a healthy catalog total.

Keep out-of-scope faults in a separate repair queue. They may matter to the wider business, but mixing them into the launch count makes it harder to see whether the approved range is ready. Do not expand that range just to improve the total.

Use three checks instead of one upload result

  1. Source receipt: did the latest source arrive with the expected identifiers?

  2. Catalog record: what does each resulting item show, including visible issues?

  3. Customer offer: does its destination show a product and purchase action the business can support?

Meta’s official product-item model represents price, currency, availability and item errors separately. That supports checking the commercial details individually; a record’s presence is insufficient. The model does not establish serving eligibility or the mapping used by a particular store integration.

Retain the times of the source export, catalog inspection and website check. A stock change between observations can explain a difference that would otherwise look like a faulty transfer. Establish the sequence before assigning the repair.

Count affected items without counting the same item twice

Worked example (fictional): a homewares store approves eighteen item identifiers. The latest export contains all eighteen, but the reviewer finds only seventeen corresponding catalog records. One item is missing.

Among the seventeen present items, two have price differences, one has an image issue and one shows stale availability. The image issue belongs to one of the two price-mismatched items. There are four issue observations across three affected items.

The initial reconciliation is therefore fourteen clear items, three present items requiring repair and one missing item: eighteen in total. Subtracting four issue observations from seventeen would double-count an affected item.

Next, the reviewer follows the fourteen apparently clear records to the store. One destination selects a discontinued finish. The completed review now shows thirteen passing items, four present items needing repair and one missing item. The catalog totals alone would have missed that fourth repair.

Fix the source that controls the next update

Trace each incorrect value to its owner. In the example, an obsolete promotional export caused the price differences. Replacing that source prevents the next sync from restoring the old prices. The image owner repairs the inaccessible asset, and operations corrects availability through the agreed inventory source.

Investigate the missing identifier separately: was it excluded, changed or rejected during transfer? Creating an improvised second identifier merely to reach eighteen can conceal the original fault and complicate the next reconciliation.

The destination owner repairs the discontinued-finish selection. Our guide to consistent product variants in paid ads explains why reaching the right product family is insufficient when the clicked option differs.

For each exception, keep the original value, expected value, repair location, owner and retest condition. This turns “feed problem” into an action the right person can complete.

Retest the resulting records and buying routes

Run the corrected source through its normal update route. Inspect the resulting identifier, price, currency, image and current commercial state, then follow the saved customer destination and check the selected variant, terms and available action.

Complete the same recorded checks for the rest of the approved range. One repaired price does not establish that every price or route is correct. In the example, all eighteen records and customer routes eventually pass the business review.

Hold affected items while a required record is missing or a material commercial mismatch remains. If the business proposes a smaller launch, explicitly approve the reduced range and verify the actual saved selection. Keep any unresolved processing or eligibility outcome as a separate release condition.

Give approval a clear expiry condition

Retain the dated range and name the owner of price, stock and source updates. Reopen the relevant checks when a promotion changes, an item is added, a sync fails or a customer reports a mismatch.

Our guide to adjusting ad spend when stock is low covers the spending decision when usable stock changes. A catalog repair and permission to spend are connected decisions with different evidence.

Bring the selected range, source history and unresolved item examples to our Meta Ads management team. We can help locate the next repair in the campaign selection, product source or buying journey.

 
 
 

Comments


Discover clics solution for the efficient marketer
More clics

Never miss an update

bottom of page