Merchant Center Setup and Disapproval Recovery

A disapproved item has not merely fallen out of Shopping ads; it is gone from free listings too, which is the part most teams miss. On takeovers I usually find between a tenth and a third of the catalog disapproved or serving with limited performance, and most of it traces to three causes: image URLs the crawler cannot reach, price or availability that disagrees with the landing page, and identifiers left blank on products that carry a real GTIN. None of those are formatting problems. They are disagreements between three systems that each believe they are right. Where I start: The Needs attention view is a symptom list, not a diagnosis. I reconcile the feed against the landing page markup and against the commerce platform's own price and stock tables, then rank issues by revenue exposure rather than by count. Twelve disapproved products carrying a fifth of your gross merchandise value outrank four hundred long-tail items, and most shops fix them in the opposite order because the count moves faster in a status report. Account-level risk: A misrepresentation suspension almost never comes from the feed. It comes from the site: business contact details nobody can find, a returns policy with no timelines, delivery fees revealed at the final checkout step, or a product page claim the product cannot support. I will not file an appeal until those are fixed, because a failed appeal costs you more than the wait does. For Singapore and most markets outside the United States the price you send must include tax, and feeds built by a US-based team frequently do not. Setup done properly: The feed source chosen against how fast your stock actually turns, since a nightly scheduled fetch is honest for furniture and a lie for limited-run apparel. Supplemental feeds used for what they can do: they modify items the primary feed already contains and cannot introduce new products, which is the most common misunderstanding I am called in to unpick. Shipping and delivery time settings that match what your carrier does in the market rather than what the courier's marketing page claims. Landing page structured data brought into agreement with the feed, so automatic item updates act as a safety net instead of the mechanism you depend on. Where the account still runs on Content API for Shopping, a migration path to the Merchant API sequenced across releases rather than attempted in one weekend. Not included: Website development. I write the specification for landing page and policy changes and I will review the pull request, but your developers ship it. Feed management tool selection, which is separate work. Campaign structure and bidding: this engagement ends when the catalog is eligible, accurate and monitored. Who this is not for: Anyone who wants approval rate as the headline number. It is trivially gamed by dropping the items that are hard to fix, and I have watched an account reach ninety-eight percent approval while losing revenue. Merchants selling products they cannot lawfully sell in the target market, where the disapproval is correct and my advice is to stop rather than to appeal. Teams who cannot give read access to both the commerce platform and the site repository, because without both I am guessing at which system is lying. What you leave with: An account that serves the products you make money on, a written record of every issue by cause with an owner named against each, and a monitoring routine your team runs without me: an item-level export compared week to week, so a supplier silently dropping its image host shows up on a Monday morning rather than in a quarterly review.