A product feed is the structured data uploaded to Merchant Center that defines each product. It’s made up of fields such as title, price, stock status, brand, and product identifiers.
Feed quality directly determines visibility. If a required field is missing or wrong, the product isn’t approved; if a recommended field is missing, the product is approved but its search matching can stay weak. The two situations point to different problems and need to be fixed differently.
The Merchant Center product data specification documentation defines the format and requirement rules for fields such as id, title, description, link, image_link, availability, price, brand, gtin, mpn, and condition. The Shopping ads policies documentation states that compliance is checked through ongoing review that combines Google’s automated systems with human evaluation.
What Are the Required Fields?
If any of these fields is missing or incorrect, the product gets rejected by Merchant Center.
Identity and basic information fields
id (up to 50 characters) uniquely identifies each product. title (up to 150 characters) and description (up to 5,000 characters) are the text fields that describe the product. link points to the product’s own page.
Image, stock, and price fields
image_link must be at least 500x500 pixels. availability (in_stock, out_of_stock, preorder, backorder) indicates stock status, and price must follow the ISO 4217 currency format. A mismatch in these fields creates a conflict between the product’s advertised status and its actual state on the site.
Product identifier fields
brand (up to 70 characters) is required except for movies, books, and music products. gtin is strongly recommended when available; if it isn’t, mpn becomes required. condition (new, refurbished, used) is required especially for used or refurbished products; skipping this field causes the product to be treated as new by default.
Why Do the Recommended Fields Matter?
Recommended fields don’t get a product rejected, but they do affect search matching and click quality — which is why they shouldn’t be dismissed as unnecessary.
Additional images and rich content
You can add up to 10 additional image links, a video, or a 3D model link. These fields boost match quality especially in categories where visual comparison matters (clothing, furniture).
Unit price and attributes
Unit price information along with attributes like size, color, and material strengthen the match between the user’s search query and the product. For example, a search for “red leather bag” can match weakly if these fields aren’t filled in — if the product title only says “bag,” the search engine may struggle to associate the color and material information in separate fields with that generic word in the title.
How Does Feed Quality Break Down?
The most common failure mode is a change on the site that doesn’t get reflected in the feed in time.
Price and stock sync are critical
If the price changes on the site but the feed isn’t updated, the price shown in the ad can conflict with the price on the site. This puts both user trust and the account’s policy compliance at risk — and the real danger isn’t any single product, it’s this gap building up unnoticed.
Bulk update errors have wide-reaching effects
If a bulk update — like a category renaming or template change — has an error, it can affect not just one product but hundreds at once. That’s why bulk changes should be tested on a small sample before going live.
Pre-Launch Checklist
To check feed quality before launch:
- Are all required fields (id, title, price, availability, brand, gtin/mpn) filled in and correctly formatted?
- Does image resolution meet the minimum requirement?
- Is price and stock data in sync with the site, and does the update frequency match how fast the catalog changes?
- Is the list of rejected products reviewed regularly?
Common Mistake
The most common mistake is putting careful effort into the feed only at initial setup and never auditing it afterward. Google’s evaluation isn’t a one-time event — a product can be rejected individually at any point if a policy or format violation is later detected.
Summary
The product feed is the core data source for Shopping and Performance Max campaigns. An error in a required field gets a product rejected outright; a gap in a recommended field leaves it visible but with weaker match quality. Feed quality isn’t a one-time setup — it’s a process that needs continuous monitoring.