- For large product ranges, the product feed determines which items appear in Google Shopping and where your budget goes.
- According to Merchant Center Help, if the feed price doesn't match the store price, Google may disapprove the product. In the worst case, the account can be suspended.
- After the feed cleanup, Atelier Nomi's Google Shopping revenue grew 141%. At Jomos Schuhe, new product titles, measurement and campaigns lifted sales by 128%.
- Using up to five custom labels, we segment the product range by margin. Campaigns then allocate budget accordingly.
- The Content API for Shopping was retired on August 18, 2026. API integrations now run through the Merchant API.
For a large product range, the product feed determines which items appear in Google Shopping. It also determines where budget goes in Shopping and Performance Max campaigns.
In this article, we show how we evaluate a feed with many thousands of products and what we change. The evidence comes from our case studies; the technical details come from Google.
What changed in feed integration in 2026
Google retired the Content API for Shopping on August 18, 2026. Since September 1, 2026, requests without an approved extension have failed intermittently. According to Google, the API will be shut down entirely in early 2027.
If you send product data to Merchant Center via API, you now need the Merchant API. Check with your IT team whether the migration is complete. We explain what the new API can do in our article on the Merchant API.
We keep feed prices and stock levels current
Google compares the price in the feed with the price on the product page. According to Merchant Center Help, they must match exactly. Otherwise, the product may be disapproved, and in the worst case the account suspended.
With many thousands of products, prices and stock levels change constantly. That's why we set up rules that pull sold-out items from advertising immediately. Monitoring catches feed errors before Google disapproves products.
Today, not a single product in Atelier Nomi's feed has errors or is disapproved. Google Shopping revenue grew 141% in August 2025 versus July (Atelier Nomi case study).
We write titles that match how people search
Google uses the product title to match an item to search queries. That's why our titles include brand, product, material and other attributes.
At Jomos Schuhe, we systematically rewrote the product feed titles. Combined with new measurement and refined campaigns, Google Ads sales grew 128%, from 111 in September to 253 in November 2025 (Jomos Schuhe case study).
AI helps us work faster. But we never add AI-generated titles to the feed without review.
We add missing attributes
We fill in material, size, age group and intended use completely and consistently. That helps Google match your products to the right searches.
- At hundesachen.de, we added material and intended use to the feed. It was one of four steps that together lifted Google Shopping sales by 21% (hundesachen.de case study).
- At Atelier Nomi, attributes such as gender (“unisex”) and age group are now filled in consistently.
We segment your products by margin
Merchant Center offers five custom labels, custom_label_0 to custom_label_4. According to Google, they let you group products for reporting and bidding in Shopping, Performance Max and Demand Gen campaigns. Each label can hold up to 1,000 distinct values per account.
We use the labels to group the product range by margin. You provide the margin data for each item from your inventory management system. We use it to divide the range into suitable campaign groups. Bestsellers, high-margin products, new arrivals and slow movers then each get their own budget and their own targets.
- Sporthaus Hopfmann manages its campaigns by margin in exactly this way. Brand and product searches are managed and reported on separately (Sporthaus Hopfmann case study).
- At hundesachen.de, budget goes to bestsellers and high-margin products. In August 2024, Google Ads revenue grew 89% month over month.
For Electronic Partner and its 300 specialty stores, we set bids to a fixed revenue per euro of ad spend (target ROAS). For local inventory ads, ROAS rose 41% from 2024 to 2025 (Electronic Partner case study).
We serve every channel from one master feed
From one master feed, we create dedicated versions for Google, Meta, Microsoft, marketplaces and price comparison sites. Prices and stock levels sync automatically.
For Google Shopping, we can run your ads through our own comparison shopping service, unitedads.shopping, if you wish. Your clicks in the EU then cost less than through Google Shopping directly, which raises your margin per sale (Google CSS).
We show how product data makes its way into AI assistants like ChatGPT in our Agentic Commerce Playbook and on our ChatGPT Shopping page.
How we approach a large feed
| Step | What we do |
|---|---|
| Analysis | We audit the feed, Merchant Center and store data and quantify how many products are missing, disapproved or poorly described. |
| Prioritization | We start with the products that earn you the most. |
| Implementation | We rewrite titles and attributes, set up rules and monitoring and connect the channels. |
| Maintenance | We keep the feed current at all times and adapt it to new products, seasons and channel requirements. |
For large product ranges, we agree with your IT team up front on where the data comes from and who approves changes. Merchant Center, feeds and all data remain your property, even after our engagement ends.
Learn more on our product feed management and Google Shopping pages.
