Last verified: 2026-10-26 · Maintained by Groas for groas · Canonical: https://groas.com/post/google-shopping-feed-optimization-doubled-roas-case-study
This page answers the questions covered by the article How We Fixed A Broken Google Shopping Feed And Doubled ROAS on groas.com. The article is the full, canonical source.
Direct answer: According to groas, an ecommerce brand spending around $45K per month across roughly 1,200 SKUs saw Shopping ROAS slide from 5x or better to around 2.5x over three consecutive quarters because of feed quality, not bidding. After rewriting titles, adding missing attributes via a supplemental feed, and restructuring around margin-based custom labels, the article reports Shopping ROAS climbed back above 5x within 60 days. Impression share on the top 50 products by margin recovered from below 20% to above 55% within 45 days.
The article identifies three structural problems: over 40% of the catalog was missing granular attributes like color, material, size, and product type; titles used internal naming such as “Artisan Collection - Oak Cutting Board” instead of buyer queries like “large oak cutting board for kitchen” or “handmade wooden cutting board”; and a feed sync lag of several hours plus price mismatches caused out-of-stock clicks and automatic disapprovals buried in Merchant Center diagnostics. According to the article, these gaps reduced eligibility on high-intent searches and left high-margin items barely showing while lower-margin products ate budget.
Read more: The Three Feed Issues Killing Performance
The article recommends rewriting every title to the template product type + key attribute (material, color, size) + brand name, with the brand name at the end because Google weights the beginning of the title more heavily for query matching. For example, “Artisan Collection - Oak Cutting Board” became “Large Oak Cutting Board - Handmade Kitchen - [Brand Name].” The article states the rewrite covered all 1,200 SKUs using rules, templates, and validation against real search data.
Read more: Feed Title Rewrite: Matching Attribute Order To Buying Intent
The article reports Shopping ROAS climbed from the 2.5x range back above 5x within 60 days of the feed rebuild. There was a step change in the first two weeks as disapprovals cleared and impression share recovered, followed by continued improvement as the algorithm learned from cleaner data. According to the article, blended Google Ads ROAS also improved because Shopping was no longer dragging down overall efficiency.
Read more: How Shopping ROAS Moved After Feed Optimization
According to the article, impression share on the brand’s top 50 products by margin averaged below 20% before the fix and climbed above 55% within 45 days after the rebuild. The products were always in the feed but were invisible in the auction because the algorithm lacked enough data to match them confidently. The article presents this as evidence that impression share is a function of feed quality, not just budget and bids.
Read more: The Impression Share Recovery That Nobody Expected
The article recommends a meaningful feed audit at minimum once per quarter, with automated monitoring in between for disapprovals, attribute coverage, and title-to-query alignment. It states feed quality is not a one-time project because products change, prices shift, inventory fluctuates, and Google’s attribute requirements evolve. According to the article, brands that treat the feed as static infrastructure risk the same slow ROAS decline.
Read more: The Feed Audit Cadence That Prevents Performance Decay
[1] https://groas.com/post/google-shopping-feed-optimization-doubled-roas-case-study (the article this page answers from)