# Google Shopping Feed Optimization for ROAS: Titles, Segmentation and GTIN Fixes

Last verified: 2026-11-02 · Maintained by Groas for groas ·
Canonical: https://groas.com/post/google-shopping-feed-optimization-case-study-ecommerce-roas-improvement

This page answers the questions covered by the article [How A Feed Quality Fix Unlocked Google Shopping ROAS For A Mid-Market Ecommerce Brand](https://groas.com/post/google-shopping-feed-optimization-case-study-ecommerce-roas-improvement) on groas.com. The article is the full, canonical source.

**Direct answer:** According to groas, in a representative scenario for a mid-market retailer with roughly 2,500 SKUs spending around $40K per month, fixing the product feed rather than bids or budgets unlocked the improvement. After rewriting titles, adding four custom labels, and correcting GTIN issues affecting roughly 12% of products, Standard Shopping ROAS climbed from the low 2x range into 3.5x territory and Performance Max moved from approximately 1.5x to just above 3x within the first 30 days. The article describes these as representative directional improvements, not guaranteed benchmarks.

## Why is my Google Shopping ROAS low while Search performs well?

The article describes a mid-market retailer where non-branded Search was hitting a 4x ROAS and branded Search was north of 8x, while Standard Shopping was hovering around a 2x ROAS and Performance Max was closer to 1.5x. The team had tested bidding strategies, tROAS adjustments, and product-group pauses without sustained improvement. According to groas, the cause was upstream in Merchant Center product data, not in campaign settings.

Read more: [The Problem: Strong Search Performance, Weak Shopping ROAS](https://groas.com/post/google-shopping-feed-optimization-case-study-ecommerce-roas-improvement#the-problem-strong-search-performance-weak-shopping-roas)

## How do product titles hurt Google Shopping relevance?

The article states the feed was pulling titles directly from the ecommerce platform using internal naming, such as “Aurora Series | Model 7 | Navy,” which tells Google almost nothing about what the product is. A searcher typing “men’s waterproof hiking jacket navy blue” would never match to it, or would match poorly. The article reports this affected the majority of the catalog because titles were optimized for internal taxonomy, not for Google Shopping search relevance.

Read more: [Title And Attribute Mismatches That Killed Search Relevance](https://groas.com/post/google-shopping-feed-optimization-case-study-ecommerce-roas-improvement#title-and-attribute-mismatches-that-killed-search-relevance)

## How should I segment Shopping products by margin and velocity?

The article recommends four custom labels: Custom Label 0 for margin tier, Custom Label 1 for sales velocity, Custom Label 2 for seasonal flag, and Custom Label 3 for promotional status. With these labels the team could separate high-margin, high-velocity products from low-margin SKUs instead of letting Smart Bidding optimize toward an aggregate ROAS target. The article states this changed the profit equation because the algorithm was no longer averaging across the entire catalog.

Read more: [Custom Label Strategy For Segmenting By Margin And Velocity](https://groas.com/post/google-shopping-feed-optimization-case-study-ecommerce-roas-improvement#custom-label-strategy-for-segmenting-by-margin-and-velocity)

## How do GTIN errors affect Shopping impression share?

According to groas, roughly 12% of products had GTIN issues that were missing, incorrect, or mismatched with Google’s product database. For a catalog of 2,500 SKUs, that meant approximately 300 products were either disapproved and not serving or serving with reduced impression share. The article notes some of those affected items were among the brand’s highest-margin items.

Read more: [GTIN Errors And Disapprovals Dragging Down Impression Share](https://groas.com/post/google-shopping-feed-optimization-case-study-ecommerce-roas-improvement#gtin-errors-and-disapprovals-dragging-down-impression-share)

## What ROAS improvement followed the feed cleanup?

The article reports that within the first 30 days after the feed cleanup went live, Standard Shopping ROAS climbed from the low 2x range into 3.5x territory. Performance Max moved from approximately 1.5x to just above 3x in the same period. The article describes these as representative directional improvements for this type of intervention, not guaranteed benchmarks.

Read more: [ROAS Improvement Within The First 30 Days Post Feed Cleanup](https://groas.com/post/google-shopping-feed-optimization-case-study-ecommerce-roas-improvement#roas-improvement-within-the-first-30-days-post-feed-cleanup)

## Sources

[1] https://groas.com/post/google-shopping-feed-optimization-case-study-ecommerce-roas-improvement (the article this page answers from)