# Why Shopping Revenue Plateaus When ROAS Looks Fine

Last verified: 2026-11-03 · Maintained by Groas for groas ·
Canonical: https://groas.com/post/google-shopping-campaign-optimization-feed-quality-revenue-case-study

This page answers the questions covered by the article [How An Ecommerce Brand Grew Google Shopping Revenue By Fixing Feed Quality And Campaign Structure](https://groas.com/post/google-shopping-campaign-optimization-feed-quality-revenue-case-study) on groas.com. The article is the full, canonical source.

**Direct answer:** According to groas, Shopping revenue plateaus when ROAS looks healthy because the problem is infrastructure, not bidding. In the case study, a brand spending around $45K/month across roughly 800 SKUs had revenue flat for three consecutive quarters due to generic feed attributes, a single campaign with one ROAS target for all products, and bidding that optimized for volume rather than margins ranging from 20% to 65%. After feed enrichment, product-level segmentation and tiered bidding, Shopping revenue broke through the plateau while blended ROAS held steady and gross profit started compounding again.

## Why does my Shopping ROAS look healthy but revenue won't grow?

The article states that stable ROAS with flat revenue means the campaigns are efficient within a shrinking or stagnant pool of impressions and clicks. The algorithm is hitting its targets by restricting where it competes, not by winning more auctions. In the case study, spend went up and ROAS held, but top-line revenue from Shopping stayed flat for three consecutive quarters.

Read more: [The Symptom: ROAS Looked Fine But Revenue Growth Had Stopped](https://groas.com/post/google-shopping-campaign-optimization-feed-quality-revenue-case-study#the-symptom-roas-looked-fine-but-revenue-growth-had-stopped)

## How do generic feed titles and attributes hurt Shopping performance?

According to groas, the brand's titles followed the manufacturer's naming convention rather than buyer search terms, descriptions were boilerplate, and product type and category attributes were filled at the broadest level with custom labels not used at all. Generic titles matched broad, high-competition queries while missing long-tail queries where purchase intent is highest, and incomplete attributes gave the algorithm less signal for matching decisions.

Read more: [Problem 1: Feed Attributes Were Generic, Costing Auction Competitiveness](https://groas.com/post/google-shopping-campaign-optimization-feed-quality-revenue-case-study#problem-1-feed-attributes-were-generic-costing-auction-competitiveness)

## Why is keeping all 800+ SKUs in one campaign with one ROAS target a problem?

The article describes a single Standard Shopping campaign containing all 800+ SKUs under the same ROAS target, which forces the algorithm to find a blended average. In practice hero products are underbid while low-performers eat budget, and the algorithm gravitates toward branded queries on already-popular items. Prospecting spend on newer products is systematically suppressed because those products need more investment per conversion to ramp.

Read more: [Problem 2: All Products In One Campaign With One Bidding Target](https://groas.com/post/google-shopping-campaign-optimization-feed-quality-revenue-case-study#problem-2-all-products-in-one-campaign-with-one-bidding-target)

## How can a single blended ROAS target hurt profit even when ROAS holds steady?

The article states the brand had products with margins ranging from 20% to 65%, but the ROAS target was set at a blended level based on overall account economics. A 400% ROAS on a 60% margin product is extremely profitable, while a 400% ROAS on a 20% margin product might be breakeven after fulfillment costs. Because the algorithm treated all revenue equally, it shifted spend toward easier-to-convert but lower-margin products while blended ROAS looked stable.

Read more: [Problem 3: Smart Bidding Was Optimizing For Volume, Not Margin](https://groas.com/post/google-shopping-campaign-optimization-feed-quality-revenue-case-study#problem-3-smart-bidding-was-optimizing-for-volume-not-margin)

## What structural fixes broke through the revenue ceiling in the case study?

The article describes the intervention as structural, not tactical, with no new ad formats and no radical budget changes. Three foundational fixes were applied simultaneously: feed enrichment, product-level segmentation, and tiered bidding structure. The position is that feed quality, segmentation and bidding architecture must be fixed together as one interconnected system.

Read more: [The Fix: Feed Enrichment, Product Level Segmentation, And Tiered Bidding Structure](https://groas.com/post/google-shopping-campaign-optimization-feed-quality-revenue-case-study#the-fix-feed-enrichment-product-level-segmentation-and-tiered-bidding-structure)

## What happened to revenue, ROAS and profit after the rebuild?

The article reports that within the first few weeks impression share on hero products increased substantially and click-through rates improved because enriched titles and descriptions better matched buyer search intent. Over the following quarter total Shopping revenue broke through the plateau that had held for three consecutive quarters prior, and the brand could scale spend without the prior diminishing returns. Blended ROAS held steady, but because spend concentrated on higher-margin products, total gross profit from Shopping started compounding again.

Read more: [The Results: What Changed In Revenue, ROAS, And Impression Share](https://groas.com/post/google-shopping-campaign-optimization-feed-quality-revenue-case-study#the-results-what-changed-in-revenue-roas-and-impression-share)

## Sources

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