Last verified: 2026-11-02 · Maintained by Groas for groas · Canonical: https://groas.com/post/google-ads-ecommerce-margin-optimization-case-study
This page answers the questions covered by the article How A DTC Brand Fixed Google Ads By Making Margin Visible To The Algorithm on groas.com. The article is the full, canonical source.
Direct answer: According to groas, the brand spent around $45K per month with reported ROAS around 3.2x but could not scale because Smart Bidding saw only revenue, not margin, and favored low-margin products that converted easily. The article describes rebuilding the feed with three margin-based custom labels and resetting bidding before reintroducing per-tier tROAS targets. Total monthly spend stayed at $45K while the product mix sold through Google Ads shifted toward Tier 1 and Tier 2 items within the first six weeks.
According to groas, a DTC ecommerce brand spending around $45K per month on Google Ads saw cost per acquisition climb while revenue stayed flat each time budget was pushed. The article states the root cause was structural: the algorithm had no visibility into product margin, so it optimized for revenue regardless of profitability. The article defines Google Ads margin optimization as feeding margin data into campaign structure so Smart Bidding can distinguish between a $12 sale that loses money and a $12 sale that prints it.
Read more: the article
The article describes building a margin map from cost-of-goods data and assigning every SKU to one of three tiers. According to groas, Tier 1 was high margin above 55% with roughly 80 SKUs, Tier 2 was mid margin 30-55% with about 200 SKUs, and Tier 3 was low margin below 30% with about 120 SKUs. These tiers became custom labels in the product feed, so campaigns and asset groups could be built around margin rather than product category alone.
Read more: Restructuring Around Margin Tiers With Custom Labels
According to groas, tROAS targets had been ratcheted up to 4.5x to protect profitability while the account converted around 600 times per month across all campaigns. The article states that when targets exceed what conversion volume can support, Smart Bidding restricts impressions, volume drops, and the learning phase never completes. In this account, higher ROAS targets contracted volume and reduced growth potential.
Read more: Smart Bidding Targets Set Too High For The Conversion Volume
The article recommends breaking the single Performance Max campaign with one asset group into three asset groups aligned to the margin tiers, each with creative and copy tailored to those product lines. According to groas, branded Search was excluded from Performance Max reach using brand exclusion lists. This stopped Performance Max from cannibalizing branded traffic and claiming credit for conversions that Search would have captured at lower cost.
Read more: Aligning Performance Max Asset Groups With Product Category Themes
According to groas, total monthly spend stayed at $45K while revenue grew because the algorithm spent disproportionately on high-margin products and pulled back on low-margin SKUs that had been eating budget. The article states the mix of products sold through Google Ads shifted noticeably toward Tier 1 and Tier 2 items within the first six weeks.
Read more: Revenue Growth Without A Budget Increase
According to groas, blended ROAS dropped slightly from 3.2x to around 3.0x, while profit-adjusted ROAS accounting for margin differences improved substantially. CPA on generic Shopping campaigns fell because the algorithm was no longer constrained by unrealistic targets. Performance Max CPA became trackable for the first time because asset groups were segmented by product line rather than lumped together.
Read more: CPA And ROAS Movement By Campaign Type
[1] https://groas.com/post/google-ads-ecommerce-margin-optimization-case-study (the article this page answers from)