ROAS (Return on Ad Spend) is the most commonly used metric in e-commerce advertising. But optimizing for ROAS alone can actively hurt your business. After working with 100+ e-commerce brands, we have seen how ROAS-obsessed strategies destroy profitability while looking great on dashboards.
The ROAS Illusion
ROAS measures revenue generated per euro spent on advertising. A 5x ROAS means every €1 of ad spend generates €5 of revenue. On the surface, this looks like a healthy return. But ROAS tells you nothing about whether that revenue is actually profitable.
Consider two products:
Product A: Price €100, COGS €90, margin €10 (10%). At 5x ROAS with €20 CPA, you earn €100 revenue but spend €20 on ads against a €10 margin. You lose €10 per sale.
Product B: Price €100, COGS €40, margin €60 (60%). At 3x ROAS with €33 CPA, you earn €100 revenue and spend €33 on ads against a €60 margin. You profit €27 per sale.
The ROAS-optimized algorithm would allocate more budget to Product A because it delivers higher ROAS. But Product B generates actual profit. This is the ROAS illusion — it rewards efficiency in revenue generation without considering the cost structure behind that revenue.
The problem compounds at scale. As you increase spend on high-ROAS but low-margin products, you may grow revenue while simultaneously shrinking profits. We have seen brands double their revenue year-over-year while becoming less profitable.
Contribution Margin: The Metric That Matters
Contribution margin is the revenue remaining after subtracting variable costs: COGS, shipping, payment processing, and advertising. It represents the actual profit contribution of each sale after accounting for the direct costs of making and marketing it.
The formula is straightforward:
Contribution Margin = Revenue - COGS - Shipping - Payment Fees - Ad Spend
For e-commerce, we calculate this at the order level and aggregate by channel, campaign, and product category. This gives you a clear picture of which campaigns generate profit and which destroy it.
The shift from ROAS to contribution margin as your primary KPI changes every optimization decision. Campaign budgets, bidding strategies, product prioritization, and even promotional calendars should be guided by margin contribution rather than revenue efficiency.
To implement this, you need accurate product-level margin data flowing into your analytics and advertising platforms. This requires coordination between your finance team (for COGS data), your operations team (for shipping and fulfillment costs), and your marketing team (for campaign-level spend data).
Implementing Profit-Based Bidding
Both Google Ads and Meta Ads support value-based bidding strategies that can optimize for profit rather than revenue. The implementation requires feeding margin-weighted conversion values to each platform.
Google Ads — Value-Based Bidding: Instead of sending the order total as the conversion value, send the gross margin (revenue minus COGS). This tells Google's Smart Bidding algorithm to prioritize conversions with higher margins. A €60 margin sale becomes more valuable to the algorithm than a €100 revenue sale with only €10 margin.
Implementation options include modifying your data layer to push margin values, using Google Ads conversion value rules to apply category-level margin adjustments, or using a server-side enrichment in sGTM to replace revenue with margin before sending to Google.
Meta Ads — Custom Conversion Values: Similar to Google, you can send margin-adjusted values through the Meta Pixel and Conversions API. The value parameter in the purchase event should reflect gross margin rather than revenue. Meta's delivery algorithm will then optimize for margin-weighted ROAS.
Important caveat: When you switch to margin-based values, your reported ROAS numbers will drop significantly (because the conversion value is lower). This is expected. Communicate this change to stakeholders before implementation to avoid confusion.
Product Segmentation by Profitability
Not all products should receive equal advertising investment. We segment e-commerce catalogs into four quadrants based on margin and demand:
Stars (high margin, high demand): These are your most profitable products. Allocate 40-50% of your ad budget here with aggressive bidding. Use dedicated campaigns and creative that highlight product quality and value.
Cash cows (low margin, high demand): These drive volume but minimal profit per unit. Use them strategically for customer acquisition with strict CPA caps. New customers acquired through commodity products can be nurtured into higher-margin purchases over time.
Hidden gems (high margin, low demand): These products have excellent margins but lack visibility. Invest in awareness and consideration campaigns to build demand. Use content marketing, social proof, and targeted prospecting to drive trial.
Dogs (low margin, low demand): Do not advertise these products. They drain budget without generating meaningful profit or volume. Exclude them from Shopping campaigns and dynamic product ads.
This segmentation should be implemented through custom labels in your product feed, allowing you to create separate campaigns or asset groups for each quadrant with appropriate bidding strategies.
Building a Profit Dashboard
To operationalize profit-based optimization, you need a dashboard that connects ad platform data with financial data. We build these using a data pipeline that merges three sources:
- Ad platform data: Spend, impressions, clicks, and conversions by campaign, ad group, and product
- E-commerce platform data: Orders, revenue, product-level COGS, shipping costs, and return rates
- Financial data: Payment processing fees, fulfillment costs, and overhead allocation
The resulting dashboard shows contribution margin by channel, campaign, and product category. It reveals which campaigns generate profit and which ones only look good on ROAS reports.
Key metrics to track include contribution margin per order, contribution margin ROAS (CM-ROAS), new customer contribution margin, and profit per click. These metrics provide a complete picture of advertising profitability that ROAS alone cannot deliver.
We typically build these dashboards in Looker Studio connected to BigQuery, with automated data pipelines pulling from Google Ads API, Meta Marketing API, and the client's e-commerce backend. The pipeline runs daily, giving the team fresh profitability data for optimization decisions.