ROAS vs POAS: Which Metric Should You Actually Optimize For?

ROAS vs POAS: Which Metric Should You Actually Optimize For?

ROAS (Return on Ad Spend) is the metric every ad platform reports by default. POAS (Profit on Ad Spend) is the metric that actually tells you whether you're making money.

Here's the complete breakdown — what each measures, when to use which, and why the difference matters more than most advertisers realize.

The Definitions

ROAS = Revenue ÷ Ad Spend

A 5x ROAS means you generate $5 in revenue for every $1 spent on ads. Standard, widely used, easy to calculate.

POAS = Gross Profit ÷ Ad Spend

Gross Profit = Revenue − Cost of Goods Sold (COGS)

A 2x POAS means you generate $2 in gross profit for every $1 spent on ads.

The Critical Difference: Revenue vs. Profit

The Example That Makes It Click

Two campaigns. Same ROAS.

Campaign A:

• ROAS: 6x ($60,000 revenue / $10,000 spend)
• Product margin: 15%
• Gross profit: $9,000
• POAS: 0.9x ← losing money

Campaign B:

• ROAS: 3x ($30,000 revenue / $10,000 spend)
• Product margin: 55%
• Gross profit: $16,500
• POAS: 1.65x ← profitable

If you optimize for ROAS, you scale Campaign A and pause Campaign B. You're scaling a money-losing campaign and cutting a profitable one.

This is not a theoretical scenario. This happens constantly in eCommerce accounts with mixed-margin product catalogs.

When ROAS Is Enough

ROAS is sufficient when:

• Your product catalog has uniform margins (all products have roughly the same margin %)
• You're in early-stage testing where comparing channels directionally is what matters
• You don't yet have reliable COGS data broken down by product

ROAS as a directional indicator is fine. ROAS as a profitability optimization target is dangerous when margins vary.

When You Need POAS

POAS matters when:

• You sell products with significantly different margins (a $20 item vs. a $200 item with different costs)
• You have accessories, bundles, or subscription upgrades with different margin profiles
• You've scaled beyond $50k/month in ad spend and every point of margin matters
• Your campaigns are optimizing toward high-revenue / low-margin products

How to Implement POAS Tracking

Step 1: Get your COGS data by product SKU (from your accounting system, Shopify, or ERP)

Step 2: Calculate gross profit per conversion:

Gross Profit = Revenue × (1 − COGS %)

Step 3: Pass gross profit as the conversion value to Google Ads and Meta (instead of revenue)

Step 4: Set your Target ROAS bid in Google/Meta based on POAS, not revenue ROAS

• Target ROAS (in platform) = 1 ÷ (Target Net Margin Ratio)
• But input gross profit as the conversion value, so the platform is effectively optimizing for POAS

Step 5: Monitor POAS weekly alongside ROAS

The Full Metrics Stack: MER + ROAS + POAS

The three metrics serve different levels of your analytics:

MER (Total Revenue / Total Spend) — Strategic level. Is my overall marketing investment efficient? This is the board metric.

ROAS — Campaign level. Which campaigns and channels are generating the most revenue per dollar? This is the monitoring metric.

POAS — Product/campaign profitability level. Which campaigns are actually profitable after accounting for margins? This is the optimization metric.

Use all three together. If MER is declining while ROAS looks stable, you have an attribution or mix problem. If ROAS is strong but POAS is weak, you're over-indexed on low-margin products.

What Is a Good POAS?

A POAS above 1x means you're generating positive gross profit from advertising. Targets by business type:

Note for SaaS/subscription: Short-term POAS can be below 1x if you have strong LTV data justifying payback period. This is where LTV:CAC analysis becomes essential.

The Practical Recommendation

Start with ROAS — it's what every platform natively supports and it's a reasonable directional metric.

Move to POAS optimization once:

1. You have reliable margin data by product or campaign
2. You're running a mixed-margin catalog
3. You've scaled enough that margin leakage is meaningful

The transition isn't difficult — it requires margin data and the willingness to pass profit values rather than revenue values as conversion signals.