POAS > ROAS: Why Profit on Ad Spend Should Run Your Buying

ROAS is the most-quoted metric in performance marketing and one of the most misleading. It counts revenue, not profit — so a campaign can post a "great" 5x ROAS and still lose money once cost of goods, shipping, payment fees, and discounts come out. POAS (Profit on Ad Spend) fixes that by optimizing to the one number that actually grows your bank account: the money you keep. In 2026, with rising CAC and margin compression across DTC and lead gen alike, optimizing to revenue is how good-looking accounts go quietly broke.

ROAS vs POAS at a glance

• ROAS = revenue ÷ ad spend. Ignores margin entirely. A vanity number that platforms love because it is always big.
• POAS = gross profit ÷ ad spend. Optimizes to money you keep. The number your accountant cares about.

On a catalog with mixed margins, ROAS-led bidding pours budget into your high-revenue, low-margin products — the ones that move volume but barely clear cost. POAS-led bidding redirects spend to the products that are actually profitable, which is almost always a completely different set of SKUs.

A worked example: why a 5x ROAS can still lose money

Metric	Product A (low margin)	Product B (high margin)
Revenue from ads	$10,000	$10,000
Ad spend	$2,000	$2,000
ROAS	5.0x	5.0x
Gross margin	22%	68%
Gross profit	$2,200	$6,800
POAS	1.1x	3.4x

Identical ROAS, wildly different outcomes. Product A barely breaks even before you pay for shipping and a customer service ticket; Product B prints money. A ROAS-only target treats them as twins. POAS tells you the truth — and tells the bidding algorithm where to lean.

The POAS break-even formula every marketer should know

Your break-even POAS is always 1.0 (every dollar of profit covers a dollar of spend). Your break-even ROAS, by contrast, moves with margin: it equals 1 ÷ gross margin %. At a 25% margin you need a 4x ROAS just to break even; at 70% you break even at 1.43x. Most teams set a single blanket ROAS target across the whole account and never reconcile it against margin — which is exactly how budget drifts into unprofitable products while the dashboard looks healthy.

How we operationalize POAS

POAS is only as good as the profit data you can feed back into the platforms. Our build:

1. Calculate true contribution margin per SKU — net of COGS, shipping, payment fees, and average discount.
2. Pass profit as conversion value instead of revenue, using value rules and dynamic value parameters in the conversion tag.
3. Feed it server-side so the signal is clean — see our server-side tracking guide for why browser-only data corrupts value-based bidding.
4. Use offline conversion import for lead gen so closed-deal value, not form fills, trains the algorithm.
5. Bid to value — Maximize Conversion Value / tROAS fed with profit, not Maximize Conversions.

Key takeaways

• ROAS measures revenue; POAS measures profit. Only one pays your salary.
• Break-even POAS is always 1.0. Break-even ROAS = 1 ÷ gross margin.
• Identical ROAS across two products can hide a 3x difference in real profit.
• POAS requires margin data piped into bidding — usually via server-side tracking and offline conversion import.
• The shift typically reallocates 20–40% of budget toward genuinely profitable SKUs.

We make POAS the default operating metric on every account — it is the second layer of our 4-layer profit system and a natural follow-on once tracking is clean. See it applied in our DTC e-commerce growth work, or get your margins audited in a free 48-hour audit.