Metrics & KPIs

AOV vs POAS: Revenue Size vs Profit Efficiency

In short: Both metrics evaluate order-level economics, but they answer different questions. AOV measures how much a typical order is worth in revenue; POAS measures how much profit that ad spend actually returned after cost of goods. AOV can rise while POAS falls if the bigger orders are lower-margin. AOV needs only revenue and order count; POAS needs a working cost feed. Use AOV to size the opportunity, use POAS to decide whether the spend was actually worth it.

By the AdFlint research team · Fact-checked against current Google and Meta platform behavior · Last reviewed July 2026

AOV

Total revenue divided by the number of orders in a period, the average amount a customer spends per transaction.

It is the lever connecting conversion rate to ROAS: raising it improves returns without winning a single additional customer. Being an average, a handful of large orders can pull it well above what a typical buyer spends. The common misreading is tracking it without the median or product mix behind it, then crediting a promotion-driven swing to a creative change.

Full definition

POAS

Gross profit attributed to advertising divided by ad spend, replacing the revenue numerator in ROAS with revenue minus cost of goods.

It exposes what revenue reporting hides: a catalog where cheap high-margin items and expensive low-margin ones look identical on ROAS but differ completely in contribution. Feeding margin into bidding usually shifts delivery toward better products. The main pitfall is a stale or incomplete margin feed, which produces confident optimization toward numbers that no longer reflect actual product costs.

Full definition

Side by side.

The differences that actually change what happens in your account.

	AOV	POAS
What you divide	Revenue divided by order count.	Revenue minus cost of goods, divided by ad spend.
Data dependency	Order and revenue data only, always available from any storefront.	Needs accurate, current cost-of-goods data per SKU, which many stores don't maintain.
What a rising number tells you	Customers are buying more per basket, through bundles or upsells.	Ad spend converted into more actual profit, not just more revenue.
Blind spot	Says nothing about margin, so a big discounted basket can look identical to a full-margin one.	Without a live cost feed, it confidently reports on numbers that no longer reflect real product costs.
Failure mode	Chasing bigger baskets with discounts, so revenue grows while margin dollars shrink.	An incomplete COGS feed, missing shipping or fees, inflates the profit figure.
Where it reports	Ecommerce platform or analytics order reports.	Ad platform only if margin data is connected, otherwise a spreadsheet blending spend against a profit feed.
Bid strategy tie-in	Not a native bid input, mostly a diagnostic number.	Feeds margin-aware bidding when margin-adjusted values are passed as the conversion value.

What actually separates them.

AOV is calculated from revenue and order count alone; POAS additionally requires cost of goods sold, so it cannot be computed from ad platform data by itself.

Raising AOV through discounts or bundles can lower POAS at the same time, because the extra revenue may carry thinner margin.

AOV is a store-wide or campaign-wide average that ignores ad spend entirely; POAS is inseparable from spend because spend is its denominator.

There is no native Target POAS bid strategy, so profit-aware bidding usually means feeding margin-adjusted values into Target ROAS as a workaround.

AOV moves with product mix and pricing even without a single new customer acquired; POAS moves with product mix, pricing, cost changes, and spend efficiency all at once.

Which one should you use?

Use AOV when

You want to know if a promotion or bundle is getting customers to buy more per order.
You are diagnosing whether a ROAS drop came from lower conversion rate or smaller basket size.
You do not have reliable cost-of-goods data by SKU yet.
You are comparing checkout or merchandising changes, like cross-sell or free-shipping thresholds, independent of ad spend.

Use POAS when

You sell a catalog where margin varies a lot by product and ROAS alone hides that.
You have a maintained cost feed, covering COGS, fees, and shipping, connected to reporting or bidding.
You are deciding which products deserve more budget based on actual profit contribution.
You suspect a winning ROAS campaign is actually selling your lowest-margin items.

Common questions.

Can a campaign have a high AOV but a bad POAS?

Yes, and it happens often. Big basket sizes from bundling low-margin add-ons or discounted items raise AOV while shrinking the profit each order returns, so a campaign can look strong on AOV and weak on POAS at the same time.

Do I need POAS if I already track ROAS?

Only if margin varies meaningfully across what you sell. If every product carries roughly the same margin, ROAS and POAS move together and tracking both adds little; if margins differ substantially across your catalog, ROAS can rank campaigns in the wrong order.

Why doesn't my ad platform show POAS natively?

Ad platforms only know spend and whatever conversion value you send them; they have no visibility into what a product cost you to make or source. To get POAS you either feed a margin-adjusted value back as the conversion value, or calculate it outside the platform against a cost feed.

Can I use AOV to estimate POAS?

Not reliably. AOV tells you order size, not margin, so two campaigns with identical AOV can have very different POAS depending on which products made up those orders.

Or stop choosing between them.

AdFlint picks the setting, writes the ads, and keeps optimizing inside the Google and Meta accounts you already own.

Try AdFlint free See how it works

Related comparisons

All comparisons