Definition
Profit on ad spend (POAS) is the gross profit generated by advertised orders divided by the ad spend that produced them. Where ROAS counts revenue, POAS first strips out cost of goods, shipping, payment fees and other per-order costs, then compares what is left with the media cost. A POAS above 1 means the ads returned more profit than they cost; below 1 they consumed margin.
Formula
POAS = (Attributed revenue - Cost of goods and per-order costs) ÷ Ad spend
- Attributed revenue
- Order value credited to the ads by the platform or by the store's own attribution, stated net of discounts
- Cost of goods and per-order costs
- Product cost, packaging, outbound shipping paid by the store, payment processing and an allowance for returns on those orders
- Ad spend
- Media cost for the same campaigns and period
Two campaigns with the same ROAS and different POAS
- Campaign 1 spend
- $10,000
- Campaign 1 attributed revenue
- $40,000 (ROAS 4.0)
- Campaign 1 product and per-order costs
- $26,000
- Campaign 1 POAS
- (40,000 - 26,000) ÷ 10,000 = 1.4
- Campaign 2 spend
- $10,000
- Campaign 2 attributed revenue
- $40,000 (ROAS 4.0)
- Campaign 2 product and per-order costs
- $32,000
- Campaign 2 POAS
- (40,000 - 32,000) ÷ 10,000 = 0.8
Illustrative figures. On a ROAS report the two campaigns look identical. On a POAS report the first returns 1.4 dollars of gross profit per ad dollar and the second returns 0.8, which means it is paying to sell products at a loss. The difference is entirely in what the orders contained.
Why it matters
POAS matters because ecommerce catalogues rarely carry one margin. A campaign that sells clearance stock, heavy items with subsidized shipping or third-party brands with thin markups can post a handsome ROAS while draining cash, and a campaign selling high-margin own-brand goods can look mediocre on ROAS while quietly funding the business. Feeding margin data into the bid strategy, so that the platform optimizes toward profit rather than revenue, changes which products get budget. That shift is often the largest structural gain available in an established ad account, and it cannot be found by looking at ROAS.
Where it goes wrong
- Using a store-wide average margin instead of product-level cost: the whole point of POAS is that margins differ, and an average hands the same blind spot back
- Leaving out returns: categories with high return rates give back a share of every attributed order, and profit that later refunds is not profit
- Ignoring the operational cost of the data: POAS needs cost of goods per SKU synced to the ad platform or the analytics layer, and stale cost data quietly produces wrong bids
- Reading POAS without the volume: a campaign with a POAS of 3 on 1,000 dollars of spend earns less total profit than one with a POAS of 1.5 on 40,000 dollars, so the ratio has to be read next to absolute profit
Questions about POAS
How is POAS different from ROAS in practice?
ROAS asks how much revenue came back per ad dollar. POAS asks how much gross profit came back. The two agree only when every product carries the same margin. In a real catalogue they diverge, and POAS is the one that lines up with the bank balance.
Do ad platforms support bidding on profit instead of revenue?
Yes, indirectly. Google Ads and Meta accept custom conversion values, so a store can send gross profit per order as the conversion value instead of revenue and then run value-based bidding against it. The platform still calls the result ROAS, but the target now represents profit.
What POAS should a store aim for?
Any figure above 1 means the ads generated more gross profit than they cost, but overhead, salaries and software still need paying from that profit. The working target depends on what the store's fixed costs demand of each marketing dollar, and it is set from the store's own cost structure rather than from an outside number.