Glossary/POAS
What is POAS?
POAS (profit on ad spend) is the contribution a campaign produced divided by what it cost, where ROAS divides revenue by the same cost. Because contribution is revenue after cost of goods, marketplace commission, transaction and payment fees, seller-funded vouchers, the free-shipping cost share and fulfillment, POAS is always smaller than ROAS — typically by a factor of three to five on a Thai marketplace cost stack. The practical difference is that POAS has a fixed, interpretable break-even at 1.0, whereas ROAS breaks even at a number that changes with every product’s margin. A campaign at POAS 1.0 returned exactly what it spent; a campaign at ROAS 1.0 lost most of what it spent.
01/Formula
Formula
POAS = contribution attributed to ads / ad spend
= (revenue x contribution rate) / ad spend
= ROAS x contribution rate
Break-even POAS = 1.0, on every product, always.
Break-even ROAS = 1 / contribution rate, different per product.Example
A Shopee campaign: ฿10,000 spend, ฿62,000 attributed revenue. The advertised products retain 27.7% contribution before ads. ROAS = 62,000 / 10,000 = 6.20 POAS = 6.20 x 0.277 = 1.72 Profit after ads = 62,000 x 0.277 − 10,000 = ฿7,174 The same campaign at a 15% contribution rate: POAS = 6.20 x 0.15 = 0.93 -> loses ฿700.
02/In detail
How is POAS different from ROAS?
They differ in the numerator only, and that one substitution is what makes POAS comparable across a catalogue. ROAS puts attributed revenue over ad spend, so its pass mark moves with margin: a 40%-contribution accessory passes at 2.5 while a 12%-contribution electronics SKU needs 8.33, and a shop-wide ROAS target therefore overfunds one and starves the other. POAS puts attributed contribution over the same spend, which normalises out the margin difference and leaves a single line at 1.0 that means the same thing on every product, every campaign and every platform. That is the reason a portfolio decision — which campaign should get the next baht — is far easier to state in POAS than in ROAS.
What is a good POAS?
Above 1.0 means the campaign paid for itself, but 1.0 is not the target — it is the floor, and a floor that flatters. Two corrections push the useful target higher. First, attributed contribution is not incremental contribution: some of the credited sales would have happened without the ad, so the honest numerator is smaller than the reported one. Second, POAS as normally computed is an average across everything the campaign bought, while the decision to spend one more baht depends on the marginal return, which is lower. On our own Shopee panel of 80+ Thai shops, the six dayparts ranged from a POAS of 2.25 to 3.63 on an attributed basis — comfortably above break-even as averages, and silent about whether the next baht in the best window would have returned anything.
POAS reading < 1.0 the campaign consumed more contribution than it produced = 1.0 break-even, before any incrementality adjustment 1–2 positive but thin; check the marginal, not the average > 2 healthy on an attributed basis; still an average
Why do platform dashboards not show POAS?
Because the platform does not hold the inputs. Computing POAS requires unit cost of goods, which no marketplace knows, plus a correct attribution of category commission, transaction and payment fees, seller-funded voucher share, the free-shipping cost split and a returns reserve down to the individual order line. Marketplaces report what they can compute from their own ledger, which is attributed revenue over spend. That is not a conspiracy, but it does mean the metric shown by default is the one whose pass mark the seller has to supply from outside the system — and most do not, which is how a healthy-looking ROAS coexists with a shrinking bank balance.
03/Why it matters
The trap, in one paragraph.
POAS is the only common denominator across a mixed catalogue. Without it, comparing a campaign on a 40%-margin accessory with one on a 12%-margin appliance means comparing two ROAS figures whose pass marks differ by a factor of 3.3 — an arithmetic trap that reliably sends budget to the products least able to carry it. With POAS, both campaigns are read against the same line, and budget allocation becomes a ranking problem rather than a per-campaign judgement call.
Common mistake
Treating POAS above 1.0 as proof a campaign is worth funding. POAS is an attributed average: it credits sales the ads did not cause, and it is held up by the early, cheap, well-converting traffic a campaign buys first. A campaign averaging POAS 1.7 can be losing money on its marginal baht while the reported figure stays comfortably positive.
04/In DataGlass
How POAS is used in DataGlass.
DataGlass computes contribution per order line from reconstructed order economics — cost of goods, category commission, transaction and payment fees, vouchers, free-shipping cost share, fulfillment and a returns reserve — so POAS is available per campaign, per product and per platform rather than as a shop-level estimate. Campaign recommendations are ranked on expected contribution gained, which is the same quantity POAS reports as a ratio.
05/Sources
- [1] Shopee Thailand — seller commission and fee schedule
Commission bands, transaction and payment fees and Shop Voucher mechanics — the deductions that separate the revenue in ROAS from the contribution in POAS.
- [2] DataGlass — Marketplace benchmarks (open dataset, CC BY 4.0)
The measured POAS range of 2.25 to 3.63 across six dayparts, computed on 9.11 million Shopee clicks with each campaign’s own advertised product’s realised margin.
Keep reading