Performance Marketing

POAS vs ROAS: Which One Should Your Campaigns Bid To?

ROAS counts revenue, POAS counts profit. See how each is calculated, when bidding to profit beats bidding to revenue, and how to send margin data to ads.

TL;DR
  • ROAS divides revenue by ad spend; POAS divides profit by ad spend. Two campaigns with the same ROAS can make very different amounts of money if the products they sell carry different margins.
  • Bidding to ROAS is fine when margins are similar across your catalog. When margins vary widely, or discounts and shipping costs swing from order to order, bidding to profit stops the algorithm from chasing low-margin revenue.
  • To bid to POAS, send profit instead of revenue as the conversion value, reset your targets to the new scale, and keep a revenue-valued conversion for reporting. Do it gradually and with clean cost data.

What is the difference between ROAS and POAS?

Return on ad spend (ROAS) tells you how much revenue each ad dollar brought in. Profit on ad spend (POAS) tells you how much profit each ad dollar brought in.

  • ROAS = attributed revenue ÷ ad spend
  • POAS = attributed profit ÷ ad spend

The word “profit” needs a definition before POAS is useful. Most stores use one of two:

  • Gross profit: revenue minus cost of goods sold. Easy to get, because Shopify stores a “cost per item” on each variant. It is the gross margin view.
  • Contribution profit: revenue minus every variable cost of the order: product cost, shipping, fulfillment, payment fees and a returns allowance. This is the contribution margin view, and it is the more honest one.

With contribution profit in the numerator, a POAS of 1.0 means the first order exactly paid for the ads. Above 1.0, the order made money after ad spend. Below 1.0, it lost money unless repeat purchases make up the gap. Some teams subtract ad spend from profit first, which moves break-even to 0. Either convention works. Pick one and use it everywhere.

The two metrics are linked by a simple identity: POAS = ROAS × margin. A campaign at 4.0 ROAS selling products with a 30% contribution margin runs at 1.2 POAS.

Same ROAS, different profit: an illustrative example

Two campaigns each spend $1,000 and each bring in $4,000 in revenue. These numbers are for illustration only.

Campaign A Campaign B
Ad spend $1,000 $1,000
Revenue $4,000 $4,000
ROAS 4.0 4.0
Contribution margin of products sold 50% 20%
Contribution profit $2,000 $800
POAS 2.0 0.8
Profit after ad spend +$1,000 −$200

In a ROAS report they look identical. In your bank account, one makes $1,000 and the other loses $200.

Why does bidding to ROAS overspend on low-margin products?

Automated bidding optimizes toward the conversion value you send it. If that value is revenue, the algorithm treats every revenue dollar as equal. A $100 order of a 15%-margin product looks just as good as a $100 order of a 60%-margin product.

Over time, the algorithm drifts toward whatever converts most easily at your target. That is often:

  • Low-margin hero products that sell on price.
  • Items on heavy promotion, where the discount already ate the margin.
  • Bundles priced aggressively to raise average order value.
  • Heavy or bulky products whose shipping costs aren’t visible in revenue.

The platform is doing what you asked. Revenue is simply the wrong instruction when margins vary.

There is a second, quieter issue. A single ROAS target applied to a mixed catalog is always wrong for most products. If your blended break-even ROAS is 3.0 but some product lines need 5.0 to break even, a 4.0 target loses money on those lines every day. Our guide to break-even ROAS shows how to calculate the floor for each line.

When should you switch from ROAS to POAS bidding?

POAS is more accurate but takes more work. Switch when the extra accuracy changes decisions.

Your situation ROAS bidding POAS bidding
Margins are similar across products (within a few points) Works well; set the target from break-even ROAS Little extra benefit
Margins vary widely across the catalog Overspends on low-margin items Better fit
Frequent promotions or sitewide discounts Treats discounted orders as full value Better fit
Shipping cost varies a lot by product Hides heavy-item losses Better fit, if shipping is in the cost data
No reliable product cost data Only option for now Not ready yet
Low conversion volume per campaign Simpler, more stable Possible, but test carefully

If margins are fairly uniform, calculate break-even ROAS from your margin, set targets above it and review against store profit.

How do you bid to profit in Google Ads and Meta?

Ad platforms have no “POAS” setting. You bid to profit by changing the value.

Option 1: Send profit as the conversion value

At the moment of purchase, calculate the profit of the order and send that number as the conversion value instead of revenue. The calculation needs a cost per SKU, plus your estimates for shipping, fees and returns if you want contribution profit.

Where to do the calculation:

  • On the server, not in the browser. If you put product costs in your site’s data layer, anyone can read your margins in the page source. A server-side container (see our guide to server-side tagging for Shopify) can look up costs privately and attach profit to the event before it reaches the ad platform.
  • Through offline or server imports. Some stores upload order-level profit after the fact, including refunds.

In Google Ads, keep a revenue-valued purchase conversion as a secondary action for reporting, and make the profit-valued one the primary action that bidding uses. Your “target ROAS” setting then works as a target POAS: a target of 150% means $1.50 of profit per $1 of spend.

For Meta, value-based bidding works on the purchase value you send through the pixel and the Conversions API. Shopify’s native Meta integration sends order value, so sending profit usually means a custom server-side setup. Meta’s in-platform ROAS will then show profit-based numbers, so label your reports clearly.

Option 2: Keep revenue bidding, but split campaigns by margin

If you can’t change the conversion value yet, put margin into your structure instead. In Google Merchant Center, assign each product a custom label such as “margin-high,” “margin-mid” and “margin-low” in your product feed. Build separate Shopping or Performance Max campaigns per label, each with a target ROAS set from that group’s break-even point.

It is cruder than profit bidding, but it stops one blended target from governing very different economics.

Reporting without changing bidding

Google Ads also offers “conversions with cart data.” When you pass the items in each order and add cost of goods sold to your Merchant Center feed, Google Ads can report gross profit by campaign. Google describes this as reporting and insight. It is a good way to see which campaigns are profitable before you change how they bid.

What can go wrong when you bid to POAS?

Profit bidding fails in predictable ways.

  • Stale cost data. If supplier costs rise and your cost table doesn’t, the algorithm bids on profit that no longer exists. Assign an owner and update costs on a schedule.
  • Keeping the old target. If you switch values from revenue to profit and leave the target at 400%, you are now asking for $4 of profit per $1 of spend. Volume collapses. Reset the target first (see the worked example below).
  • Ignoring refunds. Profit sent at checkout doesn’t know about returns. Include a returns allowance in the calculation, or adjust values afterwards where the platform supports it.
  • Starving customer acquisition. First-order profit undervalues products that start long, profitable relationships. A low-margin entry product with a strong first-to-second order rate can be worth more than its first order suggests. Use Google Ads’ new customer acquisition goal, which can bid higher for new customers than existing ones, or set a lower POAS floor for proven entry products. Our article on the first-to-second purchase rate explains how to find those products.
  • Changing everything at once. Move one campaign, watch it for a few weeks, then roll out.

Worked example: moving a campaign from ROAS to POAS

All numbers below are illustrative.

A Shopping campaign spends $10,000 a month at a 400% target ROAS. Over the last 90 days, the products it sold carried an average contribution margin of 35%.

  1. Translate the current target. 4.0 ROAS × 0.35 margin = 1.4 POAS. The campaign is already running at roughly 140% POAS on average.
  2. Set the starting POAS target close to today’s level. Start at 140%, not 400%, so volume holds while bidding learns the new values.
  3. Switch the value. Make the profit-valued purchase conversion primary for this campaign and keep the revenue one as secondary.
  4. Let it learn. Leave the target alone for a few weeks. Spend should shift toward higher-margin products.
  5. Move toward your real goal. If you need 160% POAS to cover fixed costs, raise the target in small steps and watch volume.
  6. Judge the result in profit. Compare contribution profit after ad spend before and after, not ROAS. Revenue can fall a little while profit rises.

Checklist: are you ready to bid to profit?

  • [ ] Every active product has a current cost per item in Shopify (or your platform).
  • [ ] You have chosen one definition of profit (gross or contribution) and written it down.
  • [ ] Shipping, payment fees and a returns allowance are estimated per product or per order.
  • [ ] Profit is calculated on the server, not exposed in the browser.
  • [ ] A revenue-valued conversion stays in place for reporting.
  • [ ] New targets are translated from current ROAS × margin before the switch.
  • [ ] One campaign is tested before the whole account moves.
  • [ ] Entry products with strong repeat rates have a protected, lower floor.

FAQ

Is POAS better than ROAS?

It’s more accurate when margins vary, because it measures what you keep instead of what you bill. If your margins are similar across products, a ROAS target set from break-even ROAS gets you most of the way with less work.

What is a good POAS?

Using contribution profit, anything above 1.0 means the first order paid for its ads. The right target depends on your fixed costs and how much repeat revenue a new customer brings, so set it from your own numbers rather than an industry figure.

Can I use POAS in Meta Ads?

Yes, by sending profit instead of revenue as the purchase value, usually through a custom Conversions API setup. Meta will then report profit-based “ROAS,” so label dashboards clearly to avoid mixing the two.

Turn this into a plan for your store

The Free Profit Audit scans your store, sizes your biggest profit leaks and builds a 90-day plan.

Free Profit Audit