Most ROAS targets I've seen in ad accounts were picked off a benchmark chart or copied from last quarter. Neither one knows what your product costs. Break-even ROAS does. It's the floor, the return where ad spend eats every dollar of margin and leaves zero. Set your targets from that line and the platform's bidding works for your P&L instead of against it.
What break-even ROAS is and how to calculate it
Break-even ROAS is the return on ad spend where the money an order contributes exactly pays for the ad that sold it. Profit is zero. Above that line, every sale makes money. Below it, every sale costs you money, however good the dashboard looks.
The formula. Break-even ROAS = 1 / contribution margin. Contribution margin = (order value - product cost - shipping - payment fees - returns) / order value.
A worked example
Take an $80 average order. The products in it cost $28. Shipping and packing run $8. Card fees are 3%, so $2.40. About 8% of revenue comes back as refunds on stock you can't resell, which takes another $6.40 off the average order.
That leaves $35.20 per order, a 44% contribution margin. Break-even ROAS is 1 divided by 0.44, which is 2.27x. Put another way, you can spend up to $35.20 in ads to win that $80 order and still not lose money. $80 divided by $35.20 lands on the same 2.27x from the other direction.
Now hold an ad account up against it. A campaign reporting 2.0x ROAS looks fine in a weekly report. On this product it spends $40 to win each $80 order, and the order only has $35.20 to give. That's a $4.80 loss on every sale the campaign brags about.
It's the same math the ROAS calculator runs, where the input is labeled gross margin. Here it's broken into line items, because the costs people forget are the ones that move the number most.
Which costs belong in the margin
The rule I use is simple. If a cost goes up when you sell one more order, it goes in. If you'd pay it whether or not the ad ran, it stays out.
| Cost | In the margin? | Why |
|---|---|---|
| Product cost (COGS) | Yes | Every unit sold has to be bought or made. |
| Shipping, packaging, pick and pack | Yes | Paid per order. A free-shipping offer makes all of it yours. |
| Payment processing | Yes | A percentage of each order, often with a fixed fee per transaction on top. |
| Returns and refunds | Yes | Refunded revenue was never really revenue. Apparel and shoes get hit hardest. |
| Per-order app, marketplace, or affiliate fees | Yes | Anything charged per sale scales with the ads. |
| Discount codes | Yes, unless revenue is reported after discounts | Check which figure your conversion value actually passes. |
| Rent, salaries, software | No | Fixed costs come out of total profit, not the per-order line. |
| Management fees for the ad account | No, add them to ad spend | They scale with spend, not with orders. |
Two things trip people up more than the rest. The first is which revenue the platform reports. If your conversion value includes the shipping and tax a customer paid, a $90 reported order might hold $80 of product revenue. Calculate the margin against the same figure the platform reports, or your break-even will sit too low.
The second is returns. Take a 50% margin product with a 25% return rate on stock that can't go back on the shelf. The margin drops to 25%, and break-even doubles from 2x to 4x. That's a different business, and most ROAS targets never account for it.
The calculator treats the returns figure as revenue that's gone for good, with the product cost already spent. If most of your returns get resold at full price, enter a lower number that covers only the cost of handling them.
Break-even ROAS lookup table by margin
This is the table to bookmark. Find your contribution margin on the left. The row gives you the break-even ratio, the number to type into Google Ads, and the target ROAS for a 10% or 20% profit margin after ad spend.
| Contribution margin | Break-even ROAS | Google Ads tROAS | Target for 10% profit | Target for 20% profit |
|---|---|---|---|---|
| 10% | 10.00x | 1000% | Not possible | Not possible |
| 15% | 6.67x | 667% | 20.00x | Not possible |
| 20% | 5.00x | 500% | 10.00x | Not possible |
| 25% | 4.00x | 400% | 6.67x | 20.00x |
| 30% | 3.33x | 334% | 5.00x | 10.00x |
| 35% | 2.86x | 286% | 4.00x | 6.67x |
| 40% | 2.50x | 250% | 3.33x | 5.00x |
| 45% | 2.22x | 223% | 2.86x | 4.00x |
| 50% | 2.00x | 200% | 2.50x | 3.33x |
| 55% | 1.82x | 182% | 2.22x | 2.86x |
| 60% | 1.67x | 167% | 2.00x | 2.50x |
| 65% | 1.54x | 154% | 1.82x | 2.22x |
| 70% | 1.43x | 143% | 1.67x | 2.00x |
| 75% | 1.33x | 134% | 1.54x | 1.82x |
| 80% | 1.25x | 125% | 1.43x | 1.67x |
The shape of the curve matters more than any single row. Going from 60% to 70% margin lowers break-even by 0.24x. Going from 20% to 30% lowers it by 1.67x. Low-margin stores get punished hardest by every extra cost, which is why a 5-point jump in returns can sink a 25% margin business and barely register at 70%.
The Google Ads column is rounded up to the next whole percent, so a target set from it never lands a hair under break-even.
Those profit targets are a share of revenue. If you'd rather think about the return on the ad money itself, the way the ROAS calculator does, multiply break-even by 1 plus the return you want. A 40% margin with a 20% return on ad spend is 2.5 x 1.2 = 3.0x. Both methods work. Pick one and don't mix them inside the same account.
Every target you set in Google Ads or Meta should start from this number and sit above it. A target at break-even asks the algorithm to buy you zero profit as efficiently as it can.
How to set target ROAS in Google Ads and Meta
Break-even is the floor, not the target. How far above it you go depends on what the campaign needs to earn after fixed costs. With the 44% example and a 10% profit goal, the target is 1 divided by 34%, or 2.94x.
Google Ads takes a percentage
Google Ads Target ROAS is entered as a percentage, not a ratio. Google's help page on Target ROAS works it as conversion value divided by ad spend, times 100, so $5 in sales for every $1 of spend is a 500% target. A 2.94x goal goes in as 295% once you round up.
Here's what Google's own documentation says before you switch a campaign over.
- Search and Shopping campaigns need at least 15 conversions in the past 30 days to use Target ROAS. Other campaign types have their own requirements.
- Google suggests reporting conversion values for 4 weeks or 1 to 2 conversion cycles, whichever is longer, before you set a target.
- A new or reactivated bid strategy, campaigns added or removed, or a big change to conversion goals can put the strategy in Learning status. Google says that status never lasts more than 7 days, and ads keep serving during it.
- Changing the target itself doesn't trigger Learning status. The bidder reacts right away but needs 1 to 2 conversion cycles to reach the new number.
- A target set too high may limit how much traffic your ads get.
- When you judge results, leave out the most recent conversion delay period. Late conversions make the last few days look worse than they are.
My approach is to start the target at or slightly under the campaign's actual ROAS from the last 30 days, as long as that's above break-even. Then I raise it in small steps, waiting a couple of conversion cycles between each. Jumping from 250% to 400% in one move usually just shrinks volume. And if actual ROAS is already under break-even, no bid target fixes that. Go back to the margin first.
Meta takes a decimal
Meta's ROAS goal uses the same ratio, written as a decimal. Meta's bid strategy documentation defines minimum ROAS as purchase value divided by spend, so 2.33 means 233%. It's only available when the ad set optimizes for value, which means you're passing purchase values back to Meta. Meta also warns that setting the bar too high may lead to under-delivery.
So one campaign goal ends up looking like two different numbers. 2.94 in Meta and 295% in Google Ads are the same target. The number people get wrong is 1.
A 1x ROAS is break-even. A 1x ROAS means the ads paid for themselves and nothing else. You still bought the product, shipped it, and paid the card fees. True break-even is 1 divided by your margin, and it's above 1x for any business that pays for what it sells.
Already know your spend and revenue? Check your actual ROAS against this line.
Open the ROAS calculatorBreak-even ROAS vs break-even CPA
They're the same line drawn in different units. Break-even CPA is the contribution per order, $35.20 in the example. Break-even ROAS is the order value divided by that, 2.27x. If every order were exactly $80, the two would always agree.
| Metric | Formula | Best for | Example ($80 order, 44% margin) |
|---|---|---|---|
| Break-even ROAS | 1 / contribution margin | Stores with mixed order values | 2.27x |
| Break-even CPA | Order value x contribution margin | Lead gen, one-price offers, single-product stores | $35.20 |
They stop agreeing when order values vary. A store selling a $25 accessory and a $300 bundle can't run one CPA target across both, because at the same margin the bundle can afford twelve times the CPA. ROAS targets scale with the cart, so they handle mixed catalogs better. CPA targets make more sense when every conversion is worth about the same.
Running both gives you a sanity check. If break-even CPA is $35 and Google says you're paying $48 per purchase, the ROAS report will show the same problem as a number under 2.27x. The CPA calculator handles the per-order version, and it's the better fit for lead gen, where there's no order value to divide.
Where break-even ROAS misleads you
The formula's right. The inputs are where it goes wrong. These are the four I've seen do the most damage.
Platform ROAS isn't incremental ROAS
Break-even assumes every dollar of reported revenue was caused by the ad. Branded search and retargeting claim plenty of sales that would've happened anyway. If a campaign reports 3x but half those buyers were coming regardless, incremental ROAS is closer to 1.5x, and that's the number to hold against break-even. Split brand from non-brand before you judge either one.
First-order math ignores repeat buyers
A customer who buys three times in a year is worth more than their first order shows. Break-even on the first purchase is the strictest bar you can set. If your business has a real repeat rate, the better question is what a customer is worth over time and what you can pay to get one. The LTV to CAC calculator covers that side, and the CAC calculator works out blended acquisition cost.
One margin hides a mixed catalog
An account-wide 44% might be 70% on accessories and 25% on the hero product. If the ads mostly sell the hero product, the real break-even is 4x, not 2.27x. When margins vary a lot, run the calculator per product group and set targets by campaign or product group to match.
Contribution isn't profit
Break-even ROAS covers per-order costs only. Rent, salaries, and software still come out of what's left. A business running every campaign at exactly break-even is losing money overall, by the full amount of its fixed costs. That's what the target profit field is for.
When running below break-even is fine
Sometimes it's the right call. It has to be a decision with a limit, though, not an accident nobody's checked.
- You know the repeat rate. If 35% of new customers place a second $80 order within six months, each new customer brings in $47.52 of contribution, not $35.20. First-order break-even drops from 2.27x to 1.68x. Pull the repeat rate from your own order data, and don't guess.
- You're buying data on purpose. A new campaign or product launch needs conversions before Smart Bidding has anything to learn from. Set a budget cap and a date to cut it.
- You're clearing stock. Inventory you'd write off anyway costs close to nothing for this decision. Rerun the calculator with a lower product cost and break-even falls hard.
- The campaign feeds another channel. Prospecting video that grows branded search can look weak on its own numbers. Prove it with a holdout or geo test before you count it.
What's never fine is running under break-even because the target came from a benchmark report. That one's just a slow leak.
The bottom line
Work out break-even ROAS before you set any target. Put your real per-order costs into the calculator, returns and fees included, and use the target ROAS it gives you in Google Ads or Meta. If a campaign's actual return sits under the break-even line, the bid strategy won't save it. Fix the margin, the offer, or the targeting first.