How to calculate ROAS: break-even, target and blended

Almost every store knows its ROAS, and most owners cannot say whether it is good. That is because ROAS without your own margin is meaningless: a ROAS of 3.0 can be solid business or a guaranteed loss, depending on what is left of every dollar of revenue. This article covers the calculation, the break-even point for your margin, the measurement traps, and when you should optimize for profit instead of revenue.

The calculation

ROAS = revenue from advertising ÷ ad spend

Example: $3,000 in ad spend, $12,000 in attributed revenue. ROAS = 12,000 ÷ 3,000 = 4.0. For every dollar spent, $4 of revenue comes back.

Some people write ROAS as a percentage (400%), others as a multiple (4.0). Same meaning, different notation.

In Google Ads the column is called Conv. value / cost, the total conversion value divided by the total cost of all ad interactions. Google itself does not label it ROAS; it describes it as an estimate of your return on investment. Meta Ads Manager reports the same idea as a purchase ROAS metric, and the exact label depends on whether it counts website or app purchases, so check the definition in the column picker rather than trusting the name. In both tools you may have to enable the column first.

One thing about the denominator: use the full cost of the channel. If you pay agency or tool fees, they belong in there, otherwise you are flattering the channel.

Break-even ROAS: the number that actually counts

ROAS on its own only says how much revenue comes back. Whether money is left depends on your gross margin. Break-even ROAS is the point where advertising exactly covers its own cost.

Break-even ROAS = 1 ÷ gross margin

Example: a product sells for $100 net, cost of goods is $60. Gross margin 40%, so 0.40. Break-even ROAS = 1 ÷ 0.40 = 2.5.

At a ROAS of 2.5 the advertising exactly pays for itself: $1,000 of spend brings $2,500 of revenue, of which $1,000 is gross profit. Net zero.

The right-hand column uses 1 ÷ (margin − target profit share). At a 40% margin and a 10% target profit: 1 ÷ 0.30 = 3.3.

That table explains why benchmark numbers from forums are useless. A store with a 65% margin celebrates a ROAS of 2.0; a retailer with a 22% margin loses money at the same value.

What belongs in your gross margin

Most miscalculations happen here, not in the ROAS formula. Include every variable cost per order:

• Cost of goods, including freight and duty
• Shipping cost minus shipping income
• Packaging material
• Payment fees, which differ noticeably between methods
• Return costs: expected return rate × return handling cost × the share that cannot be resold
• Commissions, if you sell through marketplaces or affiliates

Rent, salaries and software do not belong in there. Those are fixed costs that occur without the order. The figure you get this way is your contribution margin, and it is the only sound basis for advertising decisions. How it relates to your other numbers is in the overview of the ecommerce metrics that matter.

Worked example with returns: product $100 net, cost of goods $60. Return rate 15%, return handling $6, and 20% of returned goods cannot be resold. Expected return cost per order: 0.15 × (6 + 0.20 × 60) = $2.70. The real margin falls from 40% to roughly 37%, and break-even ROAS rises from 2.5 to about 2.7.

Platform-reported versus blended ROAS

The ROAS in your ad account and the ROAS you can verify in your bank account are never the same number. Three causes are worth knowing before you start adjusting campaigns.

Attribution timing. Google Ads credits a conversion to the day of the click, not the day of the order. With longer consideration cycles that shifts ROAS across days, which is why the last seven days regularly look worse than they end up being.

Modelling and cross-device. Platforms estimate conversions they cannot measure directly, across devices or after consent was declined. That is methodologically defensible, but it produces more conversions than your store knows about. Google Ads offers two attribution models, data-driven and last click, and switching between them changes how credit is spread across keywords and campaigns without changing a single order.

Double counting between channels. If Google Ads and Meta both claim the same order, the sum of channel revenues exceeds your total revenue. That is normal and not an error, but you must not add the numbers up.

The practical consequence: use platform ROAS to compare inside one channel, meaning which campaign runs better than another. Use blended ROAS to judge the whole picture.

Blended ROAS = total revenue ÷ all marketing costs

This number does not lie, because it has nothing to attribute. It answers the question that matters at the end of the month: is my marketing carrying its weight overall? The trap is the mirror image of the platform view: blended ROAS credits advertising with revenue that would have arrived anyway, from organic search, direct traffic and returning customers. So read them together. Platform ROAS up while blended ROAS is flat or down usually means attribution shifted, not that performance improved.

From ROAS to POAS

POAS = contribution margin ÷ ad spend

ROAS treats every dollar of revenue as equal. In reality your products have different margins, and advertising does not automatically sell the high-margin ones.

Example: two campaigns, $1,000 spend each.

Campaign A: $4,000 revenue, ROAS 4.0, but clearance stock at a 20% margin, so $800 contribution margin. POAS 0.8, a loss.

Campaign B: $2,500 revenue, ROAS 2.5, products at a 55% margin, so $1,375 contribution margin. POAS 1.375, a profit.

Judged on ROAS you would scale A and switch off B. Exactly wrong.

A POAS of 1.0 is break-even. For POAS you need cost prices or margins per product in your system. If that does not exist, there is an intermediate step: calculate POAS manually once a quarter for your ten highest-revenue products. That is usually enough to see whether your advertising sells the right things.

A second reason pure ROAS steers too conservatively: it only values the first purchase. If a new customer buys twice on average, you are allowed to sit below break-even on the first order. Lifetime value gives you that frame, and it tells you how much a new customer may cost.

Step by step: determine your target ROAS

1. Calculate your contribution margin rate. For the whole store and separately for your three most important categories, with every variable cost from the list above.
2. Form break-even ROAS. 1 ÷ contribution margin rate. That is the hard floor.
3. Set a profit requirement. How much should be left of every advertising dollar? Formula: 1 ÷ (margin − target profit share).
4. Account for new-customer value. If you know your repeat rate, the target ROAS for acquisition campaigns may sit below break-even. Without a defensible number, skip this step.
5. Check tracking before you steer. Compare conversions in Google Ads with orders in your store. If conversions are missing, every bidding strategy builds on wrong data. The symptoms and the fixes are in clicks but no conversions in Google Ads.
6. Enter the target and check the data volume. Target ROAS only works with enough conversions per campaign and month. With few conversions, maximize conversions with a budget cap is more stable.
7. Check monthly against blended ROAS. If platform ROAS rises while blended ROAS falls, only the attribution moved.

For Performance Max campaigns the same logic applies, with one addition: because the campaign type mixes channels and can include brand search, its reported ROAS tends, as an experience value, to sit above what an equivalent standard campaign would show. So do not compare the two numbers directly.

Checklist when ROAS suddenly drops

Work top to bottom. In most cases you find the cause in the first three points.

• Tracking: does conversion tracking still count every order? Check after every theme or app update.
• Brand share: has the share of queries containing your store name fallen? Brand search has the best ROAS and lifts the average.
• New search terms: are irrelevant queries eating budget? The search terms report shows it.
• Order value: has average order value dropped? Then it is not an ads problem but a store topic.
• Competition and click prices: has cost per click risen at an unchanged conversion rate?
• Season: compare against the same period last year, not against last month.
• Availability: are the advertised products out of stock or dropped from the feed?
• Landing pages: do ads still lead to the right page, or into a redirect?
• Budget: did a campaign start running out mid-day? See limited by budget in Google Ads.

A note on budgets, because it affects the daily ROAS you read: Google states that a campaign might spend up to twice its average daily budget on a given day, and that over a month you will have spent no more than 30.4 times the average daily budget. Single days therefore look more volatile than the account really is.

What you cannot conclude from ROAS

ROAS says nothing about scalability. A ROAS of 8.0 on a $300 monthly budget does not mean $3,000 will deliver the same. ROAS almost always falls as you raise budget, because you move into less purchase-ready demand. So the practical question is not how high is my ROAS, but up to which budget does my ROAS stay above break-even. The structural side of that question, campaign types and budget allocation, is covered in the guide to Google Ads for ecommerce.

It also says nothing about whether advertising created the revenue or merely collected it. Brand campaigns on your own store name show high ROAS values because those customers would have found you anyway.

And it cannot replace a judgement about the store itself. If the ROAS is weak because visitors arrive and do not buy, the work belongs on the ecommerce conversion rate, not in the bidding strategy.