What Is ROAS?

ROAS is calculated by dividing the revenue a campaign generates by what was spent on that campaign. The formula is this:

ROAS = Revenue from advertising / Ad spend

The result is expressed as a ratio (say, 4:1, four euros of revenue for every euro spent) or as a percentage (400%). A campaign that spends €1,000 on ads and generates €4,000 in attributed sales has a ROAS of 4:1.

What the formula leaves out is exactly where the confusion starts: the cost of the product sold, staff wages, logistics, returns, or any other business expense. ROAS only compares ad revenue against ad spend. That's why it can sit comfortably next to a business that's actually losing money, something its sibling metric, ROI, does catch, because it subtracts every cost, not just advertising.

ROAS is almost always calculated inside PPC campaigns, pay-per-click, where every click has a measurable price, CPC, which together with the conversion rate determines how much budget it takes to hit a given ROAS.

One detail that's easy to overlook is the attribution window: most platforms credit a sale to an ad if it happens within a set period after the click or impression, say 7 or 30 days. Change that window and the reported ROAS changes with it, even though nothing about the campaign itself moved. Comparing ROAS across two time periods or two platforms only makes sense if the attribution window is the same in both.

An online store launches a PPC campaign with €1,000 in ad spend and generates €4,000 in sales attributed to those ads. The ROAS calculation is this:

ROAS = €4,000 / €1,000 = 4 (a 4:1 ratio, or 400%)

Looking only at ROAS, the campaign looks like a clear win: four euros back for every euro spent. But ROAS never asks what it cost to make or buy what was sold, or what it cost to fulfill those orders. If the cost of goods sold and the associated operating costs (logistics, customer service, returns) add up to €3,500, the actual profit on that campaign looks like this:

Profit = Revenue (€4,000) - Advertising (€1,000) - Goods and operations (€3,500) = -€500

The store lost €500 on a campaign with a 4:1 ROAS. ROI, calculated against total investment (advertising plus goods and operations, €4,500), comes out around -11%: negative, despite ROAS showing what looked like an excellent result. The full mechanics of the ROI calculation, with more examples and industry-specific nuance, live in the dedicated ROI article; the one thing worth keeping from this example is the core idea: ROAS measures revenue, ROI measures actual profit.

The reverse also happens, and it's worth keeping in mind so the wrong lesson doesn't stick. A second campaign from the same store also spends €1,000 but generates only €2,500 in sales, a 2.5:1 ROAS, which looks worse at first glance. But if those products carry a higher margin and their goods and operating costs add up to just €1,200, the real profit comes to €300: €2,500 - €1,000 - €1,200 = €300. With the lower ROAS, this second campaign actually turns a profit, while the first one, with the higher ROAS, lost money. ROAS alone can't tell these two situations apart; only ROI can.

This mismatch shows up more often than expected in businesses with tight margins, like physical-product ecommerce with high manufacturing or sourcing costs. A high ROAS feels reassuring, but only ROI confirms whether that campaign actually left money in the till.

ROAS is the right metric for fast decisions inside a single campaign: comparing two ads, two audiences, or two PPC channels against each other, all sitting on the same cost structure. That's where the simplicity of ROAS pays off, since it can be calculated instantly from the numbers the ad platform already reports, with no need to cross-reference accounting or product margins.

ROAS stops being reliable the moment the question shifts from "which ad performs better" to "does this campaign make sense for the business." That second question needs the real product margin, fixed and variable costs, and every other expense tied to each sale, exactly what ROAS doesn't measure and ROI does. Two campaigns with the same ROAS can produce opposite business outcomes if one sells a product with a 60% margin and the other sells one at 10%.

In services or software businesses, where the marginal cost of each additional sale is low, ROAS and ROI tend to sit close together. In businesses carrying physical inventory and thin margins, the gap between the two can be huge, exactly what the two worked examples in this article show.

The practical rule: ROAS drives day-to-day optimization inside a campaign; ROI decides whether that campaign, that channel, or that product deserves more budget. Tracking both metrics side by side, rather than leaning on just one, avoids exactly the mistake in the worked example above.

• Work out the product margin before setting a ROAS target; a 3:1 ROAS can be profitable at a 50% margin and not enough at a 15% margin.
• Use ROAS to compare campaigns, ads, or audiences against each other, always within the same channel and a similar cost structure.
• Check ROI at least once a month per campaign or product line, not only at quarter close.
• Adjust the ROAS target by season: campaigns with heavy discounts, like Black Friday, come with thinner margins, so the ROAS needed to stay profitable goes up.
• Treat ROAS differently by channel: average cost per click (CPC) and conversion rate vary between Google Ads, Meta Ads, or Amazon Ads, so the same ROAS number doesn't mean the same thing everywhere.
• Factor in returns and checkout discounts when calculating actual revenue, not just the order's gross value.
• Watch the attribution window when comparing ROAS across time periods or platforms; a longer or shorter window changes the result even when the campaign hasn't changed.

• Using ROAS as the only success metric for a campaign, without ever checking it against the business's actual ROI.
• Setting the same ROAS target across every product, ignoring that each one carries a different margin.
• Confusing ROAS with ROI in reports, presenting a revenue ratio as if it were a profitability figure.
• Leaving out returns, checkout discounts, or marketplace fees when calculating the revenue that feeds the ROAS number.
• Cutting the price or piling on discounts to inflate sales and boost ROAS short term, without checking what that does to margin, and in turn to ROI.
• Ignoring the attribution window when comparing ROAS across two different platforms, each with its own rule for what still counts as a sale caused by the ad.