ROAS calculator: How to work out return on ad spend

ROAS is revenue divided by ad spend. Enter any two of the three below and the calculator returns the missing one, so you can grade a campaign that already ran, or work out what revenue a budget has to produce to hit a target.

The optional section is the one worth opening. ROAS is built on revenue, not profit, which means the same number can be excellent for one business and a slow loss for another. Add your gross margin and the calculator shows the profit left after media, plus the break-even ROAS your margin actually demands.

Return on ad spend is the revenue an advertising source produced, divided by what you paid for it.

That is the entire metric. It became the default in performance marketing because it is fast, available inside every ad platform, and comparable across channels. You can read it daily, per campaign, per creative, and act on it before the month closes.

The convention on how to express it is split, which causes needless confusion. A ROAS of 4 means four units of revenue per unit spent. Some teams write it as 4x, others as 400%, and they are the same figure. Anything below 1x, or below 100%, means the campaign returned less revenue than it cost, which is a loss before you have paid for a single product.

What ROAS deliberately excludes is everything except media. No cost of goods, no shipping, no payment fees, no salaries, no agency retainer. That exclusion is what makes it quick and what makes it dangerous.

ROAS = Revenue from ads ÷ Ad spend

Multiply by 100 if you want the percentage form. Rearranged, the two versions you will use when planning rather than reporting:

Revenue = Ad spend × Target ROAS

Ad spend = Revenue ÷ Target ROAS

The second one is how budgets actually get set. A quarter that needs $600,000 of paid revenue at a 3x target requires $200,000 of media, and if that number is larger than the budget, the plan is wrong before it starts.

A ROAS target picked without reference to gross margin is arbitrary. The honest starting point is the ROAS at which the campaign stops losing money:

Break-even ROAS = 1 ÷ Gross margin

A business with a 45% gross margin breaks even at 2.22x. One with a 25% margin needs 4x just to get to zero. Software at 80% margins breaks even at 1.25x.

This single relationship explains most arguments about ad performance. The ecommerce brand celebrating a 4x and the retailer despairing at a 4x are both looking at the same number against completely different cost structures. Read the table:

And note what break-even means here: it covers the cost of goods and the media, and nothing else. Rent, salaries, tooling, and the agency fee all sit below it. A campaign running exactly at break-even ROAS contributes nothing to any of them.

A healthy consumer brand. You spend $25,000 on paid social and it drives $100,000 of attributed revenue at a 45% gross margin.

ROAS = $100,000 ÷ $25,000 = 4.00x

Gross profit on that revenue is $45,000. After the $25,000 of media, $20,000 is left to cover everything else. Break-even was 2.22x, so the campaign cleared its bar with room to spare.

The same 4x, in a thinner business. A retailer runs the identical numbers at a 25% gross margin.

Gross profit = $100,000 × 0.25 = $25,000, which is exactly the media cost.

The campaign returned precisely nothing. Every hour of work, every unit shipped, every customer service ticket was free labour. A 4x ROAS was the break-even point, and nobody noticed because the dashboard was green.

Planning backwards from a target. Your gross margin is 55%, so break-even is 1.82x. You want media to contribute at least $150,000 of gross profit this quarter on a $180,000 budget.

Required gross profit = $150,000 + $180,000 = $330,000

Required revenue = $330,000 ÷ 0.55 = $600,000

Required ROAS = $600,000 ÷ $180,000 = 3.33x

Now the target has a reason behind it, and the number can be defended to whoever signs off the budget.

There is no universal threshold, and any article offering one is guessing on your behalf. What exists instead is a sequence of increasingly demanding bars:

Two structural adjustments matter more than any benchmark. First, a business with genuine repeat purchase can afford a first-order ROAS below break-even, because the second and third orders carry no acquisition cost. Subscription and consumable businesses routinely buy the first order at a loss on purpose. Second, ROAS declines as spend rises. The cheapest, most intent-heavy audience gets bought first, and scaling means paying more for progressively colder demand. A 6x at $10,000 a month is not a promise of 6x at $100,000.

Four metrics that all sound like they answer the same question and do not.

MER, sometimes called blended ROAS, deserves more attention than it gets. Platform-reported ROAS is measured by the platform selling you the advertising, and every platform claims credit for conversions the others also claim. Add up the reported revenue across channels and it routinely exceeds what the business actually made. MER divides total revenue by total marketing spend, which cannot be double counted because there is only one of each. When platform ROAS is rising while MER is flat, the platforms are reallocating credit rather than creating sales.

It uses revenue, not profit. The central flaw, and the reason the calculator above asks for your margin.

It counts correlation as causation. Some of the people who saw an ad and bought would have bought anyway. Only incremental revenue is a true return, and measuring incrementality requires holdout tests rather than a dashboard.

Attribution windows change the answer. A 7-day click window and a 28-day view-through window on the same campaign can produce ROAS figures that differ by a multiple. Nothing about the campaign changed.

It rewards harvesting over building. Branded search and retargeting show spectacular ROAS because they capture demand that already existed. A portfolio optimised purely on ROAS gradually stops creating new demand and then wonders why growth stalled.

It ignores returns. In categories with high return rates, the revenue in the ROAS numerator includes orders that came straight back. Use net revenue after returns or the number is fiction.

It says nothing about scale. A 12x ROAS on $2,000 of spend is a rounding error. Read ROAS alongside the absolute profit it produced, which the calculator above shows.

Divide the revenue attributed to an advertising source by what you spent on it. Spend $25,000 and generate $100,000 and the ROAS is 4.00x. Multiply by 100 if you prefer the percentage form, which would be 400%.

The only defensible answer starts with your gross margin. Divide 1 by your gross margin to get the ROAS at which you break even, then set the target above it by however much gross profit the business needs. A 45% margin breaks even at 2.22x, so 4x is genuinely good. A 25% margin breaks even at 4x, so the same figure is worth nothing.

The ROAS at which revenue covers the cost of goods and the media, with nothing left over. It equals 1 divided by your gross margin. At a 50% margin it is 2.00x, at a 20% margin it is 5.00x.

ROAS divides revenue by media spend alone. ROI divides net gain by every cost involved, including goods, salaries, and fees. ROAS is always the higher number, which is why a campaign can show a strong ROAS and a negative ROI at the same time.

It depends entirely on your gross margin. At 45% margin a 4x leaves real profit. At 25% margin a 4x is exactly break-even and leaves nothing for salaries, rent, or fulfilment. The multiple means nothing without the margin beside it.

ROAS measures one channel using that channel's own attribution. MER, or marketing efficiency ratio, divides total company revenue by total marketing spend. MER cannot be double counted, so it is the better check on whether marketing as a whole is working.

Because the cheapest and most motivated audience gets reached first. Scaling means bidding for progressively colder demand, so the marginal ROAS on each extra dollar falls even when the campaign itself has not changed.