Target ROAS Calculator

The maths behind it

Target ROAS is the return required to hit a stated profit objective, rather than merely to break even: target ROAS = 1 ÷ (contribution margin − target profit margin on ad-driven revenue).

Equivalently, if you want advertising-driven revenue to contribute a given amount after spend, the target is contribution needed ÷ contribution per dollar of revenue.

It is also the name of an automated bidding strategy on most platforms, where the number entered becomes the algorithm's objective.

In practice

At a 55% contribution margin with a target of 15% profit on ad-driven revenue: 1 ÷ (0.55 − 0.15) = 2.50.

So a 2.5 ROAS delivers a 15% profit margin on that revenue, against the 1.82 break-even.

At $4,000 of spend, a 2.5 ROAS produces $10,000 of revenue, $5,500 of contribution and $1,500 of profit, exactly the 15% intended.

Setting the target at 3.5 instead would produce a higher margin on less volume, because higher targets restrict the auctions the campaign can win. That trade between rate and volume is the whole of target-setting.

The limitations

A higher target does not simply improve profitability; it reduces reach. Automated bidding hits a high target by bidding only on the cheapest, most certain conversions, which are often the ones that would have happened anyway.

That produces an excellent reported ROAS and very little incremental revenue, which is the most common failure mode of target-based bidding.

Putting it to use

Set the target from your own margin arithmetic rather than from a platform recommendation, and change it in modest steps. Large target changes reset the algorithm's learning and produce weeks of unstable performance.

Then judge the result on total contribution rather than on whether the target was met. A campaign hitting a 4.0 target on $800 of spend is worth less than one hitting 2.5 on $6,000.

The volume-versus-efficiency trade

Every ROAS target sits somewhere on a curve: low targets buy volume at thin margins, high targets buy efficiency at low volume. Neither end maximises profit.

Profit is maximised where the marginal order still contributes, which is at break-even ROAS on the last dollar spent, not on the average. Because reported ROAS is an average across the whole campaign, an account hitting exactly break-even on average is losing money on its marginal spend.

The practical approach is to move the target down in steps while watching total contribution rather than ROAS. Contribution rises, plateaus and then falls, and the plateau is where the account should sit. Most advertisers never find it because they optimise the ratio rather than the money.

One practical constraint: automated bidding needs a stable target to learn against, so a target changed weekly produces permanent instability rather than continuous improvement.

Changing it by no more than 15% to 20% at a time and then leaving it for at least a fortnight is what allows the algorithm to settle, and accounts that adjust more frequently than that are usually measuring the disruption rather than the strategy.

The target should also be reviewed whenever the product mix shifts, since the catalogue average it was derived from will have moved with it.