Marginal Return on Ad Spend
Marginal Return on Ad Spend is the incremental revenue generated by the next dollar of advertising spend, as opposed to the average return across all spend in a channel.
Also known as: marginal ROAS, incremental ROAS curve, next-dollar ROAS
Marginal Return on Ad Spend measures the incremental revenue generated by the next dollar of advertising spend, as opposed to the average return across all spend in a channel. It answers a more useful question than average ROAS: whether the next dollar is worth spending, rather than what past dollars averaged. It is the right efficiency lens for any allocation decision.
What Marginal Return on Ad Spend Means
Marginal Return on Ad Spend is the slope of a channel's spend-versus-return curve at the current investment level. Because most channels show diminishing returns as audiences saturate, the marginal return is usually lower than the average return once a channel approaches its efficiency ceiling. A channel with a strong average 5:1 ROAS can have a marginal return of 1.5:1 if it is already saturated, meaning the next dollar would perform far better elsewhere. The metric exists to answer the budget question average ROAS cannot: where should the next dollar go?
How Marginal Return on Ad Spend Works
It is estimated from a response or saturation curve, built through marketing mix modeling or controlled spend tests at multiple levels. The slope of that curve at your current spend level is the marginal return. Without a curve, marginal return is a guess; with one, it becomes a defensible input to budget decisions. The cleanest method is MMM, which fits a curve to historical spend and outcome data per channel. A simpler approach is a stepped spend test: raise or lower spend in defined increments and observe the outcome at each level. Three to four spend levels over six months give a credible curve for most channels.
Common Pitfalls and Misconceptions
The common error is optimizing toward average ROAS, which is the number every ad platform reports by default. Average ROAS rewards channels that scaled well historically; marginal ROAS reveals where the next dollar should go now. The second pitfall is assuming marginal return cannot be negative: in heavily saturated channels, additional spend may produce almost no incremental revenue, meaning the next dollar destroys efficiency even if the overall channel still looks profitable on average. Negative marginal returns are common in branded paid search and retargeting where the channel mostly cannibalizes organic conversion.
Marginal Return on Ad Spend in Practice
The practitioner approach is to reallocate budget until marginal return is roughly equal across all channels. If one channel's next dollar returns 4 dollars and another returns 2 dollars, shift budget from the second to the first until the curves meet. Most growth orgs that do this discover meaningful misallocation, often 20 to 40 percent of budget, that average-ROAS reporting completely hides. Marginal Return on Ad Spend should also be validated against incrementality testing on the largest line items, since a channel can have positive incrementality but a tiny marginal return, meaning it works but is fully scaled, which is a different conclusion than either metric tells alone.
Frequently asked questions
-
Why does marginal return differ from average return?
Average return blends every dollar spent, including highly efficient early dollars that bought reach on best-fit audiences. Marginal return isolates the next dollar, which lands on a more saturated audience and therefore usually performs worse than the average. The gap between average and marginal grows as a channel scales.
-
How do I estimate marginal return?
You estimate it from a response or saturation curve, built through marketing mix modeling or controlled spend tests at multiple levels. The slope of that curve at your current spend level is the marginal return. Without a curve, marginal return is a guess; with one, it becomes a defensible input to budget decisions.
-
How does marginal return guide budget shifts?
The principle is to move budget until marginal return is roughly equal across all channels. If one channel's next dollar outperforms another's, reallocate toward the stronger one until they balance. This is the textbook condition for optimal allocation and rarely matches what average-ROAS dashboards recommend.
-
Can marginal return be negative?
Effectively yes. In heavily saturated channels, additional spend may produce almost no incremental revenue, meaning the next dollar destroys efficiency even if the overall channel still looks profitable on average. Negative marginal returns are common in branded paid search and retargeting where the channel is mostly cannibalizing organic conversion.
-
Is marginal return relevant for B2B?
It is, especially for paid media and ABM advertising where audiences are finite. Pushing more budget into a small target account list quickly hits diminishing returns because frequency rises but unique reach plateaus. Marginal analysis catches that before money is wasted on excess impressions to saturated audiences.
-
How do you build a saturation curve?
The cleanest method is marketing mix modeling, which fits a curve to historical spend and outcome data per channel. A simpler approach is to run a stepped spend test: raise or lower spend in defined increments and observe the outcome at each level. Three to four spend levels over six months give a credible curve for most channels.
-
How does marginal return relate to incrementality testing?
They answer different questions but complement each other. Incrementality testing measures whether a channel produces any genuine lift at all; marginal return measures how much lift the next dollar produces given current saturation. A channel can have positive incrementality but a tiny marginal return, meaning it works but is fully scaled. Both signals matter for budget decisions.