← All tools
UA utility · Budget allocation

Marginal ROAS calculator.

Blended ROAS is reporting. Marginal ROAS is allocation. See what each budget increase actually returned — and the step where scaling stopped paying.

Enter each spend level you ran and the revenue its cohorts produced by the same age (for example D30). Add rows for more steps. The example numbers are illustrative.

Your numbers

Spend steps (per day)
Last step marginal ROAS——
Blended ROAS at top—All spend, all cohorts
Highest step above target—Spend level to hold at

—

Blended vs marginal ROAS by spend level
BlendedMarginalTarget
SpendRevenueBlendedΔ SpendΔ RevenueMarginal

Making the comparison fair

Compare cohorts at the same age, from periods long enough to have settled, and as close in time as possible so seasonality and creative changes do not masquerade as scale effects. If a step changed more than budget — a new creative, a new market — treat its marginal ROAS as a blend of both changes.

How it's calculated

The method, in plain terms.

  1. Blended ROAS = Σ revenue ÷ Σ spend
  2. Marginal ROAS (step k) = (Revenue_k − Revenue_k−1) ÷ (Spend_k − Spend_k−1)
  3. Compare cohorts at the same age (e.g. D30)
Worked example

Moving daily spend from $10K to $15K lifts D30 cohort revenue from $4.5K to $6.0K. Blended D30 ROAS still reads a healthy 40% (down from 45%), but the extra $5K bought only $1.5K — a marginal ROAS of 30%.

Defaults on this page are illustrative starting points, not benchmarks for your game. Replace them with your own numbers.

The thinking behind it
Terms on this page

Marginal ROAS · Blended ROAS · Scale ceiling

Embed this calculator

Free to embed on your blog, wiki or course page with the credit link intact.

FAQ

Common questions.

What is marginal ROAS?

Marginal ROAS is the return on the incremental block of spend: the additional cohort revenue divided by the additional spend when budget moves from one level to the next, measured at the same cohort age. It answers “what did the last increase buy?”, which blended ROAS cannot.

How is marginal ROAS different from blended ROAS?

Blended ROAS averages all spend, so cheaper early spend and strong mature cohorts can keep it healthy while the newest spend loses money. Marginal ROAS isolates the step change, which is why it is the right metric for deciding whether to increase budget again.

What data do I need?

For each spend level you ran long enough to produce comparable cohorts, the daily (or weekly) spend and the revenue those cohorts generated by a fixed age such as D7 or D30. Use the same age for every level, and make sure the cohorts are mature enough for that window.

Can I get marginal ROAS by slicing blended ROAS by campaign?

Not reliably. Segmenting the average more finely still mixes spend that would have happened anyway with the incremental spend. You need to compare what changed when spend changed — or run a controlled test.

What target should I set?

Use the ROAS at your chosen cohort age that historically leads to payback within your acceptable window. If D30 ROAS of 35% typically becomes 100% by day 180 and a 180-day payback is acceptable, 35% at D30 is your marginal hurdle.

Calculator by The Game Scientist ↗