What is Marketing Mix Diminishing Returns?
This models diminishing returns — the well-established fact that pushing more budget into a channel generates progressively lower ROAS on each additional dollar, as the best-performing inventory or audience gets used up first. It answers a more advanced question than simple ROAS: not 'is this channel working', but 'is the next dollar into it still worth it'.
Why it matters
- It's the modeling approach that fixes the main weakness of proportional budget allocation, which wrongly assumes ROAS stays flat as spend increases.
- It shows when a channel is approaching saturation, before overall account ROAS actually starts declining.
- It gives a specific number — marginal ROAS — to compare against other channels' marginal returns when deciding where the next budget dollar should go.
The formula
Marginal ROAS = [k × (Current Spend + Additional Spend)^α − Current Revenue] ÷ Additional Spend, where k = Current Revenue ÷ Current Spend^α
- Current Spend, Current Revenue
- — the channel's current monthly spend and the revenue it generates
- α (alpha)
- — the diminishing-returns factor — lower means returns fall off faster as spend increases
- Additional Spend
- — the extra budget being considered for this channel
How to use this calculator
- 01Enter your channel's current monthly spend and the revenue it currently generates.
- 02Estimate the diminishing-returns factor (alpha) — 0.6-0.8 is a reasonable starting range for most paid channels; refine it using your own historical spend/revenue data over time if available.
- 03Enter the additional spend you're considering.
- 04The result is the marginal ROAS on that next increment — compare it against your break-even ROAS and other channels' marginal returns.
Worked example
A channel currently spends $10,000/month generating $40,000 in revenue (4x blended ROAS), with an alpha of 0.7, considering an additional $5,000.
- k = $40,000 ÷ 10,000^0.7 ≈ 63.4
- New revenue at $15,000 spend = 63.4 × 15,000^0.7 ≈ $53,131
- Marginal revenue = $53,131 − $40,000 = $13,131
- Marginal ROAS = $13,131 ÷ $5,000 ≈ 2.63×
While the channel's blended ROAS is a strong 4x, the marginal ROAS on the next $5,000 is only about 2.6x — still likely worth spending if that clears your break-even ROAS, but a clear signal the channel is starting to saturate.
Industry benchmarks
Compiled August 2026
Alpha 0.8 – 0.95
Slow saturation
Channel can absorb significant additional spend before returns drop much.
Alpha 0.5 – 0.7
Moderate saturation
Common for most mature paid channels.
Alpha below 0.5
Fast saturation
Small niche audiences or highly competitive auctions.
Alpha of exactly 1.0
No diminishing returns
The unrealistic assumption behind flat proportional budget splits.
Common mistakes
- Assuming blended ROAS applies to the next dollar spent — it almost never does once a channel has meaningful volume.
- Guessing alpha without ever calibrating it against your own actual spend/revenue history at different budget levels.
- Applying one alpha across very different channels, when saturation speed varies significantly by platform and audience size.
- Reading the marginal ROAS as a prediction rather than a directional estimate — the curve is a simplification of a market that also moves for reasons unrelated to your spend.
How to improve your diminishing returns
Calibrate alpha using your own historical data
If you have spend and revenue data at multiple budget levels over time, fit alpha to that data instead of using a generic estimate — this significantly improves accuracy.
Compare marginal ROAS across channels before allocating
The channel with the highest marginal ROAS on the next dollar — not the highest blended ROAS — is where new budget should go.
Compare current channel allocation →Stop spending where marginal ROAS crosses break-even
The correct ceiling for a channel is the point where the next dollar returns exactly your break-even ROAS. Past that, additional budget buys revenue at a loss even while blended ROAS still looks healthy.
Calculate your break-even ROAS →Frequently asked questions
Why not just use blended ROAS to decide where to spend more?+
Blended ROAS reflects the average across all your current spend, including the best-performing early dollars — the next dollar is typically less efficient than that average, which is exactly what this model corrects for.
How do I get a real alpha value instead of guessing?+
Fit the power-law curve to at least 3-4 historical data points of spend and revenue at different budget levels — a spreadsheet regression or your analytics platform's own modeling tools can do this more precisely than a manual guess.
Is this the same as marketing mix modeling?+
It borrows the saturation curve that sits at the heart of marketing mix modeling, but a real MMM also models seasonality, adstock carryover, price, competitor activity and baseline demand across every channel at once. Treat this as one component of that idea, used on its own.
Can marginal ROAS go negative?+
Under this power-law model, no — additional spend always adds some revenue, just progressively less of it. In reality it can, when extra spend drives frequency high enough to irritate the audience or bids into unprofitable inventory. That is a limit of the model, not evidence it cannot happen.
Related calculators
Put this calculator on your site
Free to embed, no attribution required beyond the link the widget already carries. It posts its height to the parent window as marketing-calculators:height if you want to size the frame exactly.
<iframe src="https://markcelerate.com/embed/marketing-mix-diminishing-returns-calculator" title="Marketing Mix Diminishing Returns Calculator" width="100%" height="520" style="border:1px solid #ddd;border-radius:6px" loading="lazy"></iframe>