Measurement · 01 Demo · synthetic data
MMM budget reallocation
Where should the next dollar go?
Channel response curves with saturation. Move spend between channels and watch predicted revenue, blended MER, and each channel's marginal return update live.
MMM budget reallocation
Weekly plan across four channels. Drag spend; the model responds.
Predicted weekly revenue
$588k
0.0% vs. starting plan
Total spend
$155k
Starting plan $155k
Blended MER
3.80×
Revenue over total spend
Next dollar to
Paid social
From Retail media
Paid search
Marginal 1.19×
- Contribution
- $96k
- Average ROAS
- 2.14×
Paid social
Marginal 1.71×
- Contribution
- $114k
- Average ROAS
- 1.90×
Online video
Marginal 1.44×
- Contribution
- $16k
- Average ROAS
- 0.82×
Retail media
Marginal 1.18×
- Contribution
- $61k
- Average ROAS
- 2.05×
Base sales without paid media are fixed at $300k a week. Dashed line: the slope at the current spend, which is the marginal return. Curves are Hill saturation functions with made-up parameters, standing in for a fitted model's posterior means.
Synthetic numbers generated in your browser. Not client or employer data.
How to read it
What the demo is showing
Each curve is a channel's modeled response to weekly spend. The dot is the current plan. Where a curve flattens, the next dollar buys less.
Average return (revenue over spend) flatters a channel that is already saturated. Marginal return (what the next dollar adds) is the number that should move budget.
"Rebalance at the same budget" shifts spend toward the highest marginal return until the channels are roughly equal at the margin. A real plan adds guardrails: minimum spend, ramp limits, and the model's uncertainty.
The shift from MTA to macro-measurementHow I'd approach an MMMNext demo: Geo test readoutAll demos