Doubling Down Doesn't Double

Split a fixed quarterly budget across five marketing channels — each with a hidden saturation curve. Allocate blind in round one, then watch every channel's curve draw itself and reallocate in round two. The moat: each channel's knee — the exact spend where "put more here" quietly stops working — becomes a visible bend in the line, not a surprise in next quarter's CAC report. Runs entirely in your browser (0 uploads, works offline).

How to use it

Round one: move the sliders to split the budget across the five channels — with the curves still hidden.
Hit Reveal the curves — each channel's returns curve draws in, with its knee marked and your spend as a dot (red once you're past the knee).
Round two: reallocate now that you can see where each curve bends, and re-score. Chase a higher efficiency % — how close you got to the mathematically optimal split.

What this clears up (the fundamentals)

Returns are concave, not linear — a channel's output is plateau × spend ÷ (spend + knee). The first dollars buy a lot; past the knee, each extra dollar buys almost nothing.
The knee is where doubling down dies — at the half-saturation point, marginal return has already dropped to a quarter of where it started. "It worked last quarter" is exactly why it stops working when you pour more in.
Spreading beats concentrating — the optimal split equalises the marginal return across channels (water-filling). The next dollar always belongs wherever a curve is still steep, not wherever the last dollar landed.
Concentration has a measurable cost — dumping a whole budget into one winning channel typically scores far below 100% efficiency, and the tool shows you exactly how much you left on the table.

Where it's used

A free top-of-funnel GTM channel-mix exercise, and the L2 demo for the miatz module "Automating the Campaign, Not Just the Copy." A fast way to feel why marketing budgets should be rebalanced, not just scaled. It's a miatz build-lab concept playable; learn to build it yourself.

FAQ

What's a saturation curve?

A curve that flattens as spend rises: doubling spend less than doubles returns. Here each channel uses plateau × spend ÷ (spend + knee), so returns approach a ceiling (the plateau) and the marginal payoff shrinks with every extra dollar.

What is the "knee"?

The spend at which returns reach half the plateau — the point where the curve visibly bends from steep to flat. Past it, the marginal return has already fallen to a quarter of its starting value, so extra spend is mostly wasted.

How is the optimal split computed?

By marginal-return equalisation ("water-filling"): the split where the next dollar would earn the same in every channel. Any other split is beatable by moving a dollar from a flat channel to a steep one — which is exactly what the efficiency score measures.

Is anything uploaded?

No. Every allocation, curve and score is computed in your browser — nothing is transmitted or stored. Turn off your Wi-Fi and it still works.

Limits

A teaching model: the curves are smooth and independent, with no cross-channel effects (brand halo, retargeting overlap), no time lag, and no diminishing audience beyond the single saturation term. Use it to build the instinct for concave returns, then bring channel-level data to your real media plan.

Part of the miatz playgrounds. Explore the rest from the playground catalog.

Bookmark this page (Ctrl+D, or ⌘D on Mac) — it works offline the next time you need it.

Doubling down doesn't double

How to use it

Round one: move the sliders to split the budget across the five channels — with the curves still hidden.

Hit Reveal the curves — each channel's returns curve draws in, with its knee marked and your spend as a dot (red once you're past the knee).

Round two: reallocate now that you can see where each curve bends, and re-score. Chase a higher efficiency % — how close you got to the mathematically optimal split.

What this clears up (the fundamentals)

Returns are concave, not linear — a channel's output is plateau × spend ÷ (spend + knee). The first dollars buy a lot; past the knee, each extra dollar buys almost nothing.

The knee is where doubling down dies — at the half-saturation point, marginal return has already dropped to a quarter of where it started. "It worked last quarter" is exactly why it stops working when you pour more in.

Spreading beats concentrating — the optimal split equalises the marginal return across channels (water-filling). The next dollar always belongs wherever a curve is still steep, not wherever the last dollar landed.

Concentration has a measurable cost — dumping a whole budget into one winning channel typically scores far below 100% efficiency, and the tool shows you exactly how much you left on the table.

FAQ