Marketing mix model (MMM)

마케팅 믹스 모델 (MMM)

Estimate channel effects from aggregate spend and outcome patterns.

···
html
<div class="mmm"><div class="head">SPEND → ESTIMATED RESPONSE <b id="out">+62</b></div><svg viewBox="0 0 360 145" preserveAspectRatio="none"><path d="M20 125 Q68 71 125 47 Q205 17 345 11" fill="none" stroke="var(--accent)" stroke-width="5"/><path d="M20 125H345" stroke="var(--line)" stroke-width="2"/><line id="guide" x1="125" x2="125" y1="47" y2="125" stroke="var(--muted)" stroke-dasharray="5 5"/><circle id="point" cx="125" cy="47" r="9" fill="var(--fg)"/></svg><div class="ctrl"><span>MEDIA SPEND</span><input id="range" type="range" min="0" max="100" value="33"><b id="spend">33</b></div></div>
css
.mmm{width:min(87%,590px);font:850 clamp(10px,max(2vmin,1vw),14px)/1.2 sans-serif}.head,.ctrl{display:flex;align-items:center;justify-content:space-between;gap:10px}.head{border-bottom:2px solid var(--fg);padding-bottom:8px}.head b{color:var(--accent);font-size:clamp(17px,4vmin,30px)}svg{width:100%;height:clamp(88px,23vmin,170px);display:block;margin-top:clamp(7px,2vmin,14px)}.ctrl{border-top:1px solid var(--line);padding-top:8px}.ctrl span{white-space:nowrap;color:var(--muted);font-size:clamp(10px,max(1.7vmin,1vw),12px)}input{flex:1;min-width:0;accent-color:var(--accent)}
js
const r=document.querySelector('#range');function draw(){const v=Number(r.value),x=20+3.25*v,y=125-114*(1-Math.exp(-v/27));document.querySelector('#point').setAttribute('cx',String(x));document.querySelector('#point').setAttribute('cy',String(y));document.querySelector('#guide').setAttribute('x1',String(x));document.querySelector('#guide').setAttribute('x2',String(x));document.querySelector('#guide').setAttribute('y1',String(y));document.querySelector('#out').textContent='+'+Math.round(100*(1-Math.exp(-v/34)));document.querySelector('#spend').textContent=String(v)}r.oninput=()=>{clearInterval(timer);draw()};let dir=1;const timer=setInterval(()=>{let v=Number(r.value)+dir*8;if(v>=96||v<=8)dir*=-1;r.value=String(Math.max(0,Math.min(100,v)));draw()},460);draw();

MMM models the relationship between aggregate sales and channel spend over time or regions. It can account for lag, saturation, seasonality, price, and other factors.

Increasing the budget in the demo first raises estimated contribution quickly, then with diminishing returns. The curve is illustrative rather than a fitted model.

Estimates depend on input data and assumptions. Calibrate with experiments and inspect uncertainty and diagnostics before reallocating budgets.

When to use

Use it to plan budgets when individual paths cannot capture all channel effects.

Open as page ↗