One-click ad experiments

원클릭 광고 실험

Splits a campaign to compare a new automation setting against a baseline.

···
html
<div class="experiment"><div class="label">CAMPAIGN EXPERIMENT <span>50 / 50 SPLIT</span></div><div class="arms"><div>BASELINE <b id="base">12</b><i id="basebar"></i></div><div>AI MAX TEST <b id="test">13</b><i id="testbar"></i></div></div><div class="chart"><svg viewBox="0 0 300 55" preserveAspectRatio="none" aria-label="Illustrative comparison chart"><polyline id="line-a" fill="none" stroke="var(--fg)" stroke-width="3" points="0,45 75,35 150,29 225,24 300,18"/><polyline id="line-b" fill="none" stroke="var(--accent)" stroke-width="3" points="0,45 75,37 150,31 225,23 300,15"/></svg></div><div class="caption" id="caption">COMPARE AFTER THE TEST WINDOW</div></div>
css
.experiment{width:min(88%,500px);font:800 clamp(10px,max(2vmin,1vw),16px)/1.2 sans-serif}.label{display:flex;justify-content:space-between;gap:7px;border-bottom:2px solid var(--fg);padding-bottom:6px}.label span{color:var(--accent)}.arms{display:grid;grid-template-columns:1fr 1fr;gap:clamp(6px,2vmin,15px);margin-top:clamp(8px,2vmin,15px)}.arms div{border:1px solid var(--line);background:var(--surface);padding:clamp(7px,2vmin,16px)}.arms b{display:block;font-size:clamp(17px,4.5vmin,32px);margin:3px 0}.arms i{display:block;height:clamp(6px,1.3vmin,10px);width:50%;background:var(--fg);transition:width .45s}.arms div:last-child i{background:var(--accent)}.chart{height:clamp(38px,10vmin,75px);border-bottom:1px solid var(--line);margin-top:clamp(8px,2vmin,14px)}svg{display:block;width:100%;height:100%}.caption{margin-top:6px;color:var(--accent)}
js
const states=[[12,13,'EARLY READ — KEEP RUNNING'],[18,17,'MID TEST — KEEP RUNNING'],[24,27,'COMPARE AFTER THE TEST WINDOW']];let n=0;function draw(){const [a,b,label]=states[n++%3];document.getElementById('base').textContent=String(a);document.getElementById('test').textContent=String(b);document.getElementById('basebar').style.width=(a*3)+'%';document.getElementById('testbar').style.width=(b*3)+'%';document.getElementById('caption').textContent=label;document.getElementById('line-a').setAttribute('points','0,45 75,35 150,29 225,24 300,'+(55-a));document.getElementById('line-b').setAttribute('points','0,45 75,37 150,31 225,23 300,'+(55-b))}draw();setInterval(draw,1500);

In September 2025, Google announced one-click experiments within the setup flow for AI Max for Search campaigns. Advertisers can compare an existing campaign with an AI Max test arm.

The demo splits a fictional baseline and test setting and draws their outcomes over the same period. The figures are illustrative and make no real performance claim.

Define the primary metric, duration, and audience split before launch. Avoid stopping on an early fluctuation or comparing different conversion definitions.

When to use

Use it to compare new automation with a baseline campaign; lock the experiment plan first.

Open as page ↗