Behavior-linked experiments

행동 데이터 연동 실험

Read funnel, replay, and survey signals beside experiment results.

···
html
<div class="linked"><div class="result">EXPERIMENT B <b>+ LIFT</b></div><div class="sources"><span>FUNNEL: PAYMENT DROP</span><span>REPLAY: REPEAT TAP</span><span>SURVEY: FEE UNCLEAR</span></div><div class="join">↓ NEXT HYPOTHESIS</div></div>
css
.linked{width:min(83%,470px);font:800 clamp(10px,max(1.7vmin,1vw),14px)/1.2 sans-serif}.result{border:2px solid var(--fg);padding:clamp(8px,2vmin,17px);display:flex;justify-content:space-between}.result b{color:var(--accent)}.sources{display:grid;gap:5px;margin-top:7px}.sources span{padding:clamp(5px,1.2vmin,10px);background:var(--surface);border-left:4px solid var(--accent);animation:attach 4.2s infinite}.sources span:nth-child(2){animation-delay:.45s}.sources span:nth-child(3){animation-delay:.9s}.join{text-align:center;color:var(--accent);margin-top:7px;animation:join 4.2s infinite}@keyframes attach{0%,12%{opacity:.15;transform:translateX(15px)}26%,80%{opacity:1;transform:none}100%{opacity:.15}}@keyframes join{0%,12%{opacity:.55}30%,90%{opacity:1}100%{opacity:.55}}

Behavior-linked experiments connect conversion results to what visitors actually experienced. Tools now bring analytics, replays, and surveys around shared event definitions.

In the demo, drop-off, repeated clicks, and survey feedback attach to a result card one by one. These clues can form the next hypothesis.

Qualitative signals do not establish causation. Validate assignment and outcome calculations separately, and apply privacy controls to replays.

When to use

Use it when a test has ended but its context and next hypothesis are unclear.

Open as page ↗