Watching Exit Intent and Timed Popups for Harm
Exit intent and timed popups interrupt reading, trap focus, and reappear after dismissal. A watching brief for keeping promotional popups within bounds.
Popups are set once and forgotten, then they multiply. The welcome offer, the exit intent save, the seasonal campaign, each added by a different tool, until a shopper meets three interruptions before reading a single product description.
Watching popups means counting them and testing how they behave. This guide covers the checks: frequency, dismissal, focus, timing, and the quiet metric that matters most, whether they still earn their interruption.
Count every popup a new visitor meets
List every interruptive element by tool and trigger: time based, scroll based, exit intent, cart based. Stores are routinely surprised by their own inventory, because each tool reports only itself and no one owns the combined experience.
Then experience the sequence as a new visitor on desktop and mobile. A stack that looks reasonable as a list feels very different at second ten of a first visit, and the mobile sequence is usually worse, since exit intent there fires on scroll up and back actions during normal reading.
Test dismissal and focus behaviour
Every popup needs a clear close control with a real label, dismissible by keyboard and by Escape, that returns focus to where the shopper was. A close control that is a faint graphic in a corner, reachable only by pointer, fails a large share of visitors outright.
Check the trap question: while the popup is open, can keyboard focus wander behind it into the page, and when it closes, is focus lost to the top of the document. Both failures are common in cheap popup tools and both read as the site breaking.
Respect dismissal across the session
A dismissed popup should stay dismissed for a sensible period, and the same visitor should not meet the same offer again on the next page. Reappearing popups are the top complaint in feedback about store popups, ahead of the offers themselves.
Watch the settings after campaign changes. Launching a seasonal variant often resets frequency caps set on the original, so the checking routine after any popup edit includes one full browse as a returning visitor.
Measure whether the interruption pays
Each popup should justify itself with its own numbers: shown, dismissed, converted, and what happened to browsing depth after it fired. A popup with high dismissal and no measurable lift is not free. It costs attention on every visit it interrupts.
Retire losers on a schedule. A quarterly popup review that deletes one underperformer does more for the store experience than another round of offer testing on top of an ever growing stack.
Related reading
If this topic is useful, these earlier guides go deeper on connected questions: Watching Embedded Forms in Newsletter and Signup Blocks, Watching Consent Dialogues That Cover the Page, and Watching Sale and Clearance Templates as They Rotate.
Frequently asked questions
How many popups should a new visitor meet?
As few as possible, ideally one considered interruption per session. The practical step is an inventory: most stores discover they run more popups than anyone intended, added one tool at a time.
What makes a popup accessible to dismiss?
A clearly labelled close control reachable by keyboard, Escape key dismissal, focus that returns to the shoppers place on close, and no focus wandering behind the popup while it is open.
Why do popups reappear after dismissal?
Usually because frequency caps live in each tool separately, or a campaign edit reset them. Testing as a returning visitor after any popup change catches it before customers report it.
How should a store decide to keep a popup?
On its own evidence: conversion attributable to it versus dismissal rate and any drop in browsing after it fires. A popup that converts nothing and interrupts everyone should be retired.