Exit intent
In one sentence
Exit intent is a signal that a visitor is about to leave, most commonly detected by cursor movement toward the browser controls, and it triggers a last-moment intervention, though it is a probabilistic signal with substantial false positives that barely works on mobile.
Not to be confused with Bounce rate.
Definition
Exit intent is a signal that a visitor is about to leave, most often detected when the cursor moves toward the browser controls at the top of the window. It fires a last-moment intervention such as an overlay or an offer.
The deeper version is that it is the most established behavioral trigger in web conversion, which makes it the best-measured comparison for any intent-based activation.
How it is detected
- Desktop: cursor velocity and direction toward the top of the viewport, where the tab bar and address bar sit.
- Mobile: unreliable, because there is no cursor. Proxies include scroll-up velocity, back-button interaction and inactivity, all weaker signals.
- The mobile weakness is significant and frequently glossed over, given how much traffic is mobile.
The false positive problem
- An upward cursor movement has many causes: reaching for a bookmark, switching to another open tab, moving to the search bar to look something up in support of the purchase.
- A visitor moving to check a competitor is not leaving permanently.
- Firing on every upward movement produces interruptions that feel arbitrary.
Why combining signals works better
- Exit intent alone triggers on visitors who never engaged.
- Exit intent combined with scroll depth or dwell time targets visitors who demonstrated interest and are now leaving without converting.
- Fewer, better-judged triggers outperform more triggers, because the cost of a badly-timed interruption is high.
What it is typically used for
- Discount and offer overlays.
- Newsletter capture.
- Cart-abandonment recovery, which is where measured performance is strongest.
- Assistance offers, which is the relevant category for voice.
The measured performance
- Exit-intent overlays convert at roughly 2.87 to 3.94 percent generally.
- Cart-abandonment specific overlays perform substantially better, at around 17.12 percent.
- The gap is instructive: intervention at a moment of specific hesitation outperforms generic intervention at departure.
Common misconception
That exit intent is a reliable prediction of departure. It is a probabilistic signal with substantial false positives, an upward cursor moves for many reasons, and it barely works on mobile, where there is no cursor to read at all.
Why it matters commercially
It is the best-measured behavioral trigger available, which makes its published performance the realistic frame for expectations about any intent-based activation, including on-site voice. If you want a defensible number for what a well-timed on-page intervention does, this is where it comes from.
In voice specifically
When the intervention is a spoken assistance offer rather than a dismissible text overlay, the cost of a false positive rises, because a voice interrupting is more intrusive than a box a visitor can close. That raises the bar for combining exit intent with an engagement signal rather than firing on every upward cursor flick.
Where AsqVox fits
Exit intent is one of the live activation triggers the Orb uses, alongside scroll depth and dwell time. Combining signals rather than firing on exit intent alone is the approach that respects the false positive problem, so an assistance offer reaches a visitor who already showed interest rather than one who was only reaching for a bookmark.
Visual
The signal, and what it actually predicts
One signal, four intentions. Only one of them is a visitor actually leaving.
Statistics
Every figure carries its source and year. Vendor numbers are labelled as vendor numbers, and where no reliable figure exists this page says so rather than borrowing one.
Exit-intent overlays convert at roughly 2.87 to 3.94 percent.
2.87 to 3.94%vendor claimWisepops, Popupsmart and OptiMonk, 2026 - All three are vendor-published aggregates from their own customer bases, published across 2025 and 2026. Read them as directional evidence that well-timed on-page intervention works, not as a precise benchmark. This is the generic case, before any hesitation-specific targeting.
Cart-abandonment specific overlays are reported at around 17.12 percent.
17.12%vendor claimWisepops, Popupsmart and OptiMonk, 2026 - Same vendor sources and the same caveat. The gap against the generic figure is the instructive part: intervention at a moment of specific hesitation substantially outperforms generic intervention at the point of departure.
Desktop detection relies on cursor movement toward the browser controls. Mobile has no cursor equivalent, and the available proxies are materially weaker. This is a structural limitation rather than an implementation quality issue.
-no reliable figureIt matters because mobile is most of the traffic. The weakness cannot be engineered away, so a vendor promising comparable mobile exit-intent performance is promising something the signal cannot deliver.
There is no published false positive rate for exit intent detection, which is the number needed to judge how often it interrupts visitors who were not actually leaving.
-no reliable figureThis is the missing figure that would let anyone size the interruption cost honestly. Its absence is why combining signals is a judgment call rather than a calculation.
There is no published benchmark for combined-signal triggering measured against exit intent alone, so the improvement from adding scroll depth or dwell time is argued rather than quantified.
-no reliable figureThe case for combining signals is sound and widely held, but it rests on reasoning about false positives, not on a measured lift figure anyone has published.
Examples
In practice
A team ships exit-intent triggering alone and sees engagement rise while complaints rise faster, because the trigger fires on visitors moving toward a bookmarked tab. Adding conditions that require demonstrated engagement, through scroll depth and dwell time, reduces trigger volume substantially and improves both engagement rate and sentiment.
The everyday version
Exit intent is spotting that someone is about to leave, usually by watching the mouse head toward the top of the screen. It is a guess. The mouse goes up for lots of reasons, and on a phone there is no mouse at all, which is why combining it with other signals works better than relying on it alone.
Usage
Who says it
Who uses the term
- Conversion and growth teams, routinely.
- Vendors, as the category term for overlay and popup products.
Where it turns up
In a conversion program
- In conversion tooling, trigger configuration and testing programs.
Common misuse
What it gets used for that it should not
- Relying on it alone, without corroborating engagement signals.
- Assuming comparable mobile performance, where the signal is structurally weaker.
- Optimizing trigger volume rather than trigger precision.
Questions people ask
What is exit intent?
Exit intent is a signal that a visitor is about to leave, most commonly detected on desktop by cursor velocity and direction toward the browser controls at the top of the window. It fires a last-moment intervention such as a discount overlay, a newsletter capture, a cart-abandonment offer or an assistance offer. It is the most established behavioral trigger in web conversion.
Does exit intent work on mobile?
Poorly. Mobile has no cursor, so there is no reliable equivalent of the upward cursor movement desktop detection depends on. The available proxies, scroll-up velocity, back-button interaction and inactivity, are all materially weaker signals. This is a structural limitation rather than an implementation problem, and it matters because mobile is most of the traffic.
How well does exit intent convert?
Vendor aggregates put generic exit-intent overlays at roughly 2.87 to 3.94 percent, and cart-abandonment specific overlays at around 17.12 percent. Those come from Wisepops, Popupsmart and OptiMonk across 2025 and 2026, so read them as directional rather than precise. The lesson in the gap is that intervention at a moment of specific hesitation beats generic intervention at departure.
Why combine exit intent with other signals?
Because exit intent alone triggers on visitors who never engaged, and an upward cursor moves for many innocent reasons, from reaching for a bookmark to checking a spec in the search bar. Combining it with scroll depth or dwell time targets visitors who demonstrated interest and are now leaving without converting, so fewer, better-judged triggers replace many arbitrary ones.
Last reviewed 4 August 2026. Written and reviewed by Dhruv Dholakia, founder of AsqVox.