How does exit intent work on a desktop?
It watches the mouse. Browsers fire an event when "the cursor of a pointing device (usually a mouse)" moves out of an element, as MDN's reference puts it. Popup tools use that: when the pointer leaves through the top edge of the window, towards the tabs, the address bar or the close button, the popup opens. Popup Maker describes its version as detecting "when you move your pointer outside of the top edge of the browser window", and marks it as desktop only.
That signal is imperfect too, but it has a logic: someone reaching for the tab bar is often about to go somewhere else.
Why doesn't that work on a phone?
Because there is no pointer to leave the window. A finger touches the screen and lifts; nothing moves towards the address bar first. Wisepops, which sells mobile exit intent, says so itself: desktop exit intent relies on mouse tracking, so on phones its tool uses other signals instead.
So every mobile exit-intent tool picks one or more stand-ins. These are the ones the vendors we checked describe, and what each one really tells you:
| Signal | How the tool uses it | What it also catches |
|---|---|---|
| Quick scroll up | Phones hide the address bar while you scroll down, so a fast scroll up is read as reaching for it. Popup Maker fires when someone quickly scrolls up more than 10% of the page after scrolling down; the percentage can be changed. | Someone going back to reread a price, a photo or the opening hours. |
| Back button | The first press of Back opens the popup instead of leaving; a second press leaves. | Someone who wanted the previous page on your own site. |
| Switching tab or app | The page loses focus, and the popup shows when the visitor returns. | Someone who went to WhatsApp to ask their partner, or to check their calendar. Wisepops notes the popup appears when they come back, not when they leave. |
| Timer | The popup opens after a set number of seconds. Popup Maker calls it "a simple time delay fallback specifically for mobile devices". | Everyone still on the page, leaving or not. Wisepops: "Not interacting with a page doesn't mean I'm about to exit." |
Wisepops also says the scroll-up signal needs tuning: "If it's too sensitive, it will trigger as soon as the user scrolls up a little." Too insensitive, and it rarely fires at all. Either way, the popup is reacting to a gesture, not to a decision to leave.
Do exit-intent popups convert worse?
On the numbers the popup vendors publish, yes: exit intent is the weakest of the common triggers. It isn't mobile popups as such. Both datasets below found that popups shown on phones convert better than desktop ones.
- Wisepops, about 1 billion views of sign-up popups: exit-intent popups left an email on 3.94% of views. Popups shown on landing: 4.72%. After scrolling: 5.37%. Popups the visitor opened with a click: 54.42%. Across devices, mobile popups converted at 4.98% against 3.67% on desktop.
- Omnisend, 1.24 billion popup views in 2025: "'engagement-based' triggers generally outperformed exit intent", and popups designed for mobile converted at 2.2% against 1.4% for desktop-only ones.
These are vendors measuring their own customers' sign-up popups, and Omnisend's text and charts disagree on some of its trigger figures, so we only quote its overall finding. The direction is the same in both: on a phone, people respond to popups that come at a sensible moment, and an exit popup is the one that comes when they've already decided to go. We covered the other popup figures, including the EU rules on fake countdown timers, in Black Friday popups vs one question.
What does Google say about popups on mobile?
Two things matter here: the long-standing interstitial rule, and a newer spam policy that is closer to how mobile exit intent works.
The interstitial rule. Since 10 January 2017, Google has said that pages where "content is not easily accessible to a user on the transition from the mobile search results may not rank as highly". Its first example is a popup that covers the main content, "either immediately after the user navigates to a page from the search results, or while they are looking through the page." Small banners "that use a reasonable amount of screen space and are easily dismissible" are fine, as are cookie and age checks. Google's current guidance says intrusive dialogs "may lead to poor search performance" and recommends banners "that take up only a small fraction of the screen".
That rule is about getting to the content after a search. An exit popup comes later, so it's less clear how much it counts. But a full-screen box that appears after a scroll up is a popup that covers the content "while they are looking through the page", which is exactly the wording Google uses.
Back button hijacking. On 13 April 2026, Google added back button hijacking to its spam policies, with enforcement from 15 June 2026. The policy defines it as a site that "interferes with user browser navigation by manipulating the browser history or other functionalities, preventing them from using their back button to immediately get back to the page they came from." Pages that do it "may be subject to manual spam actions or automated demotions", and Google warns that it can come from "the site's included libraries", which includes a popup script.
A back-button exit popup works by adding an entry to the browser history and listening for the Back press (the popstate event, in MDN's terms), so the first press shows the popup instead of going back. Google's announcement names the obvious cases, such as sending people to pages they never visited or showing them ads. Whether a newsletter popup on the first Back press counts is Google's call, not ours. It does stop someone using their back button to "immediately" get back, which is the word Google chose. We wouldn't put a site's search traffic on that bet.
Browsers push back as well: MDN notes that, to combat unwanted popups, browsers "may not fire the popstate event at all unless the page has been interacted with." So the back-button trigger misses many of the visitors who bounce quickest.
What to use instead
The useful idea behind exit intent is "speak up before you lose them". On a phone, you can't spot the last moment, so speak up earlier, at a moment the visit itself gives you, and in a form that doesn't block the page.
- Pick a moment of interest, not of leaving. The second page about the same service, a pricing page after a short read, a product someone has opened twice, a return visit. In Omnisend's data, popups triggered by pages viewed or by scrolling did better than exit intent.
- Keep it small. A card in the corner or a line inside the page, which Google's guidance supports. On a phone, nothing that covers the content.
- Ask, don't demand. One question with two to four answers to tap, including one for people who are just looking. No email field first. See how to qualify leads with one question for what to ask.
- Make the answer change the next step. "Today" gets the phone number and WhatsApp; "planning ahead" gets a call back at a time they pick or a guide by email; "just looking" gets left alone.
- Leave the back button alone. Let people go when they want to.
- Let the visitor open it. A visible button such as "Get a quote" or "Ask a question" is the click-triggered popup from the Wisepops figures: the one people chose to open.
A worked example
This is an illustration, not a customer story. A garden design firm gets most of its visits on phones, from people searching for garden ideas and local designers. It runs a full-screen exit popup: "Wait! Get our free planting guide." It opens on a quick scroll up, which on a phone often means someone going back to look at a photo of a project.
The firm swaps it for one question in a small card, shown when someone opens a second project page: "Are you planning a garden?"
- "Yes, this year": offer a call back at a time they pick, or the phone number.
- "Maybe next year": offer the planting guide by email, with a clear opt-in.
- "Just looking for ideas": nothing more. They keep reading.
The guide still gets offered, but only to people who said it's relevant, and nobody is stopped while scrolling back to a photo.
Where Signal fits
This is how Signal's popup works. It's small, sits in the corner, and only appears when Signal's estimate of what the visitor wants is unsure and the answer would change what to offer. You decide which pages matter and how long someone has to stay before it speaks up, and it never asks the same thing twice. Then it offers the next step you set for that answer: a call, a call back, WhatsApp, a booking link, or a guide by email. You can run it in shadow mode first, to see what it would have asked and offered on your real traffic before visitors see anything. Plans start at €39 a month; see how it works.
When is an exit popup still worth it?
- On desktop traffic, where the mouse signal exists. Keep it small, show it once, and never at checkout.
- For one offer that suits everyone, such as free delivery, when there's nothing worth asking first.
- When you've tested it against something calmer and it wins on your own numbers, counting sales or enquiries, not only emails.
On mobile, if you keep one, use the scroll-up or timer trigger rather than the back button, keep it to a banner, and show it once per visitor. For more on why visits don't turn into enquiries in the first place, see website traffic but no leads.
Questions people ask
Can you detect exit intent on mobile? Not directly. Phones have no cursor leaving the window, so tools guess from scrolling up, the back button, switching apps or a timer.
Do exit-intent popups hurt SEO on mobile? A full-screen popup that covers the content while someone reads goes against Google's interstitial guidance. One that catches the back button comes close to Google's back button hijacking policy, enforced since June 2026. A small, easy-to-close banner is fine.
What's a good alternative to an exit-intent popup? A small question with answers to tap, shown when the visit shows interest, with a different next step for each answer. Or a button people can open themselves.