Popups are costing you trust: what to use instead
Popups and exit-intent popups still collect leads, but at a cost most reports never show. What they return, when they still earn a place, and what replaces them.
- Popups still collect leads, at a low rate and a real cost. In the largest dated vendor dataset we found, the median popup turns 0.83% of displays into leads, while 40.5% of displays end with a close.
- Exit timing adds little on its own. The same dataset found no statistically reliable advantage for exit-intent popups. Page and offer matter far more: 4.57% on cart and checkout pages against 0.23% on blog pages.
- They still earn a place in a narrow case. A first-order discount for an online store shopper who has shown interest works. A generic signup overlay on arrival, on mobile, on content pages does not.
- The replacement is relevance, not a quieter popup. Sticky bars and inline offers remove the interruption; intent-based microexperiences also change the message to match the buying stage of the visit.
Popups, including exit-intent popups, still collect leads, but rarely enough to justify what they cost on most pages. The median popup converts 0.83% of its displays into leads, 40.5% of displays end with the visitor closing it, and exit timing shows no reliable advantage over any other trigger. Keep a popup where the visitor has already shown interest and the offer is worth something to them today; everywhere else, replace it with an offer that sits in the page and matches the intent of the visit.
Those figures come from Popupsmart’s popup benchmark (updated 25 September 2026), which covers 28,674 campaigns from 11,533 businesses and 499.6 million popup displays between January 2024 and September 2026. It is vendor data from a company that sells popups, so if anything it flatters the format.
This article answers three questions for anyone who pays for the traffic a popup interrupts: what popups return, what they cost in trust, mobile experience and search, and what replaces them. The conversion event in the main thread is a lead (a visitor submitting an email address or phone number). Where an example is an online store, the event is a purchase and is labelled as such. For exit behavior specifically, the companion piece is the most effective exit intent strategy.
What popups return today
A typical popup converts fewer than one in a hundred displays into a lead, and the spread around that figure is explained by page and offer, not by timing. In Popupsmart’s benchmark, the median popup that asks for an email or phone number converts 0.83% of its displays into leads. The top 25% of campaigns reach 1.96% and the top 10% reach 3.87%.
The benchmark set is 10,768 campaigns with at least 1,000 displays and 14 active days. Three cuts matter for a budget decision:
- Page. Popups on cart and checkout pages convert 4.57%, against 1.07% on the home page and 0.23% on blog pages. That is about 20 times more on pages where the visitor has already shown intent.
- Offer. Signup popups that offer a discount code convert 1.46%, against 0.75% for signup popups without one.
- Trigger. Popups on the most sensitive exit-intent setting converted 1.10%, against 0.78% without exit intent, a gap the report describes as showing “no statistically reliable advantage”. Delay length showed no measurable effect either.
This changes the question the live debate usually asks. “Are exit-intent popups outdated?” assumes the timing is the lever. In this data it is not. A popup performs when it appears in front of someone who already wants what it offers, and the exit moment tells you nothing about that. The older claims that exit-intent popups lift conversions by a fixed percentage do not come with a dataset or a date, so they are not repeated here.
These are medians across one vendor’s customers, and a lead here is any submission with an email address or phone number. The figures say nothing about lead quality, purchases or revenue. Use them for orders of magnitude and for the direction of each cut, not as a target for your own site.
What popups cost that the report does not show
The cost of a popup is the interruption it imposes on everyone who did not want it, and almost none of that cost appears in the popup’s own report. The same benchmark that reports a 0.83% median lead rate also reports that 40.5% of popup displays end with the visitor closing the popup.
Usability research has described the mechanism for years. Nielsen Norman Group’s guidance on modal dialogs (April 2017) lists what a modal does: it interrupts the user’s workflow, blocks the content in the background and gives the visitor “an extra goal: to dismiss the dialog”. Its conclusion: “No one likes to be interrupted, but if you must, make sure it’s worth the cost.”
Its later review of ten problematic popup trends (June 2019) is more specific: do not show a popup before the visitor has reached the main content, do not ask for an email before any interaction, show only one at a time, and reserve modal overlays “for delivering crucial information, only at appropriate times”. Both pieces are older than the benchmark, and nothing in the newer data contradicts them.
For a team paying for traffic, the practical cost is this: the visitor you paid to bring to a landing page meets a request before the page has made its case. The closing of that popup is recorded nowhere in your campaign report, and the conversion the page might have earned is simply missing. That is the trust cost in the title. It is real, and it is hard to put a number on, which is why it has to be tested and not assumed.
- Stops reading, comparing or checkout
- Demands a decision now
- Hides the content the visitor came for
- Adds a task: find the close button
- Same message whatever the visitor has done
A popup performs when it appears in front of someone who already wants what it offers, and the exit moment tells you nothing about that.
Common mistake: judging a popup by its own signup rate. A popup always collects some addresses. The question is whether the site produces more qualified leads, or more revenue per visitor for an online store, with the popup than without it. Only a holdout group that never sees the popup can answer that.
What Google says about popups and search
Google does not ban popups, but it tells site owners to avoid the intrusive kind and names the alternative. Its Search Central page on avoiding intrusive interstitials and dialogs (last updated 10 December 2025) defines them as “page elements that obstruct users’ view of the content, usually for promotional purposes” and states: “Intrusive dialogs and interstitials make it hard for Google and other search engines to understand your content, which may lead to poor search performance.”
The recommendation on the same page is direct: “Instead of full page interstitials, use banners that take up only a small fraction of the screen to grab your users’ attention.” Dialogs that are legally required, such as age gates and consent, are treated as a separate case.
How large is the ranking risk? Smaller than popup critics often claim. Google’s page experience documentation (last updated 22 September 2026) lists “Do your pages avoid using intrusive interstitials?” among its self-assessment questions, and also says that “beyond Core Web Vitals, other page experience aspects don’t directly help your website rank higher in search results”. The honest reading: a full-screen interstitial can stop search engines from understanding the page, and a small dismissible banner is not a ranking problem.
Mobile is where the risk concentrates. In the Popupsmart data, 63% of popup displays happen on mobile, where an overlay takes most of the screen and the close button is small. Mobile visitors also convert on popups at a higher median rate (1.02% against 0.57% on desktop), so the device where popups collect the most leads is also the one where they obstruct the most content.
The advertising industry reached a similar verdict on the format. The Coalition for Better Ads lists pop-up ads among the ad experiences that fall beneath its Better Ads Standards on both desktop and mobile web, based on research in which more than 150,000 consumers have taken part. Those standards govern ads, not a site’s own signup overlay, but they describe the same reaction from the same people.
Consent dialogs, age verification and other legally required notices are not marketing popups, and popups a visitor opens on purpose (clicking “Get the guide”) interrupt nobody. Neither belongs in this debate.
When popups still work, and when they do not
Popups still work when three conditions hold together: the visitor has shown interest, the offer is worth something to them today, and the popup is easy to dismiss. The clearest case is a first-order discount for a first-time shopper in an online store. The benchmark figures line up with that: a discount code roughly doubles the median signup rate (1.46% against 0.75%), and cart and checkout pages convert at 4.57%.
| Situation | Popup verdict | Why |
|---|---|---|
| First-time shopper in an online store, viewing products or cart | Can work | A first-order discount has immediate value. Judge it on revenue per visitor (purchases), not signups. |
| Stock alert or price-drop signup on a product page | Can work | The visitor asked for it, or clearly would. Better still as an inline form. |
| Newsletter signup on arrival at a blog article | Rarely works | The reader has not read anything yet. Blog pages convert 0.23% in the benchmark. |
| Paid traffic arriving on a landing page | Avoid | You paid for the visit to see the page. The overlay competes with your own call to action. |
| B2B visitor reading pricing or comparing options | Avoid | The goal is a qualified lead. A generic overlay adds addresses, not qualified pipeline. |
| Generic “wait, before you go” exit popup | Avoid | Exit timing shows no reliable advantage, and the message ignores why the visitor is leaving. |
Even the discount case needs a caveat. A code shown to every shopper is also shown to the ones who would have bought at full price, so signups and even conversion rate can rise while margin falls. For an online store the measure is revenue per visitor against a holdout group, counted in purchases.
An exit signal is not worthless. It becomes useful when it is one input among several: a visitor who viewed pricing twice and then moves to leave is a different case from one who bounced after five seconds. How to use exit behavior that way is covered in the most effective exit intent strategy.
What to use instead: five options compared
There are four practical replacements for an interruptive popup, and they differ on two things: how much they interrupt, and how much they know about the visitor. The first decides the cost. The second decides the return.
| Option | Interruption | Relevance to intent | Mobile and SEO risk | Data needed |
|---|---|---|---|---|
| Interruptive popup (on load or timed) | High. Blocks the page | None. Fires on a clock | Highest. The format Google advises against when it obstructs content | None |
| Exit-intent popup | High, but late in the visit | Low. Knows the visitor is leaving, not why | Medium. No cursor on mobile, so triggers are approximations | Cursor or scroll signal |
| Sticky bar | Low. Content stays visible | Low. Same message for all | Low if small. Matches Google’s banner advice | None |
| Inline or embedded offer | None. Part of the page | Medium. Matches the page topic, not the visitor | Lowest | None |
| Intent-based microexperience | Low. Small card, page stays usable | High. Chosen by predicted buying stage | Low. Page content is unchanged | First-party behavior in the session |
Qualitative comparison based on how each format works and on the Google and Nielsen Norman Group guidance cited above. It is not a benchmark of results.
Sticky bars and inline offers solve the interruption and nothing else. They are the right swap for a newsletter signup or a sitewide promotion, and they are what both Google and Nielsen Norman Group recommend in place of an overlay. Their limit is that the message is still the same for a first-time reader and for a buyer on a third visit.
That limit matters most on sites where few visitors are ready on the first visit. A fixed offer either asks too much of early visitors or says too little to ready ones, the situation described in what to do with visitors who are not ready to book a call and in nurturing cold traffic with tailored on-site journeys. Whatever you choose, the page itself still has to carry the argument: see how to optimize a landing page without hurting UX.
From exit timing to visitor intent
Intent-based personalization replaces the question “when should we interrupt?” with “what does this visitor need now?”. An exit signal says someone is leaving. It does not say why, or how close they were to converting. Treating all exits as the same moment of urgency is where most exit-intent popups go wrong.
Behavior during the visit says much more: pages viewed, depth of engagement, interaction patterns, time spent, return visits. Read together, those signals separate a visitor who is still learning from one who is comparing options and one who is ready to act. Each needs something different:
- Awareness. Context and a short explanation. Pressure here ends the visit.
- Consideration. Proof: how others decided, what the result was.
- Decision. Reassurance and a clear, short path to the conversion.
This is the mechanism, not a product. You can approximate it by hand with rules (show proof on pricing pages, hide the signup offer from returning customers). Rules hold up for two or three segments and then become hard to maintain, which is where prediction takes over. How AI can decide which website content to show each visitor covers that step in detail.
One caution applies to every option in this article, the personalized ones included: relevance has to be measured, not assumed. If a low rate is your starting point, first check whether the traffic or the website is the cause before changing how you ask.
How a Pathmonk microexperience differs
Pathmonk applies this mechanism without rules to write: it predicts the buying stage of each visit and shows a matching microexperience in the page. It reads first-party, cookieless behavior in the session in real time, predicts whether the visitor is in awareness, consideration or decision, and serves a microexperience for that stage. No developer work is needed and the page content is not changed.
Three things separate it from a popup:
- It is triggered by predicted buying stage, not by a timer or a cursor movement. How Pathmonk detects the stage of the customer journey explains the method.
- It does not block the content. The card sits in the page and the visitor can keep reading, comparing or checking out. See what a microexperience is.
- It is measured against a control group. The result is the difference between exposed and unexposed traffic, the test this article asks of any popup.
The product is shown on the CRO and personalization page.
Case: different content by intent, same next step
Print & Cheques Now Inc., a Canadian B2B e-commerce supplier of cheques, used Pathmonk to respond differently to visitors based on real-time intent while keeping one buying journey. The conversion event is a purchase.
Visitors showing early or uncertain behavior saw short videos and explanatory prompts. Visitors with stronger buying signals saw testimonials and trust cues. Whatever the intent, every microexperience led to the same next step: the product selection flow.
Conversion rate rose from 3.55% for unexposed traffic to 5.11% for traffic exposed to Pathmonk, a 44% increase, measured as an A/B comparison over the test period. The lift appeared at every stage: +26.3% in awareness, +30.7% in consideration and +35.1% in decision. The case study does not state the length of the test.
If your only goal is a newsletter list and you run an online store with a real first-order offer, a well-placed popup or an inline form is simpler and may be enough. One customer result is also not a forecast: the figures above belong to one site and one test.
Key takeaways
- Popups still collect leads, but the median popup converts 0.83% of displays, and 40.5% of displays end with a close (Popupsmart, September 2026, vendor data).
- Exit timing is not the lever. The same data found no statistically reliable advantage for exit-intent popups.
- Page and offer explain performance: 4.57% on cart and checkout pages against 0.23% on blog pages, and 1.46% with a discount code against 0.75% without.
- Google does not ban popups. It advises against intrusive interstitials that obstruct content and recommends small banners in their place.
- Keep a popup where the visitor has shown interest and the offer has immediate value, such as a first-order discount in an online store, and judge it on revenue per visitor against a holdout.
- Sticky bars and inline offers remove the interruption. They do not make the message more relevant.
- Intent-based microexperiences change the message to match the buying stage, without blocking the page, and should be measured against a control group like any other option.
FAQs about popups and exit intent
Are exit-intent popups still effective or outdated?
They are not completely outdated, but they are no longer effective by default. In Popupsmart’s benchmark of 28,674 campaigns, updated in September 2026, exit-intent popups showed no statistically reliable advantage over popups without exit intent. What drives results is the page and the offer, not the exit moment.
Why do many exit-intent popups feel intrusive or ineffective?
They depend on exit behavior alone. An exit signal does not explain why a visitor is leaving or how ready they were to convert, so the same message reaches a curious reader and a near buyer. Most visitors close it: 40.5% of popup displays in the Popupsmart data end with a close.
What is the difference between exit intent and user intent?
Exit intent is a timing signal: the visitor is about to leave. User intent is built from behavior across the visit, such as pages viewed, depth of engagement, interactions and return visits. The first says when, the second says what the visitor needs.
When does a popup or exit intent still make sense?
When the visitor has already shown interest, the offer has immediate value and the popup is easy to close. A first-order discount for a first-time shopper in an online store is the clearest case. Exit intent is useful when layered on top of intent signals, for example reassurance for a visitor who viewed pricing and then moves to leave.
Do popups hurt SEO?
They can. Google states that intrusive dialogs and interstitials make it hard for search engines to understand your content, which may lead to poor search performance, and recommends banners that take up only a small fraction of the screen. Small, dismissible banners and legally required notices are not the problem.
What is a good popup conversion rate?
In Popupsmart’s benchmark, the median popup that asks for an email or phone number converts 0.83% of displays into leads, the top 25% reach 1.96% and the top 10% reach 3.87%. This is vendor data and counts leads, not purchases or revenue.
What is a better alternative to traditional exit-intent popups?
For a simple signup, a sticky bar or an inline offer removes the interruption. For conversion, intent-aware microexperiences adapt earlier in the visit and respond to demonstrated behavior, so the visitor gets context, proof or a prompt to act depending on their buying stage, without the page being blocked.
How does intent-based personalization improve results compared to exit intent alone?
It gives clarity and reassurance to high-intent visitors and guidance, not pressure, to lower-intent visitors. Print and Cheques Now, a Canadian B2B e-commerce supplier, saw its purchase conversion rate rise from 3.55% for unexposed traffic to 5.11% for traffic exposed to Pathmonk, a 44% increase in an A/B comparison.
- Popupsmart, popup conversion benchmark report (published 20 October 2025, updated 25 September 2026). Vendor data (popupsmart.com)
- Google Search Central, avoid intrusive interstitials and dialogs (last updated 10 December 2025) (developers.google.com)
- Google Search Central, understanding page experience in Google Search results (last updated 22 September 2026) (developers.google.com)
- Nielsen Norman Group, Popups: 10 problematic trends and alternatives (30 June 2019) (nngroup.com)
- Nielsen Norman Group, Modal and nonmodal dialogs: when (and when not) to use them (23 April 2017) (nngroup.com)
- Coalition for Better Ads, the Better Ads Standards (accessed 6 October 2026) (betterads.org)