10 Website Popup Examples That Actually Convert

What every good popup example has in common

Before the list: four things are true of all ten, and if you get them wrong the example won't work no matter how well you copy it.

• It arrives after something, not before. A delay, a scroll, a second page, an exit signal, a completed order. A popup that fires on arrival is asking a stranger for a favour before saying hello.
• It asks for one thing. One field, or one tap. Every extra field is a decision, not a keystroke — see the progressive profiling guide for how to spread the rest across later asks.
• It has a visible way out. A real close button and, ideally, a worded decline link. Hiding the exit doesn't raise conversions; it raises rage-clicks.
• It stops when it's done its job. Frequency capped, and the conversion gate switched off so converters don't see it again.

The ten jobs at a glance

1. The discount capture popup

The job: turn a first-time browser into a subscriber by trading a first-order discount for an email address. This is the workhorse of ecommerce popups and the one most sites get almost right.

When to fire it: around 12 seconds after page load, on product and collection pages, to new visitors. Not on the cart, not on the checkout — someone already buying does not need a reason to pause and go hunting for a better code.

The rules behind it:

• Content type: popup, one Email input, Required on.
• Trigger: Time Delay, 12 seconds — counted from page load.
• Targeting: page URL does not contain /cart, plus a second rule that it does not contain /checkout.
• Audience: new visitors (the default window is 30 days, and you can change it).
• Frequency: once every 3 days. Turn the frequency rule on first — the audience and conversion controls only appear once it is.
• Show again after conversion: off.

2. The newsletter signup popup

The job: build a list where you have no discount to give — a blog, a media site, a B2B product. The trade here is content for attention, so the copy has to do the work the discount would have done.

When to fire it: at 50% page scroll on article pages, or on the second page of a session. Both signals mean "this person is actually reading", which is the only qualification that matters for a newsletter.

The rules behind it: a popup, or a standalone form, with a single Email field; trigger Page Scroll at 50%, or Page Count at 2 pages (counted within the current browser session); frequency once every 7 days; conversion gate off. Set the email validation mode to check format and prevent typos, or, on a B2B site, to accept only business emails so you stop collecting throwaway addresses. Keep the division of labour straight: the popup captures and stores the addresses, which you read in the campaign's Collected input data table — the Friday email itself goes out from whatever email tool you already write your newsletter in.

A newsletter capture built in the editor. Two fields is already one more than this job needs — drop the name unless you genuinely personalise subject lines.

The mistake that kills it: the headline "Subscribe to our newsletter". It describes the mechanism, not the payload. Name what arrives, how often, and what doesn't arrive. The word "weekly" converts better than any button colour you will ever test.

3. The exit-intent save

The job: make one last offer to someone who is leaving anyway. Because it fires on the way out, it costs you nothing in interrupted browsing — which is why it's the safest popup on this list to run site-wide.

When to fire it: on the exit signal, desktop only. There is no reliable mouse-out gesture on a phone, so the mobile version of this campaign is a separate one with the device rule set to mobile and either a 60% scroll trigger or an inactivity trigger instead.

The rules behind it: popup, trigger Exit Intent, device rule Desktop, frequency once a day, conversion gate off. Build the mobile twin as its own campaign with device Mobile and Page Scroll at 60% — same copy, different trigger. There's a fuller breakdown of the signal and its failure modes in the guide to exit intent popups.

The mistake that kills it: running one exit-intent campaign for everything and wondering why half your traffic never sees it. Check the shows-by-device breakdown in analytics; if mobile is near zero, that's the reason, not your offer.

4. The cart-abandonment offer

The job: catch a session that has items in the cart before it closes. This is the highest-value moment on the whole site, and the popup is doing something an email can't: arriving while the intent is still live.

When to fire it: on the cart or checkout URL, on inactivity (roughly 30 seconds of nothing happening) or on exit intent. Both signal hesitation rather than abandonment, which is the window you actually want.

The rules behind it: popup with page URL contains /cart, trigger Inactivity, frequency once a day. Set the submit action to Link to so the button takes them straight back to the checkout URL. If you want the same rescue on the checkout step, that's a second campaign with its own URL rule.

The mistake that kills it: discounting someone who was going to buy anyway. Reach for reassurance first — shipping, returns, sizing help — and keep the money for the exit-intent campaign one step earlier in the funnel.

5. The announcement bar

The job: say one factual thing to everybody without interrupting anybody. Sale dates, a shipping cut-off, a delivery delay, a new feature. It is the only example here that is allowed to appear on every page, all the time.

When to fire it: one second after page load, everywhere, every visit. It's a fixture, not an event.

The rules behind it: there is no separate "bar" content type — you build it as a popup sized wide and short (something like 1000 × 90 px in the editor's Width and Height fields) and placed at the top of the screen. Trigger Time Delay, 1 second; frequency all pages all the time; and — the part everyone forgets — a date window in the targeting rules so it retires itself the morning after the sale ends. For the "ships today" version, use the weekday-and-hour schedule instead so it only appears when it's true.

The mistake that kills it: leaving it up after the deadline. A stale bar teaches regulars to ignore the top of your site, and every campaign you run afterwards pays for it. Set the date window when you build it, not when you remember.

6. The product-finder quiz

The job: convert "I don't know what I need" into a segment and a starting point. A quiz earns an email that a plain form wouldn't, because by the time the email screen appears the visitor has already invested three taps.

When to fire it: 8 seconds into a collection page, or on the second page of a session. Best on catalogues wide enough to be genuinely confusing.

Quiz questions are built in the Choices panel as single- or multiple-choice lists. Tappable options return clean values you can segment on; free-text answers don't.

The rules behind it: a quiz with three question steps, the progress bar on and a fade transition between them. Mark the choice blocks Required so a visitor can't skip past them, and put the email on the last step. Everything they answered arrives as one submission at the end. A quiz always finishes on its own thank-you step, so that's where the payoff goes: write it there and add a button block whose click action is Link to, pointing at one curated collection.

The mistake that kills it: eight questions. Three is the format; four is the ceiling. If you're not sure which one to cut, the Drop-off Analysis on the quiz's results page ranks the fields by how often they were the last thing a visitor touched before quitting — that's your cut list, already sorted.

7. The feedback survey

The job: find out why the numbers look the way they do. Analytics tells you people left the pricing page; only a survey tells you it was because the plan comparison was unreadable.

When to fire it: after the visitor has experienced something. On the order-confirmation page, after a help-centre article, or 20 seconds into a page you know is underperforming. Never on the homepage to a first-time visitor — they have nothing to report yet.

The scoring panel: star rating, CSAT, CES, thumbs up/down and NPS, each with configurable endpoint labels. Pick one — a survey with three different scales is a chore, not a question.

The rules behind it: a survey with the NPS widget on step one and an optional long-text follow-up on step two; trigger Time Delay 20 seconds; page URL contains /thank-you; frequency once every 90 days so regulars aren't polled every month. Surveys and quizzes are the two content types that also get the Results view — a four-stage funnel, completion rate, average open-to-submit time, an NPS card and a day-by-hour heatmap of when responses actually arrive. There's more on question design in the website feedback survey guide.

The mistake that kills it: making the follow-up question required. The score is the data you need; the comment is the bonus. Requiring the comment converts a 3-second answer into a writing task and you lose both.

8. The spin-to-win game

The job: make the email trade feel like a reward rather than a toll. Mechanically it's the same exchange as example 1 — an address for a discount — but the order is reversed: the visitor wins first, then hands over the email to claim.

When to fire it: 15 seconds into the homepage or a collection page, to new visitors, or on exit intent. It suits promotional periods far better than everyday trading.

A spin-to-win overlay running live on a store. Prize odds are set per prize and the outcome is decided server-side, so the result can't be rigged from the browser.

The rules behind it: a gamification campaign — spin wheel, scratch card, slot machine or claw machine, they all share the same screen model. The claim screen collects the details, and submitting it is what reveals the prize. Trigger Time Delay 15 seconds, audience new visitors, frequency once every 7 days, conversion gate off. The spin to win popup guide covers prize odds and slice design in more depth.

The mistake that kills it: a losing screen that's a dead end. If a third of players get "Try again" and nothing else, you've built a machine that annoys a third of your visitors on purpose. Give the loser a small consolation and you keep the email either way.

9. The gated-content unlock

The job: trade a genuinely useful artefact — a checklist, a template, a calculator, a swipe file — for an email, on the one page where somebody has already proved they care about the topic.

When to fire it: at 60% scroll on the specific article — deep enough that the reader has hit the part of the piece the checklist summarises. The deeper the scroll, the warmer the ask.

The rules behind it: a popup triggered at Page Scroll 60% with a page URL rule that starts with /blog/ — or the same capture built as a form and ruled to that one article. Because ChilliPopup doesn't send the file, set the submit action to Link to and point it at the asset URL, or put the download link on the thank-you screen. Delivery is instant either way, which is a better experience than "check your inbox" anyway.

The mistake that kills it: gating something the article already gave away. If the checklist is just the h2 headings retyped, the ask reads as a toll booth. The upgrade has to be a different format or a genuine extension — the spreadsheet, the template, the version they can hand to a colleague.

10. The post-purchase profiling ask

The job: collect the one field you couldn't justify asking for earlier, at the single moment when the customer is most inclined to give it — right after the money has changed hands and nothing is at stake.

When to fire it: 5 seconds after the order-confirmation page loads. One question. Never two.

The rules behind it: a popup or a two-step survey with page URL contains /order-confirmation, trigger Time Delay 5 seconds, frequency once every 365 days, conversion gate off. ChilliPopup stores the birthday against that submission; the message you promised on the day comes from your own email tool, not from the popup. This is the third rung of a progressive profiling ladder — the guide covers which field belongs on which rung and why the expensive ones have to come last.

The mistake that kills it: asking for something the checkout already captured. If your order flow has the phone number, requesting it again "for delivery updates" tells the customer your systems don't talk to each other — which, at that exact moment, is the last thing you want them thinking about.

Before you copy any of these: the settings that decide the outcome

Nine times out of ten, a popup that "doesn't work" is a rules problem, not a design problem. Here is the short checklist to run before you publish any of the ten.

Every campaign reports views, unique views, clicks, submissions, conversion and field inputs, plus a breakdown of the top values collected per field.

One more piece of honesty about measurement: there is no built-in A/B test, so don't try to read two variants against each other. Change one thing — the headline, the trigger, the delay — publish it, and compare the campaign against its own previous fortnight. Conversion is reported as submissions divided by views, which means a campaign that fires rarely to a warm audience will always flatter a campaign that fires constantly to everyone. Compare like with like, or don't compare at all.

Which two should you build first?

Not all ten. Pick one that captures an email and one that does something else entirely, so the two never compete for the same moment:

• Ecommerce store: the discount capture (example 1) plus the announcement bar (example 5). Add the cart-abandonment offer once the first two are earning.
• Blog or media site: the newsletter signup (example 2) plus the gated-content unlock (example 9) on your three best-performing posts.
• SaaS or B2B: the newsletter signup with business-email validation on, plus the feedback survey (example 7) on the pricing page.
• Running a promotion: the spin-to-win game (example 8) plus the announcement bar — and a date window on both so they retire themselves.

If you're starting from a blank editor, the fastest route is the AI generator: describe the popup in one sentence — "an exit-intent popup offering 10% off a first order, warm cream background, one email field" — and it drafts the layout, colours, fonts, copy and blocks for you to edit. It's on every plan with a fair-use daily limit. And if the pixel isn't on your site yet, the one-line install guide takes a couple of minutes; the same script serves all ten of these, so there's nothing new to embed as you add campaigns.

Related reading

Frequently asked questions