Klaviyo Popups: The 4 You Are Not Running

Nearly every Klaviyo account we inherit runs one popup. It fires at everybody, asks for an email, hands over a code, and that’s the whole programme. The account owner tests the timing, maybe the headline, and wonders why the numbers plateau.

The problem isn’t the popup. It’s the count.

Five different kinds of people land on your site, and a single popup can only see one of them. The first-time visitor. The one who’s been before and never signed up. The one who signed up and never bought. The one who bought once and vanished. The one who unsubscribed six months ago and just came back on their own. Your one popup targets the first and ignores the other four, which on most stores is the larger half of the traffic.

Klaviyo will let you target a form to a segment, exactly the way you target a campaign. Almost nobody uses it. Below is the five-popup system we run on client accounts, how to build the first one properly, the settings we use, and the one configuration detail that decides whether three of the five ever fire at all. We’re a Klaviyo Elite Master, 2025-2026, and everything here was checked against Klaviyo’s own documentation on 25 September 2026.

Key Takeaways

• You can target a popup to a segment. Klaviyo has a setting called show to specific profiles in a list or segment, with include and exclude, and it updates automatically as people join and leave. That single setting is what turns one popup into five.
• Five audiences, five forms. Anonymous visitors, anonymous visitors on a second look, known profiles you are not allowed to email, known prospects who never ordered, and known customers you are under-asking.
• Three of the five depend on Klaviyo recognising the visitor. Safari caps first-party cookies set in JavaScript at seven days, so unless you have turned on Extended ID, your Safari traffic goes anonymous after a week and most of this system quietly stops working.
• Build full page, and build desktop and mobile separately. Klaviyo describes full page forms as high-converting because they cannot be easily overlooked, and a single shared form always looks wrong on one of the two.
• The button trap that silently loses your data. On a multi-step form, the button has to be set to Submit form. Set it to Close form and the properties collected on that step are never saved.

Why One Popup Leaves Most of Your Traffic Alone

Think about who is actually on your site right now.

Some of them have never seen your brand. Some arrived, looked, and left without signing up. Some gave you their email months ago, took the code, and never used it. Some bought once and haven’t been back. And some unsubscribed at some point and are now, for whatever reason, reading your product page again.

Those five people need five different things said to them. The first needs a reason to hand over an email. The fourth needs a reason to place a second order. The fifth is raising their hand after telling you to go away, which is about as strong a buying signal as you will get for free.

A single master popup speaks to the first one and, if it’s set up correctly, deliberately hides from the rest. That’s the right behaviour for that form. The mistake is stopping there, because it means four groups of people walk through your shop and nobody says anything to them.

Klaviyo documents every piece of what follows. What their getting-started guide doesn’t do is put the pieces together, which is why so few accounts run more than one.

The Five Popups, One at a Time

Popup one: the new visitor

The standard one, and the one you almost certainly have. Anonymous traffic, an incentive in exchange for an email, then a second step for text consent.

Two things make the difference between a form that collects addresses and a form that collects customers. First, ask a question before you ask for the email. Second, make the offer look like it’s worth having. Both get their own section below, because both are worth more than any timing test you’ll ever run.

Target it by excluding your email and text lists, so it stays away from anyone Klaviyo already knows.

Popup two: the double tap

Same audience, second attempt, later in the visit. Most accounts don’t have this one and it’s the cheapest of the five to add.

Anonymous visitors split into two types. Some will sign up on the spot, because the offer is good and they’ve decided quickly. Others need to read the product page first and work out whether they care. The first form catches the first group. The second form catches the second group, and without it they leave with nothing.

Set it to a longer delay than the first, somewhere between thirty-five and sixty seconds, add a scroll condition, and turn on exit intent. On desktop Klaviyo reads the mouse moving above the viewport or hovering at the edge of the page. On mobile there’s no mouse, so it fires on a fast upward scroll. It isn’t true exit intent on a phone, and it still catches people the first form missed.

If your brand will tolerate the teaser, the small tab that sits in the corner and reopens the form, that covers a lot of the same ground. Plenty of brands won’t have it because it looks cheap next to their design, which is a fair objection. The double tap is the version that doesn’t ask you to put anything permanent on the page.

Popup three: the profile you know and cannot reach

This is the one almost nobody runs, and it costs nothing.

Somebody gave you their email. Later they unsubscribed. They’re now outside every campaign and every flow you send, which means as far as your marketing is concerned they no longer exist. Then they come back to your website on their own.

That’s a person telling you they’re interested again, and you have no way of hearing it, because your main popup is configured to hide from profiles Klaviyo can identify and your campaigns are configured to skip anyone who can’t receive them. They fall through the gap.

The segment is short: profiles that cannot receive email and text marketing, and that haven’t been sent one recently. The recency condition matters. Someone who unsubscribed yesterday does not want a form asking them to come back, and putting one in front of them is how you turn an unsubscribe into a complaint. Give it a week, or thirty days if you send heavily.

The people who resubscribe here are people you already paid to acquire once. Getting them back costs you a form.

Popup four: signed up, never bought

Known profile, on your list, zero orders. They took the welcome code and then life happened.

Two ways to play it. Show them the same code again, because a surprising number of people genuinely lost it and have no idea where to look, or show them a larger one to push them over the line. Which of those you choose is a margin decision, not a marketing one, and it depends on what a first order is worth to you.

Build the segment from intent, not from time alone. Zero orders placed, plus some sign of life in the last few days: active on site, viewed a product, started a checkout, added to cart. You want the people who are circling, not the whole dormant list.

If you want to be careful about how many people see it, exclude your own Klaviyo tracking parameters so it only reaches visitors who didn’t arrive by clicking one of your emails. There’s no sense offering a code to somebody who is already reading the email that contains it.

Popup five: the customer you are under-asking

Same mechanic, pointed at people who have already bought, and it doesn’t have to be about another order.

The obvious version targets one-time customers after enough time has passed and nudges the second purchase, which is the hardest and most valuable transition in ecommerce. But the same targeting works for anything you’re missing:

• Email subscribers who never gave text consent, asked for text consent.
• Customers on their third or fourth one-off order, shown the subscribe and save page, because they’re paying more than they need to and probably don’t know it.
• A named segment of your best customers, shown something that only they see.

None of that is exotic. It’s the same setting as popup three with a different segment in it.

How to Build the First One Properly

If you have nothing at all, this is where to start. There’s a fifty-minute screen-share on our channel that builds one from an empty account click by click, including the branding and the mobile version. What follows is the short written version and the decisions that actually matter.

Choose full page. Klaviyo offers popup, flyout, full page, embed and banner. Their own documentation describes full page forms as high-converting, on the grounds that they cannot be easily overlooked, and that matches what we see. For an ecommerce store trying to maximise signups, we go full page nearly every time.

Start from scratch, not from a template. The templates aren’t bad, they just don’t look like your brand, and you’ll end up rebuilding them anyway. Set your brand colours and fonts as presets in Klaviyo first and the rest of the build takes minutes instead of an afternoon.

Build two forms, one for desktop and one for mobile. Clone the first, set one to desktop only and one to mobile only, and tune each. Copy that fits on a laptop will not fit on a phone, and padding that looks generous on a phone looks broken on a monitor. This is ten extra minutes and it’s the difference between a form that looks designed and one that looks approximately right. Worth knowing: Klaviyo’s desktop setting also covers tablets.

Make it multi-step. Step one asks a question. Step two takes the email. Step three is the success screen with the code. That order is the whole trick, and the next section explains why it works.

Then the trap. On the step that collects a property, the button’s Action must be set to Submit form. If it’s set to Close form, Klaviyo will not save the properties collected on that step. The form will look like it’s working, the profiles will come through, and the data you built the whole thing for will be missing. Check every button.

Put the code on the success screen. There’s a school of thought that says you should withhold it and tell people to check their email, on the grounds that it trains them to open your messages. We think that’s backwards. Look at what you’re asking: close the site, open a mail app, wait, find the message, copy a code, come back. That’s friction you added to serve your own metric, not theirs. Put the code on the screen, apply it at checkout if you can, and use the space underneath to give them a reason to open the first email instead of forcing them to.

The Targeting and Behaviour Settings We Use

This is where most of the value is, and where most accounts have never looked past the delay field. Everything below sits under Targeting and behavior in the form editor.

A note on the URL exclusions, because it’s the one that costs real money when it’s wrong. If you run paid traffic to a custom landing page with its own offer and its own form, the last thing you want is your generic site-wide popup landing on top of it. Exclude those URLs and let the page do its job. Same for checkout and cart pages: nobody who is mid-purchase needs a discount form in their face, and an unexpected code at that moment is a reason to abandon and go hunting for a better one.

And the timing. We start new forms at two to three seconds, not thirty. A full page form is a pattern interrupt, and the moment of highest attention is the moment they arrive. Test it against slower versions by all means, we have, and it very rarely wins. But run the test on your own traffic before you take our word for it.

The Setting That Quietly Breaks Three of the Five

Popups three, four and five only work if Klaviyo can tell who the visitor is. No recognition, no segment membership, no form. So it’s worth knowing exactly how long that recognition lasts.

Klaviyo sets its identifying cookie to live for up to two years. The browser decides whether it gets to. Safari caps first-party cookies set in JavaScript at seven days, which is a large share of mobile traffic on most DTC stores. Eight days after someone’s last visit, Safari has thrown the cookie away and that returning customer arrives at your site as a stranger.

Klaviyo’s answer is Extended ID, which re-identifies visitors against other identifiers already present in the browser and lengthens cookie-based identification from seven days to up to a year.

If you have not turned it on, here’s what’s actually happening. Your main popup is excluding known profiles, so it shows correctly to the strangers. But your customer who last visited three weeks ago now looks like a stranger too, so they get the new-subscriber offer instead of the second-order nudge, and the three targeted popups never fire for them at all. The system reads as broken when the configuration is just incomplete.

It takes a couple of minutes to switch on and it’s the first thing to check before blaming any of the rest of this.

Offer Stacking, and Where It Stops Being Fair

Here’s a problem most stores have and few have named.

Your ads and your product pages already carry discounts. Say a product page offers 21% off as a one-off purchase, or 40% off the first order on subscribe and save. Then your popup appears and offers 10% off. To a visitor who has just been told 40%, ten is not an incentive, it’s an insult, and the form converts accordingly.

You can’t hand out 40% in a popup code without wrecking your margin. What you can do is stop presenting the 10% on its own.

Stack it. Forty per cent off, then ten per cent off what’s left, is 46% off the original price. So the form says up to 46% off, which is a bigger number than the ad that brought them there, and you’re still only giving away ten points of margin in the code.

The rule that keeps this honest: the combined number has to be real and both parts have to be genuinely claimable together. If the subscribe and save discount and the popup code cannot actually be used on the same order, you don’t get to add them up. Say up to, show the maths if you’re asked, and make sure the checkout actually does it. Framing is fair. A number the basket won’t honour is not, and it’s the fastest way to turn a signup into a complaint.

Ask One Question Before You Ask for the Email

The single change that has done most for the accounts we run is putting a question in front of the email field.

Think about a physical shop. Someone walks in, and a good assistant doesn’t ask for their phone number. They ask what they’re looking for, and then they can actually help. Your popup has exactly one chance to do the same thing, and most brands spend it on a field labelled Email.

So: what is the one thing you could ask a visitor that would let you talk to them properly from that day on?

When brands get stuck on this, the answer is almost always a goal or a problem. Not a demographic, not a favourite colour. What are they trying to achieve, or what are they trying to get rid of. Those convert best because they’re the thing the visitor is already thinking about.

The clearest example we’ve worked on is in fertility. There is a world of difference between somebody trying to conceive and somebody who already is, and if you send the wrong one the wrong email you don’t just miss the sale, you can do real damage to how they feel about the brand. Asking on the first screen means you never have to guess. Klaviyo published what that programme produced.

Two practical notes before you build it.

Create the property deliberately, before you touch the form. Open a profile in Klaviyo, add the custom property there, and decide what the values are going to be. You can type a new property straight into the form builder, and that’s how accounts end up with three spellings of the same field and segments that quietly miss a third of their members.

Give people a way past the question. A fourth button that says something like save the discount for later, styled quietly so it doesn’t compete, collects no data and still moves them to the email step. A meaningful number of people take it, and you’d rather have the subscriber without the answer than neither.

There’s a second reason the question step works, beyond the data. It’s a small commitment, and people who have just made a small commitment are more likely to finish the larger one. Asking for the email cold is a bigger ask arriving with no warm-up.

What to A/B Test, in Order

Your popup is the front door to the whole Klaviyo account. Everything downstream, every campaign and every flow, is fed by it, so a few points of improvement here compound through everything else. It should be under test permanently, not once a year.

The order we work in:

• Timing first. It’s the easiest test to build and it tells you something immediately. Two or three seconds against thirty, maybe a ten-second version in between.
• Then how the offer is framed. This is the big one by a distance. Not the discount itself, how it’s presented. Stacking, a gift with a stated value, a deadline, a mystery reward. Same margin, very different numbers.
• Then the question and the button wording. Smaller, still real.
• Then microcopy. A line of social proof, a note about the offer ending, a quiet opt-out that says something like I would rather pay full price. Cheap to add and occasionally surprising.
• Imagery last. It moves the least. Test it when you have run out of better ideas.

Two things that will save you from your own tests. Name them properly. Klaviyo’s own guidance is to name a test descriptively and give each variation a distinguishing name, and hardly anyone does it. In three months you will not remember what Copy 2 of Newsletter was, and neither will whoever inherits the account. And don’t run six variations. Klaviyo allows up to seven; two to four is the sensible range. Split a modest amount of traffic six ways and you’ll wait months for an answer, then act on noise.

But Popups Annoy People

This used to be true and it stopped being true, for an unglamorous reason. Popups always worked, so everyone kept using them, and now visitors expect one. An ecommerce site without a popup is the unusual thing. The format stopped being an interruption and became a convention.

The version of this objection we meet most often comes from acquisition teams, who worry a popup will hurt the performance of the traffic they’re paying for. In our experience it does the opposite, and we’ve yet to lose that test on an account of any size. The effect tends to be largest on smaller stores, where the landing pages are less optimised and the offer does more work.

There’s one genuine cost, and it isn’t annoyance. It’s a generic popup landing on a page that was built around a specific offer, where it talks over your own campaign and lowers the value of what you were saying. That’s a targeting failure rather than an argument against popups, and the URL exclusions in the table above are the fix.

Where This Sits in the Rest of the Work

A popup is the top of a system, and the parts underneath it decide what the signups are worth.

• The free Klaviyo audit - we look at the forms, the flows behind them and the offer in front of them.
• The LTV Parthenon - our full retention methodology, and where acquisition sits inside it.
• Email deliverability - the constraint that decides whether the addresses you collect are worth anything.
• The best retention marketing tools - the wider stack around Klaviyo.

So Where Do You Actually Start?

In this order.

Build one full page form properly, with the question step in front of the email, the code on the success screen, and separate desktop and mobile versions. Check every button says Submit form. Then turn on Extended ID, because three of the next four depend on it. Then add the double tap, which takes ten minutes and is pure incremental signups. Then the three segment-targeted forms, in whichever order matches where your money is.

Most stores have the first one and nothing else, which means four of the five are simply missing. Adding them is an afternoon of work against a channel you are already paying to fill.

If you want somebody to look at what you have now and tell you which of the five is worth building first, that’s part of our free Klaviyo audit. We go through the account with you on a call, and you keep the findings whether or not you work with us.

Frequently Asked Questions

How do I create a popup in Klaviyo?

In Klaviyo, go to Sign-up forms, make sure the Klaviyo snippet is installed on your site, then create a form and choose a format. We build full page forms from scratch instead of starting from a template, set brand colours and fonts as presets first, and use a multi-step layout: a question, then the email field, then a success screen carrying the discount code. Build a separate version for mobile by cloning the desktop one and setting each to its own device.

How many popups should a Klaviyo account have?

More than one. We run five: a main form for new visitors, a second one later in the same visit for people who did not sign up the first time, and three targeted by segment at profiles Klaviyo already recognises, namely people who unsubscribed and came back, subscribers who never placed an order, and existing customers. A single form can only reach anonymous traffic, which leaves most of your returning visitors with nothing.

What is the best popup format in Klaviyo?

Full page, for most ecommerce stores. Klaviyo offers popup, flyout, full page, embed and banner formats, and describes full page forms as high-converting because they cannot be easily overlooked. Flyouts and banners are less disruptive and convert less. Embeds are useful on a footer or a dedicated page for people who arrive intending to subscribe.

Can you target a Klaviyo popup to a segment?

Yes. Under Targeting and behavior, choose visitors in a list or segment, then show to specific profiles in a list or segment. You can include segments, exclude them, or both, and membership updates automatically as people join and leave. This is the setting that makes it possible to run different forms for different audiences rather than one form for everybody.

Does an exit intent popup work on mobile?

Partly. On desktop Klaviyo triggers the form when the cursor moves above the viewport or hovers at the edge of the page. Phones have no cursor, so on mobile the same setting fires when somebody scrolls quickly upward. It is not true exit intent, and it still catches visitors who would otherwise have left without seeing a second offer.

Should the discount code be shown in the popup or emailed?

Show it on the success screen, and apply it at checkout if your platform allows. Withholding it to make people open the welcome email adds several steps between a visitor and a purchase in order to improve one of your own metrics. Use the space under the code to give a reason to open the first email instead of making the code conditional on it.

Why do my Klaviyo popups stop recognising returning visitors?

Usually the browser cookie. Klaviyo sets its identifying cookie for up to two years, but Safari caps first-party cookies set in JavaScript at seven days, so a returning Safari visitor looks anonymous after a week. Klaviyo Extended ID re-identifies those visitors and lengthens recognition from seven days to up to a year. Any popup targeted at a segment depends on it.

Do popups hurt paid ad performance?

In our experience they improve it, and the effect is usually larger on smaller stores where the landing pages are doing less work. The real risk is not the popup itself but pointing a generic site-wide form at a landing page that already carries its own offer, where the two compete. Exclude those URLs and run a dedicated form there instead.

If you would rather have someone run this against your account, our free Klaviyo audit does that, and you keep the scorecard.