Tested against the pop-ups stores already had.
Results from three clients, described by what they sell rather than by name. Every number has its dates beside it.
- 01Does it beat my pop-up?+21% conversion rate against a Klaviyo pop-up in an A/B test, 8 to 20 October 2026.
- 02Are the sign-ups any good?Unsubscribes and first orders held level with the old pop-ups.
- 03Is what they say worth anything?Of buyers who said they prefer shakes, 44% bought one. Of buyers who said no, 12% did. 23 Sep to 25 Oct 2026.
- 04Does it work off the store?A newsletter and a brand email ran the same cards from a link.
In a straight test, it beat a Klaviyo pop-up.
The store split its traffic between its Klaviyo pop-up and ours, with the same offer on both. Its conversion agency ran the test and called it for SVYP on 20 October. Since then ours is the one every visitor sees.
Figures as the agency declared them.
It also replaced the old pop-up at two stores.
The same retailer's home store, and the health-food brand. Before and after, on each store's own data.
of the homeware retailer's home-store visitors submitted the pop-up, against 2.52% for the Klaviyo one it turned off8 Jul to 8 Aug against June 2026, same 15% offer. Submissions per Shopify visitor, not distinct people. New subscribers per visitor held level
as many new subscribers a day at the health-food brand as its old pop-up got in the same weeks of 2025: 104 a day against 2723 Sep to 11 Oct, 2026 against 2025. Daily counts: the store's visitors for both years are not in our data
The people it signs up behave like the old ones did.
A pop-up that collects more emails is only worth it if those people stay and buy. At both stores where we can compare, they did.
of the homeware retailer's new subscribers unsubscribed within 14 days, against 9.8% from the pop-up it replacedCohorts from June and 3 to 19 July 2026. The gap is too small to call
of the health-food brand's brand-new subscribers ordered within 14 days, against 18.2% from its old pop-upSame weeks, 2026 against 2025. The rate held while the volume grew 3.8x
of the 5,817 people who gave the health-food brand an email have already orderedFirst 33 days, 23 Sep to 25 Oct 2026
What shoppers told the cards matched what they bought.
Every answer is saved on the shopper's profile. At the health-food brand we checked the answers against later orders.
Said yes, against said no.
Among people who went on to buy, those who said yes to a card bought that product far more often than those who said no. Each answer is a list the brand can send to.
Share of buyers whose orders held that product, among 911 people who signed up with a mix of yes and no answers and then bought, 23 September to 25 October 2026. Every gap here is statistically significant. One card, on gut health, showed no real gap and is not shown.
And the answers move with the season.
At the homeware retailer, the same cards in July and in September. Gifting and garden lighting rose into spring and the heater card fell.
Share answering yes among shoppers who said yes to some cards and no to others, July 2026 against 1 to 27 September 2026, about 850 answers a card in September. Each change is statistically significant.
The same cards run from a link.
The link can go in an email or a WhatsApp message. Two clients tried it in October.
The health-food brand put them in an email to its list.
About 50,000 subscribers got it, and the link to the cards was its main click: 562 of the 707 people who clicked anything.
existing subscribers swiped the cards and claimed the code that day, against 0 to 6 on a normal day9 Oct 2026. Customers the brand already had, not new sign-ups. Total orders that day were no higher than an ordinary day
A business newsletter ran them in one morning edition.
Readers who answered went into a prize draw. 8 to 13 October 2026.
See it on your own store first.
Sources: the A/B test as the store's conversion agency declared it, 8 to 20 October 2026; the homeware retailer's Shopify and Klaviyo data and SVYP's, June to September 2026; the health-food brand's Klaviyo data and SVYP's, 23 September to 25 October 2026, against the same weeks of 2025; SVYP's data for the newsletter, 8 to 13 October 2026.