All Latest 650 A/B Tests

In this checkout experiment, an upgrade offer was added directly above the contact and delivery form. Below the express checkout buttons, the variant showed the item already in the cart (a single bottle priced at $28) and invited shoppers to upgrade to a 3-month supply for only $8 more. The bundle was shown at $36 next to a crossed-out $60, with arrows to browse other upgrade options and a prominent “Upgrade” button. As a result, the standard checkout form was pushed further down the page, behind a second “OR” divider. Imact on completed orders was measured.

In this retest experiment, delivery copy was added under each product tile, letting shoppers know they could send an email gift message today. The idea was to reassure last-minute gift buyers that the recipient would hear about their gift right away, even before the physical delivery arrives. Impact on checkout visits and sales was measured.

This was a retest with an additional checkmark style for additional contrast and visibility. 

In this experiment, the number of displayed tile results were modified (fewer vs more listing tiles shown per page). Impact on lead starts and lead completions were measured (Desktop segment).

In this experiment, the favorite icons (hearts) were: 1) highlighted more prominently using a darker border color, and 2) when users hovered over them with a mouse, the heart icon filled with a red color. Impact on favorite clicks, adds to cart and sales was measured.

In this experiment, delivery copy was added under each product tile, letting shoppers know they could send an email gift message today. The idea was to reassure last-minute gift buyers that the recipient would hear about their gift right away, even before the physical delivery arrives. Impact on checkout visits and sales was measured.

This experiment ran on an ecommerce store utilizing Shopify's checkout. The treatment compared the checkout experience with and without a limited-time countdown timer. Impact on placed orders was measured. The A/B test ran using PrettyDamnQuick software.

In this experiment, when users added something to cart, they would be automatically sent to checkout (one step forward), instead of staying on the product page. Impact on checkout visits and sales was measured. The test was triggered to PDP page visits.

This was a sibling experiment on another domain.

An additional step was added in a lead ffunnel with a question about the customer's desired "type of travel experience". Impact on lead rate was measured. (Desktop only)

This experiment ran on an ecommerce store utilizing Shopify's checkout. The treatment compared the checkout experience with and without a limited-time countdown timer. Impact on placed orders was measured. The A/B test ran using PrettyDamnQuick software.

This experiment ran on search results page of an ecommerce grocery store.

50% of users saw product cards that upon clicking would open in the same window (Control).

The other 50% saw product cards that upon clicking would open in a new tab/window (Treatment).

Impact on adds to cart and orders was measured.

We should also add that this experiment was run as part of the Trustworthy Replication Project and only "adds-to-cart" were pre-registered as the primary metric - https://osf.io/yu4gm/overview

This experiment ran on an ecommerce store utilizing Shopify's checkout. The treatment showed a single testimonial and impact on placed orders was measured. The A/B test ran using PrettyDamnQuick software.

In this experiment, users were able to auto-generate gift messages during the checkout flow based on input fields above. For this reason, the message input area was also shifted further down beyond the initial fields.

In this experiment, instead of utilizing the browser auto-complete for the street address, the Google API call was made using the postal/zip code as the input. Once users would begin typing a street address, a pull-down with matching recommendations would appear. Impact on next step visits (to checkout) and sales were measured.

In this experiment, when users added something to cart, they would be automatically sent to checkout (one step forward), instead of staying on the product page. Impact on checkout visits and sales was measured. The test was triggered to PDP page visits.

In this experiment, users who viewed a product and then visited a listing or search results page, saw "Recently Viewed" badges on top of the corresponding product tiles. Impact on adds to cart and sales was measured.

In this experiment (sibling of 656), a sticky add-to-cart button was added to the floating page navigation after users scrolled past the midpoint of a long product page, at the same moment the navigation also became a floating element. After users clicked the button, it changed into a confirmation link labeled “Item added to cart,” with a link to checkout.

In this experiment, a sticky add-to-cart button was added to the floating page navigation after users scrolled past the midpoint of a long product page, at the same moment the navigation also became a floating element. After users clicked the button, it changed into a confirmation link labeled “Item added to cart,” with a link to checkout.

In this experiment, an additional ongoing club was tested in a floating buy box on top of 4 other subscription durations. Impact on adds to cart and sales was measured.

In this experiment, product configuration options were moved from a floating buy box on the right, to wards the center of the product page - increasing their prominence. Impact on adds to cart and sales was measured.

In this experiment, returning homepage users were shown their recently viewed items in a carousel (in the upper part of the homepage). Users who viewed at least 1 product were triggered into the experiment. Impact on adds to cart and sales was measured.