153 A/B Testing Ideas for eCommerce (Homepage, PDP, Cart, Checkout)

While the whole of eCommerce speaks of conversions, so much of that success lies in the powers of A/B testing.

According to a Speero report, 40% of CRO testers make archives out of their past tests, so that they can learn from them in the future.

Exactly why we thought, why not create a comprehensive list of variants we've tested, learnt from and helped clients with to earn more conversions.

With this one, you have your entire funnel covered PLUS a section on emails—here's a quick shortcut to the main sections:

Homepage

Category Page

Product Page

Cart Page

Checkout Page

BONUS: Emails

Let's go!

eCommerce A/B Testing Elements To Boost Conversions

Ideas for Stage One: Homepage

1. Variant A: use a mega menu / Variant B: use a horizontal menu

In our experience, we’ve noticed mega menus repeatedly work with businesses that have countless categories and product offerings. 

Horizontal menus work better for businesses with limited products and those that prioritize the mobile experience. 

2. Variant A: sticky navigation menu / Variant B: non-sticky menu

At the top of the funnel, when a shopper is trying to get to know your brand in more detail, a sticky menu might just work better—it offers more navigational control across the homepage and beyond. 

3. Variant A: feature labels for primary categories / Variant B: feature labels for sub-categories

Anchor & Crew highlights labels for their main categories—though the more usual practice for brands is to label their dropdown subcategories. 

4.  Variant A: feature a rotating carousel / Variant B: feature single hero image‍

Two bits from experience: even if you’re using a rotating carousel, make sure you offer manual control through visible previous-next buttons—and even if you choose to keep it on auto-rotate, let the speed of it be slow enough for shoppers to register what’s on each slide. 

5. Variant A: copy against the image / Variant B: hero space divided into image & solid background for copy

eCommerce brand Morphe Cosmetics places their copy (unobtrusively) against the hero image. 

Larq, on the other hand, splits the hero banner between plain background for copy and then image for their eCommerce A/B testing.

6. Variant A: a short headline / Variant B: a longer headline

Fitbit, for example, keeps it super short. 

Beardbrand, on the other hand, keeps it simple yet longer in this eCommerce A/B testing idea.

7.  Variant A: make the headline tone descriptive / Variant B: ask a question to make them think

While a description typically sets more context for the shopper, a question pushes them to think about the problem at hand that has brought them to the homepage.  

8. Variant A: make the headline statement brand-specific / Variant B: make the headline statement about a product or a line of products 

9. Variant A: hero banner copy featuring a mix of ingredients/features and benefits / Variant B: hero banner copy featuring just benefits

10. Variant A: feature the benefits in bullets / Variant B: feature the benefits in a paragraph

11. Variant A: flank the hero header headline with microcopy / Variant B: flank the hero header headline without microcopy

Remember: the primary role of microcopy is to take an argument (made in the headline, let’s say) further: it can talk about value-add, a benefit, a limited time offer etc. 

12. Test the above-the-fold CTA button (for different shapes, color, texts, positions etc.)

13. Variant A: feature only one CTA on the hero banner / Variant B: feature multiple CTAs 

Crossrope, for example, currently features two CTAs on their homepage hero space—this is one of the underrated eCommerce A/B testing ideas that can help you segment your audiences better:

14. Variant A: various prominent CTAs spread across the homepage / Variant B: some CTAs more prominent than others

To make certain CTAs less prominent, you might want to skip not color-blocking them—while the more prominent ones can be color-blocked. 

15. Variant A: button CTAs through the homepage / Variant B: a combination of usual CTAs plus visual cues on the images

Ikea, for example, features a range of CTAs on their homepage (notice how the visual cues are also on point):

16. Variant A: feature an email sign-up prompt in the hero header / Variant B: feature a seasonal sale discount

17. Variant A: feature a product quiz in the primary navigation / Variant B: feature a sticky “product finder” button

Adore Beauty features a sticky product finder button on their homepage—making it easy for those on-the-go to explore without poring over pages of content. 

18. Variant A: featuring more copy throughout / Variant B: featuring more visuals

Beardbrand clearly features more copy in this example of A/B testing for eCommmerce:

Room & Board’s homepage, on the other hand, is clearly more visual-led—how you use either depends a lot on what you’re trying to establish as a brand. 

19. Variant A: feature more product visuals / Variant B: feature more people visuals

What will work will have a lot to do with what kind of business you are—for example, if you sell jewelry, bold “people” photos will work better than if you’re into selling outdoor cooking equipment. 

20. Variant A: showcase a brand video / Variant B: feature a hero product video

21. Variant B: feature images of hero products / Variant B: feature GIFs of hero products

While technically both are image formats, GIFs when used well, convey more information and help set greater context for the shopper. 

22. Try one variant with a founder’s message & the other with brand values (in the form of graphic icons with text callouts)

23. Test one variant with a social proof badge on the hero image vs. another with press mentions

Huel, for example, showcases a Trustpilot badge on their hero space as part of their a/b testing for eCommerce ideas:

24. Test between “free shipping” & “free mystery gift” callouts on the strip above primary navigation

25. Variant A: feature a link to customer reviews / Variant B: feature quotes from customer reviews

Ritual is an eCommerce brand that seems to have done both in the same space—reviews highlighted as quotes and video links too!

26. Variant A: feature only customer review callouts / Variant B: feature both reviews AND press mentions

27. Try one variant with press mentions & the other with the impact every purchase creates

Unlike many other brands, Unconditional chooses to throw light on how a purchase creates impact—which is a deeper peek for anyone who is at the top of the funnel. 

28. Test one version that features an “about” section with link & another with brand promises called out through icons

29. Variant A: feature a discounted categories section / Variant B: feature a “deal of the week” section 

Burrow cites categories along with their associated discounts in their eCommerce a/b testing efforts.

On the other hand, Ikea highlights various categories as deals of the week.

30. Variant A: feature “shop by categories” / Variant B: feature “shop by brands” and show clickable logos

31. Variant A: feature BNPL options in the footer / Variant B: highlight popular payment methods 

eCommerce brand Unconditional does the latter. 

From experience, we’ve seen that highlighting BNPL options work really well if your primary target audience is the younger crowd—after all, 56% of millennials claim to use BNPL options widely. 

32. Variant A: showcase a section on loyalty program savings / Variant B: feature a section on regular flash sales

33. Test one variant with social media buttons on the footer & another on the upper navigation panel

34. Try one variant with a social wall featuring influencers vs. another featuring real customers

35. Variant A: feature randomly picked images for the social wall / Variant B: showcase handpicked ones

We love how cleverly Crossrope brings their social wall alive on the homepage—they feature only those customer cases that have a marked visual difference. 

36. Try one variant with links to brand/values vs. another variant with featured blogs

37. Variant A: feature tiles of latest blogs / Variant B: a “learn” section in the primary navigation with detailed sub-sections

Kettle & Fire, for example, ensures learning resources feature in their non-transactional part of the navigation menu:

38. Variant A: show email signup alert above/below primary navigation / Variant B: show email signup alert on the footer

39. Test between a shorter homepage version & a longer one 

Remember: even if you want a short homepage, you’ll need to create a balance between transactional and non-transactional elements on it.

Do read: 28 Inspiring eCommerce Homepage Examples (not your usual brands)

Ideas for Stage Two: Category Page

40. Test positioning the filter vertically on the left vs. horizontally in the center at the top

41. Variant A: sub-categories appear as plain / Variant B: sub-categories are highlighted through buttons 

Nordstrom features links to sub-categories to a/b split for eCommerce:

‍

ASOS, on the other hand, features button blocks. 

42. Variant A: feature an above-the-fold banner that shows a sale/discount / Variant B: feature a banner which just talks about the category

Fabkids features the first idea in action.

100% Pure calls attention to the category you’re visiting. If you want to announce a special discount on a specific category, the banner space is ideal for drawing attention. 

43. Test one variant with the primary CTA featured beneath the product image vs. another where there is no primary CTA 

Mejuri does not use add to cart buttons on their category page—this can often compel shoppers to go into individual product pages to know more.

44.  Try one variant with the wishlisting icon & the other without it

45. Variant A: feature wishlisting icon on the image / Variant B: feature wishlisting icon next to the product details

46. Test one variant with only pricing for sorting & the other with additional features (like recommended, newest/oldest, customer rating)

47. Variant A: feature 5-star rated discounted product suggestions in the top row / Variant B: feature only discounted suggestions in the top row

48. Variant A: feature average of start ratings displayed as a number / Variant B: mention no. of reviews & make it clickable 

Sephora ensures every product carries the average rating displayed through the shading of the stars, plus the number of reviews made. 

49. Variant A: rating stars & no. of reviews constantly viewable / Variant B:  ratings & reviews info available on hover

The latter can potentially pique the shopper’s interest a little more than otherwise—and hovering to see the review can become one step in the overall category page engagement. (Beardbrand does it.)

50. Variant A: offer a “quick view” option / Variant B: introduce the “quick add” option directly

51. Variant A: feature a Sale/Special Price label & reduced price / Variant B: feature Sale/Special Price label & price anchoring 

52. Test one variation with the “Selling Fast” label vs. another with “X pieces left only!”

53. Variant A: feature the new price marked in a bold color / Variant B: also feature the old price greyed out and struck through 

Nordstrom features both new and old prices for shopper satisfaction—the new prices feature in a strong, bold color. 

54. Variant A: show old price greyed out and new lower price in a different color / Variant B: add the additional feature of discount savings being called out

Here’s how H&M displays a discount drop in price and also uses anchoring to show both the old & new reduced price in their a/b testing for eCommerce:

55. Variant A: describe size range in the image description / Variant B: use a callout “available in various sizes” upon hover 

Zalando uses the second idea in their category pages—and it’s helpful for shoppers scrolling through because this simple callout can lead them further into the respective product pages. 

56. Try one variant where “Sold Out” is used as a label vs. another that uses “Notify Me”

57. Variant A: all color options can be viewed with previous–next buttons / Variant B: the most popular colors are displayed along with a “+X colors available” label

58. Try one variant where all the color buttons are displayed irrespective of availability vs. another where the unavailable colors are struck off

59. Variant A: feature a sticky email sign-up discount alert on the left panel / Variant B: show social proof labeled “X from so-and-so place bought <name of the product” & image”

Built Athletics uses the second idea—and what we noticed is that the social proof features a product shown on the page you’re at, inciting curiosity for the shopper. 

60. Test one variant with a descriptive callout for the product on the image vs. another where there’s no callout

Huel, for example, makes sure each of their products carries a label that’s unique—this becomes an easy way to absorb information for shoppers who are speeding through. 

61. Variant A: features a one line descriptor beneath the product image along with product name / Variant B: features just the product name

62. Test one variant with a recommendation section of lesser-priced accessories vs. another with a section on new releases

63. Variant A: total number of page results is mentioned at the top / variant B: the number of results are split by sub-category

Women’s Best features the number a shopper can expect in sub-category page results as well. 

64. Variant A: feature a “load more” button at the bottom of the page / Variant B: the “load more” button is flanked by how many results a shopper is seeing currently

65. Try one variant where the shopper can choose how many results to see per page vs. another without it

How much control you give to the shopper at the category page stage, can influence how deeply they explore your product pages—if it’s too many results, the overwhelm can make them easily drop off. 

66. Variant A:  “Build Your Bundle” appears as an option in the primary navigation / Variant B: a secondary CTA “Add to Bundle” is featured for every product

Pipcorn ends up using both ideas in combination, which means you could too! (Here's some bundling examples for inspiration.)

67. Test one variant with “buy more get $x off” labels vs. another with BOGO labels

Notice how Fashion Nova hypes up BOGO labels in their a/b test for eCommerce personalization:

On the other hand, Chubbies urges to buy more for an additional discount. 

68. Variant A: feature highlights (like free shipping) on the strip above primary navigation / Variant B: feature highlights in the section right above the footer

From working with 500+ clients across the world, we’ve noticed that the first idea works better for more immediate offerings such as flash deals, a sitewide discount etc. 

The second idea is more suited for values the brand stands by—for example, if the brand is sustainable and 100% natural, this information can come right above the footer in the form of icons. 

69. Test a variant where you feature “Bestsellers” on the top row vs. one where you feature “Limited Edition” ones

70. Variant A: show the shopper a “Quick Add” prompt before size selection / Variant B: show size and “Quick Add” feature prompts together

In the following examples, notice how Fashion Nova features a quick add and when a shopper clicks on it, the size options open up.

Gymshark, on the other hand, shows both as soon as a shopper hovers over the product image. 

71. Test a variant where “Flash Sale” products are featured alongside regular discounts vs. one where only regular discounts are promoted

Golde is one brand that clearly labels some of their products as being under “flash sale”—this induces urgency and can inspire shoppers to explore those products further. 

Ideas for Stage Three: Product Page

72. Test a variant with the product name highlighted vs. another where the payment methods are more prominent

Here’s how boldly Morphe features its product names as part of their product page test ideas:

In contrast we noticed how Huda Beauty also highlights the payment terms. 

73. Variant A: feature a short product description above the primary CTA / Variant B: feature key features

Here’s how it looks in real time—The Ordinary lays down the key features.

Dripkit describes the product in a short snippet in their product page a/b testing.

Quick tip: Which format you use will depend heavily on the nature of your target audience—if your products help them solve problems (and not just add to life, like say, coffee does), then technical specifications can become crucial. 

74. Test a variant with product description above the CTA vs. one with the description below the CTA

75. Variant A: ingredients are listed in a block of text / Variant B: a link is provided to a separate page 

FAE Beauty is a brand that goes in-depth in explaining its ingredients—hence they do so in a separate page for every product. 

This is the best approach if you want to state your ingredients as a way of being known as a transparent brand. 

76. Variant A: display product description in a horizontal format right above the CTA / Variant B: display product description in an expandable-collapsible vertical format under the first fold

77. Variant A: describe how & where the product is made / Variant B: describe the results from a real customer research

The similarity between the two instances is that both create a sense of authority that shoppers derive confidence from—though the second instance works more like social proof, especially when a brand sources the results from real-time research. 

Here’s an example from Golde's product page tesing ideas:

78. Variant A: place with the image gallery in the first fold followed by CTA beneath / Variant B: keep the image gallery to the left

79. Try a variant with a static hero product photo with a next button to display other images vs. another where the images are laid out in tiles

Skullcandy uses the first idea to make their image gallery come alive on the right side of the product page—a shopper can keep clicking the “next” button to access the associated pictures. 

Away, on the other hand, takes the more usual approach of keeping the image gallery to the left—and maintains thumbnail tiles for a quick overall view of the product. 

80. Try a variant with a 360° degree view vs. another where the all parts of the product are labeled within a single image

From experience, we know that for products that are fairly simple, you can use the second approach—while for those that feature more parts or technical aspects, a 360° view can generate deeper trust. 

The Candi Factory, which creates breathable underwear, always labels their products like this:

81. Test a variant where the primary visual is a video vs. another where the primary visual is an image

Primary visual = the image that’s on display when the shopper first lands up on your product page. 

82. Variant A: place a GIF in the photo gallery / Variant B: place a video right beneath the first fold

83. Variant A: your image gallery carries a zoom option / Variant B: the images are distinctly large (compared to the description section)

Luxy Hair, for example, makes use of bold visuals to do away with the need for a zoom option. 

84. Variant A: feature an unboxing video / Variant B: use a section to show what’s in the package

For several reasons, including needing to reduce page loading speed and marketing spends, businesses look for alternatives to an unboxing video. 

The reason we love Bond Touch is because they’ve figured a middle ground—they just use thumbnail images of the different parts within a package. 

85. Variant A: a primary CTA that first prompts to “select a size” & then changes to “add to cart” / Variant B: only a primary “add to cart” CTA

Doing what Crossrope does can mean more intuitive UX—because it might not occur to shoppers that they should select a size first.

So if they add to cart and then are told “please select a size”, it just adds to the overall friction. 

86. Test one variation with a secondary CTA featuring a popular payment method vs. another one with a wishlisting option

87. Variant A: features a secondary CTA that allows shoppers to ask a question / Variant B: features a secondary CTA to wishlist for later

Hyphen Sleep calls a shopper’s attention to reaching out for support through a secondary CTA instead of the usual wishlisting option. Many shoppers who’ve found their way into the product page are intent on buying, and for a high investment product, it’s likely they would want to ask some questions before they even wishlist it. 

88. Variant A: a secondary CTA features a special VIP price / Variant B: VIP membership benefits are cited above the primary CTA (as a link)

Quick Tip: A secondary CTA featuring a special VIP price makes sense only when you don’t have a ton of other messages or labels vying for the shopper’s attention. For example, if you’re running a special price on the product, avoid using a secondary CTA like this because it can become overwhelming. 

In the following example from Fabkids, it’s clear that the brand is promoting a special price for members, and the whole above-the-fold section reflects this. 

89. Variant A: feature color swatches over the primary CTA / Variant B: feature a separate section of available colors below the fold

While featuring color swatches is the usual practice, some brands like Tanner Goods choose to highlight a color in the main product section and feature the rest in a section below—if you’re trying to convey quality and differentiation in your products, this could be a more relevant approach. 

90. Try one variant with a size finder & another with a size chart featuring measurements

At the “desire” stage of the funnel, relevant engagement can inspire shoppers to actually make a purchase—a size finder or a size quiz can engage and personalize more than just a size chart. 

91. Variant A: the cart icon shows an updated number when a product is added / Variant B: the CTA click takes shoppers to a separate cart page

92. Test a variant with gift wrap displayed as a checkbox vs. another where a list shows the gift wrap options available 

Bonbonbon chooses to feature their gift wrap options as check boxes in their product page a/b testing:

93. Try one variation with a customer review callout section right below the first fold & another with the usual product description section 

Adore Beauty calls out their “most helpful positive” review just below the first fold—this can be helpful in creating greater confidence in the product. 

94. Variant A: highlight a customer review above the primary CTA / Variant B: highlight a badge of quality above the primary CTA

Here’s how Overstock highlights a badge of quality above the primary CTA. 

95. Variant A: feature a membership savings highlight above the primary CTA / Variant B: feature a gift card offer above the primary CTA 

Bed Bath & Beyond uses the second idea on its product pages. 

96. Test a variant with a social media wall with the descriptor “Follow Us” vs. another where the descriptor reads “See It In Real Life”

We’ve noticed that the tone you use in the copy for your social wall makes a difference—if you’re a fairly established brand with other sub-brands people follow, then “follow us” has the right ring to it.

However if you’re a brand that’s still small and trying to win confidence, “see it in real life” can make more sense. 

97. Variant A: feature free shipping/X-day returns as link-based line items / Variant B: feature a content block that offers a “guarantee”

Solo Stove calls out their free shipping and returns information through a “worry free guarantee” label—this can instantly draw more attention. 

FAE Beauty features free shipping & returns as a linked line item:

98. Try a variant where you upsell a related accessory vs. another where you offer a similar product recommendation

99. Variant A: highlight the background info about the product / Variant B:  highlight the brand differentiators 

Skullcandy creates context on how, where and why a product is made.

Lush, on the other hand, highlights their brand differentiators in their product page a/b testing.

100. Variant A: upsell through a CTA button / Variant B: make a related product recommendation

Notice how Tanner Goods ups the price through a CTA for those who choose to opt for a monogram?

101. Variant A: feature multiple variably priced product recommendations / Variant B: feature some recommendations with the same price as the main product on the page

102. Try a variant with all recommendations being products on sale vs. another with only one product on sale

Testing both will help you understand shopper behavior—and in turn inform whether you should test a third alternative with a half-and-half approach where, let's say, you put two recommended products under sale and two others under their original price. 

‍103. Variant A: feature “Write a Review” button / Variant B: feature a “Review & Earn” button

104. Variant A: offer a customization field on the product page itself / Variant B: feature a secondary customization CTA leading to a sub-page where the shopper can enter their details

Baublebar sticks with the first idea. 

105. Variant A: display a product comparison table featuring similar products & prices / Variant B: feature a comparison table with a few products carrying a discount

Notice how Sephora features one product in the comparison table with a significantly different price. 

106. Test a variant that has a sticky feedback button & another where a clickable link is displayed at the end of the page 

107. Test between a short product page variant & a longer one 

For either, you’ll have to determine which sections to keep and which ones can go. 

Here are a few successful ways in which we’ve shortened product pages for customers:

- Offered a link to the product video on Youtube (vs. dedicating a section of the page to the video)

- Created a separate page for product ingredients (so that a “read more” button can take them there after they’ve read the basics on the page)

- Fashioned the product description section like an FAQ

Ideas for Stage 4: Cart Page

108. Try one variant with a wishlisting option & another without one 

109. Variant A: feature a “free shipping threshold” message / Variant B: feature a “free shipping for loyalty members” message  

Lush makes use of the first idea—and this is how it looks:

110. Variant A: feature a free shipping threshold progress bar / Variant B: feature a store credit threshold progress bar 

Larq, for example, makes use of a store credit threshold progress bar.

111. Variant A: feature a store credit threshold progress bar / Variant B: feature a free gift threshold bar

112. Test one variant featuring a choice of free samples vs. another where an applicable discount code is mentioned

113. Variant A: feature “under $x” product recommendations / Variant B:  feature complementary products

114. Test a variant which shows one product recommendation at a time vs. another that lists all the recommendations in a vertical/horizontal format

Golde ensures a shopper has access to multiple product recommendations—but can focus on each one thanks to previous-next arrow buttons. 

115. Variant A: call your recommendation section “Recommended For You” / Variant B: call your recommendation section “Things We Know You’ll Love”

While the first puts the brand in a place of authority, the second focuses on shopper preference—and there can be a marked difference of conversions between the two. 

116. Variant A: a promo code is automatically applied / Variant B: an open field is provided along with a list of “featured offers” 

117. Test one variant with the standard “apply discount code” field vs. another that features a special applicable discount 

Death Wish Coffee, for example, reserves a special discount for military first responders, government employees and teachers. 

118. Variant A: feature “free shipping & expedited shipping” callouts / Variant B: feature “delivery & pickup” callouts

119. Test a variant where you highlight “sale savings” highlighting the price & another where you just feature the old & new prices

120. Variant A: feature a button that says “checkout” / Variant B: feature a button that says “secure checkout”

Huel goes with the second variation and this is how it looks—get a picture of how you’d feel if you were going towards checkout from this page?

121. Variant A: features a checkout button as primary CTA / Variant B: also features “Continue Shopping” as a secondary button

122. Variant A: feature a secondary checkout button of a popular payment method / Variant B: feature popular payment method logos

Brilliant, an eCommerce bicycle brand, features the regular checkout button but also displays two popular payment methods as express checkout. 

123. Variant A: free product callout shows up as pop-up / Variant B: free gift is highlighted within the cart page

Solo Stove features a separate pop-up when the shopper becomes eligible for a free gift. 

124. Try a variant with an email signup discount callout vs. another with a membership rewards callout

125. Variant A: feature trusted payment methods with a “100% secure payment” callout / Variant B: feature trusted payment methods without the callout

Chubbies sticks with the first variant in their a/b testing for eCommerce:

126. Try a variant where you feature an FAQ link vs. another where you feature a shopping assistance highlight (with support number)

127. Variant A: allow shoppers to calculate shipping costs / Variant B:  mention shipping will be calculated during checkout

Spicy Lingerie enables shoppers to calculate shipping costs at the cart page stage itself. 

Ideas for Stage 5: Checkout

128. Variant A: the main checkout sections are shown in a single page in expandable sections / Variant B: a progress bar is highlighted above

129. Try a variant where you use a thumbnail product image vs. another where you use only product name & cost

130. Variant A: ask shoppers to create an account / Variant B: offer them a X points as reward for creating the account

Chubbies understands account creation is hard work—hence they declare a reward. 

131. Variant A: offer account creation points / Variant B: additionally feature icons to display other benefits 

Sephora chooses to highlight the primary benefits of creating an account—apart from calling out collected points. 

132. Test a variant where you offer only guest checkout option vs. another where you also feature express checkout

133. Variant A: feature guest checkout along with “create an account” / Variant B: feature social sign-in plus “create an account”

After all, 65% of shoppers agree that they will return to a site that automatically lets them in through social login—plus social login allows you to record essential customer information for future use—here’s a piece we wrote to help you understand this phenomenon better. 

134. Variant A: highlight the membership program / Variant B: feature the in-house credit card program with immediate savings 

135. Variant A: feature multiple known payment methods / Variant B: offer points/rewards on a relatively lesser known payment method

Here’s what Chubbies does: 

136. Variant A: feature payment method logos as trust badges / Variant B: feature brand promises as links

Mejuri works with the second idea in their one-page checkout.

137. Try a variant where you feature email alerts vs. another where you feature SMS alerts 

138. Test one variant with a live chat feature vs. another with a link to a feedback page

139. Test one variant with only live chat feature vs. another where you also mention the link to the customer service page

140. Test one variant where you ask for email before you show the checkout page vs. show checkout page directly

H&M uses the first method—given that average cart abandonment stands at 69.99%, this can come in handy to not lose out on customers buying without account creation.

Bonus: Ideas to test over Email

141. Variant A: use the receiver’s name in the email subject line / Variant B: use a warm phrase like “hello there” but no name

142. Try a variant using just your business name vs. another where “<insert name> from <business name>” is mentioned

143. Variant A: declare the discount as the first word in the subject line / Variant B: feature the discount in the second half of the subject line

144. Variant A: use the power word “guarantee” in the subject line / Variant B: feature the power word “promise” in the subject line

145. Variant A: Make the subject line about a freebie / Variant B: Make the subject line about a discount 

146. Test one variant where you offer a discount as a welcome gift vs. another where you feature multiple advantages of signing up

Ann Taylor clearly works with the first idea. 

Anthro, on the other hand, leverages the second one. 

147. Variant A: keep the focus of the shopper on immediate gains / Variant B: draw the shopper’s attention to long-term benefits

For example, for A, you could talk about an immediate discount, BOGO offer or even an exclusive price. 

For B: you could take the shopper’s attention to loyalty program discounts, what kind of gated content they’ll have access to etc. 

148. Variant A: keep the shopper at the center of the communication / Variant B: keep the offer at the center of the communication 

In the following example, Birchbox makes the shopper the focus—this is the kind of email personalization every shopper secretly looks forward to (while all of us also know how the “50% off” discount mails look like.)

149. Variant A: feature a blown up image of a product range in the email body / Variant B: feature thumbnail images of the products and product names in the email body 

150. Variant A: use a single CTA in the email body / Variant B: use multiple CTAs in the email body

Quick Tip: Ideally, a single CTA boldly draws a shopper’s attention, however there are times multiple CTAs might work better:

- if you have multiple categories of products (even then mention no more than two)

- if you’re trying to draw the shopper’s attention to multiple resources on your site (like blogs, tools or videos—in this case, mention no more than three)

151. Test a variant with the usual “shop now” as the primary email CTA vs. another that says “get $x off today”

152. Try a variant where you bring in urgency in the headline vs. another where you feature urgency in the CTA

153. Variant A: feature just a headline in the email body along with a relevant CTA / Variant B: feature microcopy along with the headline

Check this out: Email A/B Testing: Elements to Test + Mistakes to Avoid

How to analyze your A/B tests

1. Track the right metrics

When you're trying to make sense of your A/B tests, it's important that you track the right metrics—here are a few we've found to be super relevant:

- Bounce Rate (the % of visitors who hop off your storefront after viewing just one page)

- Scroll Depth (how far your shoppers scroll once they land on your site)

- Abandonment Rate (the % of tasks shoppers open but don't conclude—for example, adding to cart but not finishing the purchase)

- Conversion Rate (the % of shoppers who complete a desired action—it can further be divided into micro-conversions and macro conversions)

2. Divide your audience into segments

You can factor in various critical elements to do the segmentation—these factors include demographic, motivational, behavioral, attitudinal and preferential amongst others. Factoring in segments can help you test more deeply and analyze in a more granular way.

3. Monitor behavior analysis tools closely

There's a reason why behavior analysis tools such as heatmaps and visitor recordings help in breaking down results from A/B tests—for one, they can help you notice which pages on your site are getting more traction. They're also instrumental in indicating what's happening to visitor behavior across the conversion funnel.

Before you go

A/B testing or not, 98% of visitors who visit an eCommerce site still drop off without buying anything.

Why: user experience issues that cause friction for visitors.

And this is the problem Convertcart solves.

We've helped 500+ eCommerce stores (in the US) improve user experience—and 2X their conversions.

How we can help you:

Our conversion experts can audit your site—identify UX issues, and suggest changes to improve conversions.