14 A/B Testing Ideas For eCommerce Emails

It’s not uncommon for experts to tout the 42X ROI effectiveness of emails. 

But the truth is not all emails are that effective—and that you need to go harder in assessing your email performance more closely before you arrive at brilliant results. 

One of the starting points is A/B testing, an approach that helps finetune your email marketing strategies while helping you build a better connect with audiences. 

And that’s why in this piece, we’ll cover the most crucial variants you can test + which contexts work really well for each of them.

You'll also find food for thought on: How to Sharpen Your Email Marketing A/B Testing Process

Plus, we’ve answered some FAQ eCommerce business owners approach us with. 

(To run more targeted A/B tests on your eCommerce website, read this.)

14 Elements to A/B Test In Your eCommerce Email Marketing Strategy

1. Sender Name

Though often overlooked in the most conscious ways, the “sender name” is an element recipients intuitively consider before deciding to open an email. 

So if it somehow sounds fake or exaggerated, it’s likely that a recipient will trust your eCommerce brand—unless of course they’ve already been a customer (and have had a great experience!)

Variants to test:

i. A: Brand name / B: Individual name + brand name

✓ Testing the brand name is a good idea if you’ve been around for a while as a brand - it also works well when you have to build a connection with a newer audience. Test this especially on promotional emails, product launch emails as well as brand newsletters. 

✓ Using individual name + brand name can be a great start to email campaigns for newer brands - it also works exceedingly well with loyal audiences, who may want a single point of contact at the brand. Test this for post-purchase emails and welcome emails. 

ii. A: Brand name / B: Position @ brand name

✓ Using position@brand name can be especially effective if you’re sending support or post-purchase customer service emails—it also works if you’ve just launched an email newsletter and want to drive authority for those subscribing to it. 

iii. A: Department @ brand name  / B:  Individual name + position @ brand name 

✓ Department @ brand name is a valuable variant when you’re trying to send post-purchase support emails or even responding to specific feedback. 

✓ Individual name + position @ brand name may be most ideal for loyalty members and the highest spenders - this is also good tact for brands that promote luxury & exclusivity. 

iv. A: Department @ brand name  / B: Individual name + department @ brand name

✓ Using Individual name + department @ brand name may be more effective if you’re handling customer queries around product delays or mismatches, taking up reassurance & personalization a notch higher—this also works wonders for brands that use exclusivity as a positioning statement. 

🚀 Pro Tip: 

Take the sender name game to a different level during peak holiday season—if you’d have used just the brand name for the rest of the year, try using Brand Name + Gift Guide or Brand Name + Exclusive Deals to improve your open rates. 

Further Reading: How often should eCommerce stores send marketing emails?

2. Subject Line 

Your email subject lines are quite literally the primers that recipients consider before wanting (or not) to open your email and go through the rest of it. 

And that’s why what you test and what emerges as the winning variant has everything to do with your conversion rate success. 

Variants to test:

i. A: Use the recipient’s name / B: Use a generic statement without the name

✓ Using the recipient’s name naturally improves personalization - but you may kill that sense if you do it across subject lines for every email type. Test this specifically for welcome emails, milestone offers and referral emails to avoid sounding like you’re trying too hard. 

ii. A: Make a generic statement about a recommendation (like “You’d love <insert the product name> because…) / B: Describe key attributes like discount & quality with the recommendation 

✓ In order to use a generic statement without the name, focus on making it targeted to the season, preference or event that you’re using as a leverage—it’s exactly why we love Birkenstock emails, they don’t beat around the bush:

✓ If you’re trying to describe key attributes with the recommendation, make sure it doesn’t drag and stays punchy—this will help recipients get an instant idea about what the rest of the email is about and what they can look forward to. This works especially well for seasonal emails, winback emails and cart & browse abandonment emails. 

iii. A: Incite curiosity with a question / B: Drive authority with a statement

✓ When you’re testing a variant to incite curiosity with a question, remember that it can work for a few conditions more than others: 1) when you’re trying to build awareness 2) when you’re trying to inspire opens from a segment of high spenders 3) when you’re trying to re-engage someone that’s not bought from you for a while.

✓ To drive authority with a statement in your email a/b testing, it’s ideal to target repeat buyers without making simplistic statements like “We’re the best in the…” Instead consider talk about switching up your shipping process to a faster one or perhaps giving the best loyalty perks in the market. 

iv. A: Show an offer for the next purchase / B: Feature care instructions for the previous purchase

✓ Showing an offer for the next purchase can be a highly effective variant when you’re trying to target one-time shoppers—this can nudge people towards the second purchase.

✓ If you’re testing care instructions for the previous purchase, ensure it’s on those who’ve either bought high ticket or high maintenance items—this may work less on those who buy more affordable products. 

🚀 Pro Tip:  

Use numbers in your subject lines to drive more confidence—the number could reflect how many reviews you’ve received, how many items have sold or even numbers from customer feedback.

Further Reading: Browse Abandonment Emails: Epic Examples, Tips, & Subject Lines

3. Preview Text

The third in the lineup of what creates the first impression of an email is the preview text. 

In fact, the more complementary it is to your subject line, the more you’ll be able to create context and drive email opens. 

Variants to test:

i. A: Feature an exact limited time window (like “Ends in 3 days”) / B: Feature a generic urgency message (like “Hurry! Selling fast!)

✓ If you decide to feature an exact limited time window, you may want to specifically work with flash deal emails and those communicating about limited drops with discounted email-only rates. 

✓ On the other hand, featuring a generic urgency message may work better for weekly promotional emails or sitewide discount emails that you’re sending to repeat buyers (that extra hint of urgency in the first instance can drive more conversions from browse abandoners and ToFu subscribers.)

ii. A: Address the recipient by name / B: Address the recipient by a generic phrase (like “Loyal One” or “Comeback Queen”)

✓ Though here we’ve recommended testing the recipient name against a description, you’re free to combine these aspects in a third variant: where the recipient’s name features in the subject line and the description (like “Loyal One”) features in the preview text.

iii. A: Maintain a text length of 40 characters / B: Maintain a text length of 80 characters

✓ If you intend to keep your preview text short, know that the ideal scenario is limited time sales emails, or in cart abandonment emails where you’re highlighting a free shipping offer. For longer preview text, product bundling emails and newsletter emails carrying multiple brand highlights could carry greater impact than other email types. 

iv. A: Create a sense of mystery in the tone / B: Create a sense of urgency in the tone

‍

🚀 Pro Tip:  

Test preview text variations by device—this can be a valuable test to run around the year, because on mobile, potential shoppers scan faster, which means if the preview text is long, they may not open the email at all!

Further Reading: eCommerce email marketing: The complete guide

4. Preheader

What microcopy does after a powerful headline in your hero header, is what the preheader does in your eCommerce emails. 

The best written ones finish the thought or question the subject line has initiated, and leaves the recipient wanting to open the email. 

Variants to test:

i. A: Make a direct offer (like “Get your 15% discount coupon today”)  / B: Make an indirect offer (like “Exclusive mystery offer waiting to be opened”)

✓ If you’re making a direct offer, know that this works more effectively on repeat / loyal customers through transactional emails. Similarly, this approach is also ideal if you’re sending a product drop email or a bestseller email. 

✓ On the other hand, an indirect offer may be more ideal for subscribers who’ve just entered your list in a bid to understand your brand, products and offers better. Use this to nudge those who haven’t made a purchase yet. 

ii. A: Put a limited time window to the offer / B: Announce the terms for an additional discount 

✓ Featuring a limited time offer can be the way to go if you have an ongoing sitewide discount—creating a separate segment with those who’ve impulsively shopped before, may make this even more effective.

✓ When you’re declaring the terms for an additional discount, it’s helpful to let people know on what this would be applicable—for example, you could retain the discount % for every segment but make the applicable category dynamic based on past purchases. 

iii. A: “Back in stock” alert / B: “Soon out of stock” alert

✓ Preheader text that says “back in stock” largely works on those who’ve displayed interest in a product / product line - either by browsing multiple times and not buying it or wishlisting it. 

✓ A “soon out of stock” preheader, on the other hand, typically works better with those who’re highly engaged with a brand, and may have even made multiple purchases before (some of them even on an impulse!)

iv. A: Keep it short (up to 30 characters) / B: Make it longer (up to 60 characters)

✓ If you’re picking a longer preheader, be very clear why you’re doing it and that’s because if the reason isn’t to target, you’re risking chances of lower email opens on mobile. Typically, lengthy preheaders work better for educational emails as well as tiered discount emails. 

✓ While using a shorter preheader, know that the ideal use cases are mobile-first emails (especially if you’re highlighting an email-only offer in the subject line) as well as single product promotion emails. 

Take a look at this Aurate preheader—it’s actually just 17 characters long:

In contrast, when we read one by Res Biotic - “95% of customers recommend res to their love…” - we found it’s 47 characters long. 

🚀 Pro Tip:  

Consider using emojis or a CTA, provided that you’ve analyzed the segment you’re using this on.

It’s a fact that 24% of email subscribers will read preheader text before deciding if the email open would be valuable.

5. Send Time

This is perhaps one of the most suitable items to test across your eCommerce email campaigns. 

And that’s because there’s no perfect send time across industries and even geographies. 

Variants to test:

i. A: Send on a Tuesday / B: Send on a Thursday

✓ You’ll have to be clear about the messaging you feature depending on whether you’re testing your email sends between Tuesdays and Thursdays. Getting a sense of where your segments are psychologically will help you craft better messages (for example, Tuesday emails do better when they’re about planning for the weekend ahead—on the other hand, last chance emails and limited X-hour offer emails tend to do better on Thursdays when potential shoppers have more time on their hands to actually buy.)

ii. A: Send in the morning hours / B: Send during post-work hours 

✓ While picking send time testing between morning hours and post-work hours, it’s ideal to go back to past engagement & purchase behavior of the segment you’re testing this on. As a norm, send emails that need quicker action during morning hours and keep high consideration messaging like bestseller emails with social proof for later in the day. 

iii. A: Choose an earlier date for recent purchasers / B: Choose a later date for those who haven’t bought for a while 

✓ To choose between an earlier date for recent purchasers and later date for idle shoppers, consider tweaks in messaging even if the core communication remains the same—for example, the one scheduled for an earlier date could have dynamic content based on who purchased what recently, whereas the later email can timebox an exclusive discount positioned as a re-engagement email.

iv. A: Choose a working day of the week to send / B: Choose a day in the weekend to send

✓ When you’re considering running an email A/B test between a working day and the weekend, consider what competitor brands are doing—especially look at the ones that are converting better through email and the kinds of offers / messaging they’re leveraging.

🚀 Pro Tip:  

Before assessing different send times for your larger audience, restrict send time A/B tests to a single segment to perfect the process: look at sample size and zero in on the key metrics to track success. 

6. Body Copy

This is really the final stage of conviction wedged between recipients deciding to open an email and clicking through that CTA. 

And to do the latter, the body copy often counts almost as much as the offer or privilege you offer as part of the email. 

Variants to test:

i. A: Drive urgency / B: Drive aspiration & desire

✓ When you decide to test between urgency and aspiration, revisit what kind of product / offer you’re pushing through that email campaign. Typically, urgency works better with limited time offers on bestsellers while aspiration may be more ideal if you’re relaunching an old classic in limited pieces. You could even test a third variant that combines the two—this approach works on products with an exclusive or luxury edge. 

ii. A: Talk about features / specifications / ingredients & create rational appeal/ B: Talk about benefits & effects & create emotional appeal

✓ Whether you bring to light body copy with emotional appeal or rational appeal, make sure your core messaging remains the same, but create tweaks based on the brand’s values when you’re triggering the emotional side and focusing on the product’s make (label the image if you like) for more rational appeal. 

iii. A: Feature one section of copy / B: Feature multiple sections of copy

✓ What wins between a single section of copy versus multiple sections of copy depends largely on what the target segment is looking for in the email. If it’s trying to play up a single offer, featuring multiple sections of copy can cause cognitive overload. On the other hand, if it’s an educational email that is nevertheless recommending personalized products, multiple sections of copy will be better able to bring out key differentiators. 

🚀 Pro Tip:  

Run your body copy email A/B test across several campaigns over a longer duration to assess both short-term metrics like click-through rates as well as long-term metrics like repeat purchase rates.

7. Copywriting Tone

Whether you choose to use humor, exaggeration or quirk, research seems to have arrived at the verdict: the more the copywriting tone in an eCommerce email sounds personalized, the more will it result in opens and click-throughs. 

But even when the tone is personal, what aspects are you to really test and assess?

Variants to test:

i. A: Choose a lightly personalized tone (like “We thought you might like…”) / B: Choose a heavily personalized tone (“Just like what you love!”)

✓ Testing between a lightly personalized tone and heavily personalized tone would need you to go back to the basics again. A hyper personalized tone is more up the alley of a repeat customer, perhaps those who’re a part of your membership program. A more lightly personalized tone may work with those who’re just warming up to your brand and may still have some way to go before they become a repeat buyer. 

ii. A: Use a warm, inclusive undertone (“We got you afternoon groceries @ great prices”) / B: Use a neutral undertone (“$90 off on afternoon groceries”)

✓ Whether or not a warm & inclusive tone wins or a neutral one does will not just depend on the kind of segment you’re targeting the campaign at, but also what your overarching brand voice is. The former may more likely be the winner if your customers depend on you creating repeated assurance for them. 

iii. A: Speak in first person (“We’ve got what it takes to keep you fit”)  / B: Speak in second person (“Your health deserves only the best”)

✓ First person “we” usage in email copywriting is great when you’re first trying to establish a relationship with a subscriber—so, it’s ideal in the welcome series or when you’re sending re-engagement emails and elaborating on brand benefits. 

✓ On the other hand, “you” usage is more customer-centric and makes more sense when recommending products or highlighting exclusive savings. 

🚀 Pro Tip:  

Randomly place each new subscriber under a “tone profile” once you’ve gathered enough data on which tones seem to work reasonably well—look at their responses over the first few emails to see what pattern emerges, and this will help you better personalize your email tone for each of those customers. 

Further Reading: eCommerce Email Copywriting: 20 Game-Changing Tips

8. CTA

It’s now a well-known fact that 90% of shoppers who read headlines also read CTAs. 

The good thing is, you can leverage it too, especially when it comes to your email CTAs. 

Variants to test:

i. A: Place the CTA above the fold / B: Place the CTA after the first scroll

✓ While above the fold has naturally become the preferred CTA placement in eCommerce emails, placing the CTA after the first scroll does serve some contexts more: where you’re promoting a bundle and each product in it will need significant explanation or where you’re explaining the steps in your rewards program.

ii. A: Feature just one CTA in the whole email / B: Feature multiple CTAs

✓ If your emails carry dynamic content suited to each segment, it becomes easier to feature a CTA based on what will appeal to a particular segment. However, if that’s not the case, you can create emails where you choose up to 3 CTAs and each one corresponds to a specific segment to engage with & buy through. 

iii. A: Feature only the conventional “Buy Now” CTA / B: Vary the CTA text

✓ Many businesses confuse picking varied CTA text to linking them to different pages. This isn’t necessarily what you should do. For example, you may want to create a narrative flow in your email about a bestseller and link all the CTAs to the same landing page but depending on what the micro text is saying, the CTAs copy can change to contribute to the narrative flow. 

iv. A: Use an on-brand CTA color / B: Use a striking & contrasting CTA color

✓ When your primary intent is to establish trust or offer non-transactional information around material sourcing or supporting a cause, an on-brand CTA color works towards creating more recall. However, for new product launches, limited time product drops or even brand collabs, a contrasting CTA color works best. 

🚀 Pro Tip:  

Don’t test all the mentioned CTA variables here all at once—if you’re intent on getting quicker data, consider running a test with each variable across different segments. 

9. Visuals 

Email campaigns that feature images receive 42% more click-throughs. 

Having said that, those that carry three or fewer images seem to do the best. 

So, it is indeed one of those factors that need prioritization in your email A/B testing process. 

Variants to test:

i. A: Show people-led images / B: Show product-led images

✓  Whether your people-led visual email or product-led visual email becomes the winning variant under certain circumstances, will have to do with where the shopper is in the customer journey. We’ve noticed newer subscribers like to see a combination of both as they become more acquainted with a brand. 

ii. A: Use a single hero image/ B: Feature a text-only email

✓  If you’re apprehensive about the clickability of a non-visual email, consider testing your text-only email with different gradients or sections to create visual layering—that’s what Warby Parker does in this email on brand benefits where each text line lights up in an auto-play gif mode:

iii. A: Display a grid of only product images / B: Display a grid that alternates between images & text

✓ While running the product-only image grid test alongside the text alternating image grid one, see to it that in either case you don’t go over 3 to 4 rows in the grid, as anything more can quickly become cognitive load territory. 

🚀 Pro Tip:  

Whenever you’re using images in your emails, ensure you match the image to the action you want the subscriber to take—for example, if it’s about using a new collection, show people-led photos with the products in use, but if it’s a limited time off, show product photos with well-designed discount labels. 

You should also read: Top eCommerce Email Marketing Trends For 2025

10. Offers

Since 61% of subscribers would typically like to receive promotional discounts every week, it’s your chance to get your email A/B testing right in this regard. 

Variants to test:

i. A: Feature a tiered discount / B: Feature a BOGO offer

✓ When testing a tiered discount alongside a BOGO offer, be on top of your business goals. This is because while BOGO leads to more units being sold, tiered discounts end up having a better impact on average order value. 

ii. A: Feature a discount offer / B: Feature a free shipping offer

✓  To test a discount offer against a free shipping offer, consider seasonality very seriously. While during peak season, a free shipping offer will likely have more sheen, in the leaner periods, shoppers are happier to settle with a discount on a product they really like. 

iii. A: Display an upgrade at a discount / B: Display a bundle at a discount 

✓ No matter which turns out to be the winner between a discounted upgrade and a discounted bundle, the success really depends on the products you make these offers on. While upgrade discounts are best announced on newer products, bundled discounts may fare better on end-of-life products. 

iv. A: First time discount on a single product / B: First time discount on a subscription order

✓ When testing a first time discount on a single product, remember that it works better on high-ticket items that have a longer lifecycle. On the other hand, a first time discount on a subscription order is ideal for regular use items that will also show predictable purchase patterns. 

🚀 Pro Tip:  

Also test secondary driving points like limited time against limited quantity to see what emerges as the winner. 

Further Reading: 30 Amazing eCommerce Email Templates (from 6 industries)

11. Personalization 

There’s a whole lot that an eCommerce brand can personalize in the emails they send to customers across the funnel. 

But what should you really test to ensure your emails aren’t just opened but become a source of revenue?

Variants to test:

i. A: Show a dynamic product description (based on browsing behavior) / B: Show a snippet from the product page description

✓ Enough data is required to show and test dynamic product descriptions against generic descriptions. Naturally though, some businesses do better with dynamic product descriptions than others, including those in the nutrition, F&B and skincare spaces. 

ii. A: Personalize the subject line based on past behavior (“Last year it was <enter product name the recipient bought>,this year it’s *surprise*”) / B: Feature a generic, non-defined subject line (“Because you love surprises”)

✓ When testing a personalized subject line against a generic subject line, change only the personalization element (it could be the recipient’s name, a product they added to cart but didn’t buy etc.) while maintaining all other variables

iii. A: Show a personalized gifting guide based on past behavior / B: Show a generic gifting guide collection that features on your website

✓ When testing a personalized gift guide alongside a generic gift guide, be conscious of the number of products you feature in each. Typically, generic gift guides will show more products, which can potentially get more clicks, leading to a skew. Maintain the same number of products in both to avoid this. 

🚀 Pro Tip:  

Watch out for outdated data informing your testing decisions. One way out of this is to stay away from over-personalization, which also makes testing across more segments with less preparedness possible. 

Further Reading: 20 email personalization templates (examples from great brands)

12. Recommendations

What you put in an email influences the purchase decisions of at least 6 out of 10 subscribers. 

So, when it comes to recommendations, it’s ideal that you double down on running the most important tests. 

Variants to test:

i. A: Show a random set of products based on recent browsing history / B: Show a “Continue from where you left off…” nudge & feature products browsed only in the last visit

✓ Timeboxing what you mean by “recent browsing history” could be relevant when you’re testing one variant with products based on this. For last visit recommendations, you could also try showing bundles that include the product they last viewed, and throw in an attractive price for better outcomes. 

ii. A: Show up to 4 recommendations each with a short descriptive paragraph / B: Show up to 8 recommendations with product details & catchy image labels

✓ When testing 4 recommendations, remember that this may be more apt for high-consideration items or products that have greater specs and nuances. On the other hand, 8 recommendations may be better for more visual-led products and categories such as fashion, decor etc. 

iii. A: Show recommendations only from the recipient’s top favored category / B: Show few recommendations from top favored category and also newer categories 

✓ To show recommendations from the top favorite category against recommendations that feature a category mix, it’s good to remember that the first approach may be effective if your products have a shorter life cycle

iv. A: Show recommendations that match the price of the last purchase (within a 20% range) / B: Show recommendations that have a price mix range

✓ When testing recommendations with a price match against recommendations that come with a price mix, remember that subscribers are likely to have different price expectations from different categories (for example, someone may love a $100 perfume but won’t even consider $100 socks)

🚀 Pro Tip:  

Recommendation cannibalization is a real occurrence across eCommerce emails, where shoppers find themselves buying recommended products but not picking other products they likely would have. To avoid this, you may want to bring in a third “hidden” variant that features “no recommendations” for a control group—this can be additional data for you to optimize your email recommendations in the future. 

Further Reading: 24 Email Automation Examples For More Conversions (eCommerce)

13. Design 

It’s not uncommon to hear about how a clean, clutter-free layout is ideal for email conversions. 

But the truth also is that there are several other nuances that dictate user behavior—some of which your email A/B testing process can tweak and enhance. 

Variants to test:

i. A: Show an image-first layout / B: Show a text-first layout

✓ If you’re going to be testing an image-first layout approach, ensure you use an optimized image that loads faster than the user scrolls. Similarly, for a text-first layout, it’s ideal to stick to a concise headline that’s super easy to understand at one go. 

In some of their emails Farfetch combines the above two:

ii. A: Feature categories at the top like traditional navigation / B: Feature categories but as image tiles

✓ While testing categories as nav links and categories as tiles, you may have to weigh out certain considerations. These include mobile optimization and how tiles may actually not be easy to click unless designed in a mobile-first manner. Then again, nav links can push down content further and the email can need more scrolls to go through. 

iii. A: Show a hero image layout / B: Show a split grid hero layout

✓ While testing a hero image layout and a split grid layout, prioritize on the quality of message focus. That’s because in both, this section will either convey the lifestyle, the benefits or the features. 

🚀 Pro Tip:  

Focus on testing the most crucial design elements like grids, CTAs and information hierarchy before you move to secondary elements like background color, font etc. 

Further Reading: eCommerce Email Design: 25 Beautiful Examples (& Why They Drive Sales)

14. Social Proof

Since emails land up straight in our inboxes, they’re anyway more personal than, let’s say, a blog we read on the Internet. 

Exactly why you ought to be using them more and more for conveying social proof, which in turn generates customer trust. 

Variants to test:

i. A: Show up to 3 customer review snippets / B: Show up to 3 customer ratings flanking separate recommendations

✓ While testing review snippets alongside customer ratings, it may be worthwhile to remember that reviews work better on ToFu and MoFu customers while ratings seem to be a better snapshot for BoFu shoppers

ii. A: Feature a pic of the founder and a note from them / B: Feature a pic of an accomplished long-term customer and a note from them

✓ Featuring the founder and a note from them can be ideal for product launch emails as well as checking in with idle customers from time to time. As for featuring a loyal customer, promotional emails for new subscribers and referral emails may be cut out better. 

iii. A: Use a real-time counter (like “498 people bought this in the last 24 hours”) / B: Use a static metric (“Loved by 10, 000 + customers”)

✓ When you’re considering using a real-time counter as a variant, be careful to not apply it during seasonal, inconsistent or low sale periods as this will return unsatisfactory results. Similarly, to use a static metric, it becomes essential to use it for long-term campaigns featuring bestsellers or offers that you run at predictable times of the year. 

🚀 Pro Tip:  

Always align social proof with where a particular segment is in terms of customer journey. For example, if you test customer rating highlights in emails for new subscribers, you may find lesser engagement and this would be a skew. 

How to Sharpen Your Email Marketing A/B Testing Process

1. Test only one element at a time

Though businesses often run from this best practice, in reality, it serves to show up cleaner, clearer insights for you to improve your eCommerce emails. 

Testing multiple elements can be misleading in terms of the data you collect about what worked and what didn’t. 

However, there could be instances where you could A/B test more than one email element:

- If you have limited time or audience: a smaller list size or limited opportunities to run tests across campaigns can make it necessary for you to test multiple elements at one go—be prepared for less precise results though

- If your platform allows multivariate testing: you can explore testing elements in combination to see how they affect each other, though a disadvantage is that you can’t do this with a really small list size

- If you’re rolling out a brand-level redesign: in this case, focusing on elements that drive the maximum engagement or action may be necessary

Further Reading: 17 Proven Ideas to Improve Email Marketing Conversion Rate (+ Examples)2. Test to assess multiple crucial metrics

When you’re running email A/B tests, it’s crucial that you run them to figure how subscribers across the funnel are behaving. 

A part of this then becomes about testing to find out how your email metrics are performing. 

At Convertcart, the metrics we suggest clients assess more closely are:

🔍 Open & Click-through rates: These are primarily engagement metrics that need tracking across the lifecycle of a customer, no matter where they are in the conversion funnel

🔍 Conversion rate: This metric identifies how well your emails are targeting intended segments and what % of those are taking the action you want them to take

🔍 Revenue per email: This is a metric that lets you see which email campaigns are converting the most, so that you can apply more rigorous A/ B testing to these

🔍 Average order value: Since a higher AOV typically means more revenue per customer, this is a long-term metric that can help you decide which campaigns are most effective and what elemental tweaks are nudging people towards buying

🔍 Return on Investment: Since the ROI accounts for how ever $ spent on an email campaign is performing, it’s a crucial metric to decide budget allocation and amplify A/B testing efforts in the right places

🔍 Time to purchase: This metric shows how long it takes for a recipient to buy after they’ve received an email, making it relevant to assessing several email A/B testing elements like send time, personalization and social proof. 

3. Segment to arrive at better insights

It’s a no-brainer for every eCommerce business to want to come to an email variant that works perfectly on a customer segment. 

But for that, the first step is segmentation in the following ways:

✓ With purchase behavior: Several competing segments including high-value customers vs. idle subscribers and first-time buyers vs. repeat buyers can emerge from this segmentation exercise

✓ With levels of engagement: Look at who frequently opens and clicks on your emails (setting a timeframe may again be important), who was formerly active but not anymore (let’s say 60 days of inaction) etc. 

✓ With location or demographics: Look at which categories perform better in which locations and across which genders, or if some locations actively buy only when there’s an offer to redeem

✓ With browsing history: Look at which brands or categories are preferred by certain segments and similarly, if certain segments respond when they receive affinity-based recommendations

✓ With the customer’s lifecycle: Are you targeting someone who’s just entered your email list? Or is it someone who didn’t buy for a long time and then bought all of a sudden? Are they repeated cart abandoners?

Further Reading: 28 No-BS Ways To Get More Email Subscribers in eCommerce

4. Create a continuous cycle of testing

While this sounds daunting to a lot of businesses, it’s the reality that will help you stay on top of conversions. Here’s a short structure we suggest you follow to do this better:

✓ Create a what, how long, when framework: What to test, how long to test for and when to test in order of importance to your primary metrics

✓ Automate as much as possible: This allows you to focus on other business problems while email promotional campaigns continue to run with A/B variations at designated intervals

✓ Analyze and iterate on wins: No going back to scratch unless your tests throw up skews, because when you test the right variants you can find a winner and then iterate on that, using it as an “improved baseline” for the next test

eCommerce Leaders Also Ask:

How complex is it to perform email A/B testing?

If you consider that A/B testing is an iterative process anyway, whether you do it on a web page or an email, you’ll see why email A/B testing is actually simple. And it gets simpler as you do more of it. 

The first thing you need is a tool or platform that’ll help you get started. 

The next thing would be to clearly lay out the metrics and goals you’d want to study. 

Finally, know that it’s ideal to always test a single variable at a given time to reduce skews. 

How do I arrive at statistical significance if I run a small eCommerce business?

Small eCommerce businesses often face the challenge of having smaller lists—if you’re here reading and resonating with this, here are a few ways you can attain statistical significance:

✓ Work with sequential testing: This lets smaller brands choose batch sizes that are much smaller (say up to 1000) and test incrementally until the sample size is large enough. Since testing on segments reveals trends fairly early, sequential testing doesn’t make it necessary for smaller businesses to wait for certainty before applying email optimization tactics. 

✓ Work with a lower confidence level: While 95% is said to be of optimal statistical significance, businesses with smaller lists can bring this down to 90%. Though this increases the risk of error, it nevertheless allows insights to emerge even with smaller list sizes. 

✓ Focus on trends: While a larger list size will offer more definitive results, smaller list sizes show up propensities. For example, if a variant is performing 30% better for two weeks straight, it may be valuable to consider running it as the winner. 

Transform Email Marketing Into A Revenue Machine

Most eCommerce store owners don’t see email as a serious revenue stream.

Ask them about the importance of email marketing, and you'll hear: “we don’t really have a major strategy,” “we mostly use generic templates,” or “we just send emails to people on our list.”

BUT AT THE SAME TIME:

There are stores out there that drive 30%+ of their revenue from email marketing.

Engage can help you do the same - Book a free demo.

We’ll show you:

• workflows we can create for your store,
• proven ways to drive 30% or more $$ from email alone, and
• successful templates and strategies from your industry (and others).