Marketing / Email Marketing

30 Email A/B Testing Ideas to Accelerate and Boost Conversions

Ben Snedeker

Updated: Aug 20, 2021 · 8 min read

Toolkit for download in this article

Split screen with retro orange telephone tubes on orange background - Flat lay

Sometimes running an email campaign can have the proverbial “throw it against the wall and see what sticks” kind of feeling.

If you’re tasked with boosting your conversion rate, or your open rate, or some other metric, you go to the old reliable strategy: You try something and see how it works. If it doesn’t work, you try something else.

Of course, this approach won’t consistently move the needle.

In order to gain reliable insight, you have to be systematic. A/B testing, when done the right way, is exactly that. And for that reason, it’s one of the most used testing methods for improving conversion rates.

What is A/B email testing?

Also known as split testing, A/B testing is one of the best tools for making sense of how your audience responds to your email. The results of each split test will help you shape your future email messages so that they have better and better impact.

The idea behind A/B testing is that you want to isolate one specific difference between two otherwise identical emails. Then, you split your contact list into two groups, sending one group one email and the other group the other email. Once all the emails have been sent, you measure which email did better, group A or group B.

Now you have a sense of what works better with your audience, and because you isolated one specific variable between the emails, you can attribute the difference in success rates to that one change.

How to A/B test emails the right way

So it’s a pretty simple concept, but there are some very important basics you’ll need to nail down in order to be certain that you get the most success.

Get clear on what you want to learn

For example, are you hoping to get more conversions on a campaign? Boost the open rate? Lower the instance of spam complaints? Once you identify what you want to do, you can start thinking of ways to tweak your emails to help achieve your goals. Then, isolate one single adjustment that could have an impact on the metric.

And most importantly, once you know what you want to learn, you can be certain to identify the best measurement to take. To see more about which email metrics you should be using, download our free Guide to Email Metrics.

Select only one aspect of your email to adjust

Group A and Group B should get nearly identical emails, except for one specific difference. While this the fundamental idea behind testing, it’s also the part that many email marketers get wrong. It can be tempting to test multiple changes at once. Unfortunately, if you change more than one thing, you can only say that one entire email did better than another entire email. You can’t point to exactly why one worked over the other. 

For example, you want to be able to say things like, “When we put the call-to-action above the fold, we get better conversions.” This way you can design future emails with confidence. You can only know this kind of fact through testing an email with the CTA above the fold versus an email with the CTA below the fold. 

Split your audience fairly

You want to split (aka segment) your list such that you give your test the fairest shot possible. Segment your list into two groups as evenly as you can. This means you’ll want each segment to have equal or nearly equal numbers of recipients, and (ideally) equally mixed demographics. If your audience isn’t split evenly, you could inadvertently affect your outcome. 

30 A/B test ideas

Many marketers don’t realize that every email that goes to your list is an opportunity to run A/B testing. The more emails you send, the more opportunities are out there. So don’t discount transactional emails (thank you emails, purchase receipts, shipping confirmations, etc.): You can optimize them to convert for you. What if, for example, your welcome email could do more than just welcome? What if you could drive your new subscriber to your website to find out more about you? Or get them to follow you on Facebook? That’s something you’ll need to test.

Because your audience and your brand are each unique, you can’t rely only on the results of other brands’ testing. Just because their shade of red resulted in higher click-throughs doesn’t mean that it’s the best choice for you. You can base your email copy and design on what has worked for other email marketers—there’s nothing wrong with that—just be sure to use testing to demonstrate its effectiveness with your audience.

So, knowing that you can always make your email messages more awesome, without further ado, here’s a 30-point list of ways to test your email campaigns: 

  1. Subject line word length (many words vs. few)

  2. Emojis in the subject line vs. no emoji in the subject line

  3. Subject line tone (serious vs. humorous, or intriguing vs. straight forward)

  4. Personalized subject line (i.e. recipient’s name in the subject line) vs. generic

  5. Subject line that explicitly states call-to-action (i.e. “Your promo code is inside” or “your 20% off awaits”) vs. not explicitly (i.e. “The best things in life are free” with promo for free shipping in the email body).

  6. Statement vs. question in the subject line
    *In case you hadn’t noticed, the email subject line is one of the most critical pieces of your email to get right. It’s the first (and possibly only) part of your communication that your audience will see. If you want them to open and read your email, you better know what your audience likes.

  7. Plain text email vs. HTML (yes, sometimes plain text email is more effective)

  8. Image-heavy vs. text-heavy email content. More isn’t always better; test to find out if your audience responds to lots of text or lots of images.

  9. Long vs. short emails

  10. Salutation (i.e. greeting in the body of the email): length, personalization, tone

  11. Color scheme (test different variations and use of your brand colors with the font, background, and other design elements)

  12. Video is a hot ticket right now. Like up to 300 percent more click-thrus, for example. Use testing to see how it can work for your audience.

  13. Logo placement

  14. Logo size

  15. Positive vs. negative pitch, aka pitch toward a positive benefit vs. pitch to avoid a pain point

  16. Call-to-action (CTA) copy: generic action word (like “download,” “click,” “buy,” etc.) vs. creative action word (like “get,” “learn,” “discover,” etc.)

  17. CTA button color

  18. CTA button shape

  19. CTA button size

  20. CTA button location (top, bottom, left, right)

  21. Personalized CTA (i.e. first name directly within the call-to-action)

  22. On that note, you can experiment with a variety of design element locations, such as, images, copy, social proof, logos, buttons, and links.

  23. For transactional email, test which CTA’s compliment the transaction: Free resource v. upsell, etc. A thank you email could include a coupon for the next purchase, for example.

  24. Time of day sent: Segment your list to find optimum open times for your audience. Don’t forget to account for time zone.

  25. Day of the week sent. Some days are statistically better than others (also true for times of day, as in the above A/B test idea) for email engagement. But you should test to see what works for your audience in specific.

  26. Messaging by region: Segment your audience by region to identify if particular messages resound with one group over another.

  27. Curated content vs. original content: This is a great way to refine your newsletter, where you often have several links to articles bundled into one email. If you mix in curated content, does your engagement go up?

  28. Personalization: If you are using CRM, your gut can be to personalize every aspect of each email. Find out what your audience likes: generic salutation (“hi!”) with first name sprinkled into the body text vs. personalized salutation. There are a lot of iterations possible here.

  29. Experiment with closing line: Do you end with “thank you,” or “thanks a million?” Do you skip the closing altogether?

  30. Add a postscript (PS) vs. no postscript (especially if it's "PS, here's a {CTA}." Track the click through on the postscript to see how powerful it is)

A/B email testing is all about learning about your audience and how they want to interact with your email. By comparing just one difference between two groups, you can learn how your audience appreciates that aspect of your message.

Of course, you’ll have to make small adjustments over time in order to confidently understand how to be most effective with your audience. But the tradeoff is that once you start gathering data about your interactions, you can confidently make adjustments going forward.  

Was this post helpful?
Illustration of Keap growth handbook
How can you grow your business to the next level? Take our assessment to find out.

The Small Business Growth Assessment will reveal where you are on your path to growth and help you identify common pitfalls so you can avoid them. Plus, you’ll get FREE curated resources to get you to the next stage.

Take the assessment

You may also like

{{ deSlug(record.displayCategory || record.secondaryCategory || record.primaryCategory || '') }} | min read

Knowledge is power, get some more...

Hello, have a question? Let's chat.

Got it