Smart digital marketers know that A/B testing your email marketing is no longer an optional measure – it’s a must have to reach your prospects and engage with them.
A/B testing, which is sometimes called split-testing or bucket tests, is where you run an experiment with two variants, which we’ll call Variant A and Variant B.
In a more scientific explanation, we could call it statistical two-sample hypothesis testing.
With A/B testing, two variants (A and B) are compared, which are nearly identical, except for one small variation that might affect a user's behavior.
Sounds like you need a Ph.D., right? Nope, not at all.
An example of A/B testing is having two web pages, which are nearly identical, with minor differences between them. You send roughly the same amount of traffic to both, and watch to see which one receives the most engagement or clicks or purchases; whatever your goal may be.
A/B testing is often used in Facebook advertising – running the same campaign to the same audience, however using a different graphic or different text, to see which one receives the better response.
The trick with A/B testing your email marketing is to not compare entirely separate emails but to test two very similar yet not identical emails. For example, the variation may be the subject line, the sender name, or something else.
In this article, I’ll explain what factors you should be testing, with the aim of constantly improving one element with every email sent.
Four goals you want to achieve with your emails
When you break the goals of email marketing down, there are four actions that you want your readers to take. Being clever with A/B testing your email marketing, you can increase each of these important steps. I’ll take you through some of the ways you can increase the response rates in each of these steps.
Those goals are;
- See your email in their inbox
- Actually open the email
- Read your email
- Perform an action like clicking a link
Seems very simple, right? Let’s dive into what you can try, to improve each of these steps. If you’re a returning visitor, you can jump straight to the right section using the links above.
Email marketing has an average ROI of 3,800 percent. For every dollar invested, the average return is $38. (Salesforce)
See your email in their inbox
I personally receive between 100-130 emails per day. I read around 60% on my desktop or laptop, and 40% on my phone. I’m busy, juggling calls, emails, and other tasks, just like your prospects likely are.
What can you do to get into my inbox, and for me not to just bulk delete it (which I often do a few times a day?).
There are two major factors here, deliverability and timing.
Deliverability is affected by a number of elements. The main two elements are what system you use to send the emails, and secondly what content is in your email.
Choosing an email marketing platform to send by is very important. If you try sending straight from your own email account using Gmail, Outlook, etc there is a high probability that either your email server will start deleting your outbound emails if you go over a certain volume of sending, or the recipient servers will treat them as spam and automatically delete them.
You should always use a professional email marketing system. They typically manage spam scoring of domains, flagging of email headers, and a whole bunch of other technical elements that you don’t want to worry about.
The bonus is most of these have easy-to-use tools for a/b testing your email marketing, without needing to do anything technical.
Secondly, what content you have in your email. If your email has a large attachment (in fact, any attachment), then a significant amount of recipient servers will delete it or mark it as potential spam automatically. This means you’ll never get to your recipients' inbox, and they will be blissfully (for them) unaware you even sent an email.
Most email servers and office networks have anti-spam software. This very effective software looks for a bunch of different elements in your email content, and rates every email with a spam score.
You want to do everything possible to get that score as low as possible.
For example, if you use any of these phrases in your email subject or within the body of an email, you are increasing the chance of your email being marked as spam;
- Free Money
- Free Gift
- Extra Cash
- Rolex discounts
- (anything of an adult nature)
Be smart, and always consider what sounds at all spammy, and don’t use it. Avoid sending email subject lines in all capitals, or using a reply-to email address such as no-reply@yourdomain
Watch your timing
The next big factor in reaching your target inbox is send timing. What day and what time of day are the two big variables? I can spot an overseas sender from a mile away – they’re the ones that email me at 2.00am. When I check my emails at around 6am, I have 20 other emails like this, which I select all and delete without opening.
Step one in A/B testing your email marketing is to consider the geography of your mailing list. If all of your readers are typically in Western Australia, we’ve already managed to narrow down your send time to a particular part of the day. If your mailing list is geographically dispersed, then we have a wider hour range in a day to test.
Then we need to consider the usual habits of these recipients. Are they desk bound Monday to Friday between 9am-5pm, or are they shift workers, or are they from the always-connected segment? Do what you can to better understand your potential customers, and try variations.
Whilst we can look at various large email marketing studies, your mailing list is different to ours and different from everyone else’s. We’ve seen examples where Tuesdays between 10am-2pm work best, and others where a Friday evening email gets the best open rates.
A/B test suggestions
Here are a few points that you could focus on A/B testing your email marketing, at the seed stage;
- Test elements to improve deliverability
- Test different days and times to send
81% of SMBs still rely on email as their primary customer acquisition channel, and 80% for retention (Emarsys, 2018).
Actually open the email
So now that you have your email in your recipient’s inbox, the next goal is to have them actually open your email.
There are three major factors here. The subject line, the sender name, and the opening text.
Subject lines are usually the first thing an email user notices. If those first few words don’t captivate their attention, the chance of them opening is slim.
A few subject line no-nos, from our experience;
SUBJECT LINES IN ALL CAPITALS – there’s no need to shout.
Spammy words, especially intentionally misspelt words.
Wrong geography, such as ‘Fall specials’- we don’t have fall in Australia (or anywhere in the world outside of the USA) we have a season called Autumn.
Great subject lines to try, include;
You should always think of the recipient rather than yourself – what’s in it for them? Try something like ’15 minutes to better profits for your business’, that focuses on the actual benefit, not the feature.
Headlines such as ‘This one hack can increase closure rates by 7.4%’ create a sense of curiosity in the recipients' mind.
Everyone loves an engaging story. Try headlines such as ‘This one time I went viral’, and open the email body with the continuation of the story.
Nearly all email marketing platforms allow you to insert first names if you have that data. Try a subject like ‘Miles, just a quick question’ or some such.
You should regularly try different email subject line formulas, to increase your open rate.
This article on Digital Marketer reveals over 100 subject lines, and how they performed.
The next step is to try testing the sender name. Most email marketing systems let you adjust this.
Rather than a boring ‘Bam Creative’, we find that using ‘Miles & Bam Creative team’ or ‘Miles from Bam Creative’ often works better. It looks more personal and encourages the recipient to open.
Using a person’s name instead of a company name works, especially if the recipient is likely to recognise the name. At the end of the day, we have relationships with people, not business names.
Finally, the intro text that most email programs show (see the example below), means that recipients get a very short preview. Make it engaging rather than a boring ‘Open in web browser’ or showing a long string of HTML.
The best preview text summarises the topic of the email and encourages the recipient to open the email.
A/B test suggestions
Here are a few points that you could focus on A/B testing your email marketing, at the open stage. Try using a mixture of different elements, and see what works best;
- Test different subject lines
- Test different sender names
- Test different opening text
Email is the third most influential source of information for B2B audiences, behind colleague recommendations and industry thought leaders. (WordStream)
Read your email
The next goal in A/B testing your email marketing is to get the recipient to actually read your email. To do this, you want to focus on the design of the content, as well as the actual content itself.
Your emails should be readable without images turned on, and should be easily legible on a mobile device as well.
Marketing email design
This is a tricky one to get right; you need your email to look great on a mobile device, which typically means minimal images and obvious calls to action, and on a desktop, it needs to look more engaging at a wider and taller view.
Try using a simple one column layout, with minimal images, and then compare click-through rates for links deeper in your emails, than complex multi-column layouts with lots of images.
Marketing email content
The old adage ‘What’s in it for me?’ rings true here. Subscribers or recipients don’t care that you’ve won an award or a new staff member has started. They care about how you can help them.
This is where sharing knowledge and tips, rather than sales messages work much better. Be genuine in your approach, and give away information freely.
Similar to this exact blog post, being free with sharing your knowledge brings you more business than being secretive and uninformative.
A/B test suggestions
Here are a few points that you could focus on A/B testing your email marketing, at the read stage.
- Test different layouts to see what works best
- Try always thinking recipient first – why would they bother reading?
Perform an action like clicking a link
The likely goal you have for your marketing email is to encourage your prospects to click a link of some sort. This link could be to;
- A product or service to purchase
- A blog post, article or white paper to read
- A form to complete to show interest
Irrespective of what the action is, we need to try and get those prospects that have opened and read your email, to click that link.
So we’ve managed to encourage your readers to open your email, and then to read your email, and now we need to work on getting them to perform an action.
In some ways, this stage of A/B testing your email marketing is easier to quantify. You can work on testing various elements, such as;
- Position of CTA (call to action);
- Text of call to action, and;
- Design of call to action.
Let’s go through each one of these elements, in more detail.
Position of call to action
Let’s say you are asking people to click a link. Where it appears makes the biggest difference to overall results. It could be perfectly designed, and read well, however, if it’s tucked into a noisy section of your email, you’ll suffer from low click rates.
Try placing the link near the top AND at the bottom. If your email is long enough, you could possibly even position it again in the middle of the content.
A good way to do the above is to use different phrases and link styles too, so it doesn’t look repetitive.
Text of call to action
Rather than just a ‘Click here’, you will find you get a higher click rate if you use text that explains the destination, or use active language rather than passive.
Here’s an example;
GOOD: Read more about the top five factors that affect SEO on your business website.
BAD: You can read our latest blog post here.
Learn how to effectively use remarketing right now, and be ahead of your competitors.
We wrote a blog post, What is Remarketing? that may help.
These text changes may seem subtle, yet it’s all part of the persuasion to get a reader to click a link.
Design of call to action
Don’t forget to try various methods to link, such as a button or a large graphic or just a simple text link within the content.
Measuring the differences
Let me explain how I go testing the above.
Let’s say I’m sending a monthly email newsletter, which features four recent blog posts. For each blog post, I would use an image from the post and make that a link. I would also put the headline of the post, and link that, and finally, I would also have a link within the intro copy of that specific article.
After the campaign has been in inboxes for a few days, I would then look to see what link was clicked on the most, and which article in which order had the most clicks.
The next time I send, I would be cognisant of what worked and didn’t work over the last few emails, and do my best to improve those rates.
A/B test suggestions
Here are a few points that you could focus on A/B testing your email marketing, at the action stage;
- Test the placement of your call to action(s)
- Test different text, in buttons, graphics or actual link text.
- Use various colours, typefaces or design elements and see what works best.
There are many variants and possible methods when A/B testing your email marketing – the more you try these tests, the better you will reach your audience at a time that suits them, with content that is engaging and with a goal, such as a purchase or new lead, achieved.
Don’t forget there are various stages a recipient needs to go through, in order to fulfill your marketing goals.
She needs to see your email in her inbox
Then she needs to be motivated enough to open your email
If the reader is engaged enough, they need to read your email
Finally, they then can perform an action like reply or click a link
These goals, in simple language, are;
SEE > OPEN > READ > ACTION
Always A/B test your email marketing, no matter if it is a simple newsletter, a once off promotion or a customer on-boarding email flow, read the results, and most importantly… learn from them.