5 A/B testing ideas for your email marketing

• Updated on: 2017-08-26
• Read original article here

5 A/B testing ideas for your email marketing
February 21, 2017 McKenzie Gregory
Share this Article
 
 
We get it: Marketing isn’t always an exact science. Sometimes, it requires a little guesswork. But making assumptions about your audience can be risky, and it often leads to poor results. 
 
That’s where testing comes in. By testing whenever possible, you can let the results do the talking and deliver more relevant and effective messages than ever before.
 
Since there’s a whole world of variables you can test in email marketing, here are five categories to start with to begin learning more about your audience ASAP.
 
1. "From" name and sender address
People get an average of 150 emails a day. That’s a lot of noise you’re competing against, so it’s critically important that you do everything you can to immediately capture your recipients’ attention in the inbox. 
 
64% of people say they open an email based on the "from" name over anything else (Chadwick Martin Bailey). It makes sense if you think about it: In the inbox, you aren’t just up against your direct competitors. You’re also competing with my boss, my friends, my doctor, my adult kickball league – you name it. 
 
 
 
That's why you should pay special mind to (and make sure to test) your "from" name and sender address. 
 
For instance, whenever they host a webinar, Uberflip always sends a branded HTML invitation to their email audience. But then, a few days later, they send a reminder "from" Hana Abaza, their VP of Marketing. Even though I know the email isn't actually personal, the approach and the note often get my attention.
 
Another example? When Nashville AMA sends out an email to their members, they send using the "from" name "NAMA – Tim Earnhart" (the name of their president). But when they send general news to their entire audience, they send using the more general "from" name "NAMA." 
 
 
Using a human's name as your "from" name is a great lever to pull to encourage opens from those who haven't engaged recently. But be careful: Don’t send from a person unless they’re someone you KNOW your audience will immediately recognize.
 
2. Subject line and preheader text 
You can’t talk about email testing without bringing up subject line split testing. If you're an email marketer, there's a very good chance you've already done this at some point. But there are always new ways to experiment with your subject lines. 
 
For example, here’s a recent Emma subject line split test. Guess which one won?
 
 
The one with the emoji! However, we think the powerful thing here wasn’t necessarily the emoji, but rather the length. That subject line just simply stands out more in a crowded inbox. 
 
If you’re going to test your subject lines, keep your preheader text consistent, and vice versa. It’s best to confine the test to a single variable to get the best, most accurate results possible.
Here’s another one. Guess which won?
You're half right! 
"They’re baaaaaack" won by a huge margin with our customers...  but "Look who’s coming back to Nashville" won with non-customers.
The takeaway? If someone is familiar with your brand, you have more room to play around with your subject line than if they don't know you (and your brand voice) quite as well. 
3. Send times 
We get this question ALL the time: What day should I send? What time should I send? 
 
My answer is always the same: Send times are one of the biggest things you have to test with your own audience to figure out. 
 
The data will tell you. 
 
For instance, Emma customer  Nourishwise  wasn't sure where to start with send times when they first launched their email marketing program.
 
So they picked a baseline (9 a.m. on Tuesdays) and tested sending at different times in 30-minute increments for a few months. After some experimentation, they figured out 10:30 a.m. works best for their audience. 
 
Another customer anecdote: One of our Senior Email Specialists , Elizabeth Duffey, worked with a very niche business that sold booking software to charter boat captains.
 
They were struggling to boost their email engagement, so Elizabeth did an audit of their email strategy. After talking to their marketers, she found out that charter boat captains don’t actually get off the water until around 8 p.m....  and they were sending the bulk of their mailings in the afternoon. So they started sending around 9 p.m., and engagement soared!
 
4. Calls to action 
No matter what your brand does, you have some sort of end goal when it comes to email marketing. Because of that, everything you send should work to elicit action from your subscribers. But what kind of CTAs work, and how can testing help you get the best results possible?
 
TEST THE COLOR. We’re psychologically wired to respond in different ways to different colors (blue builds a sense of trust and security, for instance). So experiment with color to see what works best within your messaging.
 
TEST THE COPY. This is huge: Active, first-person copy tends to work best, but there’s a ton you can do here to get the most from your CTA .
 
TEST THE PLACEMENT. Does your audience want to click right away or make it to the end of your message before they engage?
 
TEST THE SHAPE. We’re all about a bold, rectangular button. But we also love a round play button that’s just begging to be clicked. Test it out!
 
5. Old versus new tactics 
This is another thing we hear all of the time: I want to try X strategy, but I’m afraid it’s going to completely tank.
 
It’s a fair concern, but that’s where testing comes in. Anytime you want to experiment, you can give yourself a safety net by using A/B testing to send the experimental stuff to only a small portion of your audience. 
 
One of our clients, Canyon Ranch , runs an annual campaign where they encourage highly engaged customers to buy a membership. Historically, they had been sending that campaign out of Outlook. The idea was that if people were going to spend THAT much money, they’d want to receive a personal email from a human. 
 
Their team was hesitant to embrace HTML messages for this VIP segment of their audience, but with some encouragement, they eventually decided to test it out on a portion of their audience.
 
Turns out, HTML worked much better than expected – and they were able to get larger returns on the campaign than ever before! 
 
 
A/B Content Testing with Emma 
Our  A/B Content Testing  feature takes the guesswork out of your marketing and gives your subscribers a vote in how you market to them.
 
 
You can test variables from 2 emails across a percentage of your audience, then Emma automatically sends the winner to your remaining subscribers based on what you’re testing – highest open rate, click rate for the entire mailing, or clicks on one particular link (or you can manually set a winner).