Lessons In Marketing: What Is A/B Testing?
A/B testing is also known as split testing. It is a way to improve your email marketing exponentially by testing different email marketing messages. It’s called A/B testing because you have version A and version B that you send to your audience. Then you look at the metrics to see which one worked best. Only a small thing is changed in version B compared to version A. That’s how you know what element caused the better or worse result. It’s important that you only test one thing at a time so you know what caused the change. Important to note, you need to have about 100 people on your list to allow you to split test, otherwise it’s not going to work well as far as statistical significance.
Assuming your target audience has signed up for your list, you will send half your list message A and half message B. Send message A to group 1 and message B to group 2. Then you’ll take a look at the metrics to figure out which worked better. For instance, perhaps you simply change the subject line in version B and nothing else. What might then happen is that version B gets five times the opens as version A. The message does not have to be completely different, and many email providers allow you to do this without really breaking a sweat. In Aweber, it’s easy.
You’ve learned something about the subject line. You can now send version B to all of group 1 who did not click, to find out if more people click on the message again. In some cases you can set up your email autoresponder service to do these tests automatically so that the best subject line gets sent again to the people who did not click in the other group. It might sound a little confusing, but it’s really not confusing if you’ve chosen the right email marketing software to use. You can easily do this and with little time and effort – you find more people are opening your messages and you are making more sales and getting more customers in the door.
What Should I Test?
There are other things you can subject to split testing. For example:
Time of day – What time of day does your audience tend to open messages? Send the same message to some of your audience at one time, and some at another time. In this case, if your list is large enough you might be able to pick up to four times. Then, whatever time automatically gets opened more often, you can send the message again to the rest who did not click. This type of testing is easy and doesn’t even require you to think of another type of headline.
What day of the week – Want to figure out what day of the week your audience opens your email promotional material? Just do a test. You might find that your audience opens newsletters more on Monday, and sale information more on Sunday. As you learn more information about your audience, take note of it, and incorporate it into your email campaigns. You can test this and see if different types of emails are opened at different rates. If you send frequently, you can already see which days are higher opens. You can look in your email reports and see what is popular.
From address – Do your subscribers respond more if the from field is your company name or your personal name? Some people may respond better if they know it’s a trusted brand they know. Others may feel better if it’s more personal. You can also test adding a [Wholesale Fashion Buyers] type message on the subject line so people have an idea of where it’s from or what it’s related to before they open. This can increase open rates as well.
Personalization – Inside your emails you might write to “Mrs. Jones” or you might write to “Sally.” Which works best? What approach gets the best response from your subscriber? This is another place where A/B testing can help. Or try it with no personalization at all and see what happens. I’ve heard both work, so which is better for your list.
Hard sell versus soft sell – You can also test the type of words you use within your email message to find out if your audience responds better to a soft sell message over a hard sell message. If you create some urgency, do you see a lot of people clicking through and opening? While not necessary all the time, you can use this in some messages and see if you get a better response.
By using split testing or A/B testing, you’ll increase the effectiveness of all your email marketing campaigns. You should most definitely do A/B testing on every single marketing campaign that you create so that over time you can collect data that will make your email marketing even more effective. Why wouldn’t you? It’s easy once you figure out how to do it, and if you challenge yourself to always come up with 3-4 good headlines, you will have a way to immediately begin testing your email campaigns. Give it a try and see what happens. You really have nothing to lose, most email providers give you unlimited sends in a month – so start now.
What about you? Have you done some A/B testing and how has it worked out for you?