How to Execute A/B Split Testing for E-mail Campaigns If you’re a Beginner

How to Execute A/B Split Testing for E-mail Campaigns If you’re a Beginner

Brenda Savoie
28 September 2017
0 Shares

There are a number of reasons for why e-mail marketing has become one of the most cost-effective ways of generating sales. It has the power to transform a struggling start up into a thriving business that’s going to stand the test of time. The power of e-mail marketing comes from the fact that you have a repeated line of communication with your customers. This gives you multiple chances to convert a lead into sales.

However, crafting subject lines and e-mail content is not a simple process. You need to fine tune the e-mail marketing sales funnel to increase conversions – a process that should never end if you wish to make the most of all incoming traffic. Media monitoring is one way of gathering inspirational ideas to generate content.

To optimise the campaign, A/B split testing can be used to find out what works and what doesn’t. Delve deeper into this article to figure out how you can use A/B split testing to your advantage on your next e-mail marketing campaign.

How A/B split testing works

The purpose of A/B split testing is to test one aspect of the campaign and one aspect alone. This is done by keeping all other variables consistent to ensure the test is fair. As an example, you could broadcast an email with two different subject lines to see which one receives a higher engagement, such as open rates.

However, the open rate isn’t the only engagement metric you should be measuring in this example. You also need to measure the conversion of each e-mail. It might be the case that one subject line gets a higher number of open rates, but generates fewer sales.

Once the results of A/B split testing come in, you can begin to paint a picture of what works and what doesn’t. The version of the campaign that produced fewer result should be discarded and the one that proved more successful can continue to run.

A/B testing also allows you to examine any number of variables for your e-mail marketing campaigns. This includes subject lines, call to actions, different graphics and so on. As Peter Gibson, Chief Marketing Officer from BestEssays explains:  “constantly A/B split testing as many variables as possible allows campaigns to be finely tuned for optimal performance”.

What sample of subscribers to test on?

As a rule of thumb it’s a great idea to test campaigns on your entire e-mail list. This gives you the most accurate results regarding how your audience will react to certain types of changes. If you were to test on a small portion of people, the sample size might be too small for the results to be meaningful enough.

Nevertheless, there are a few scenarios where opting to run A/B split testing on a segment of your list might make more sense:

     Save costs of testing: autoresponder companies tend to charge per email sent when you acquire a lot of subscribers. Therefore, it might be cheaper to run tests on a segment of your list.

     Avoid unsubscribes: if you feel that your A/B split testing might upset a few people on your list, then use a caution based approach and only test on a limited portion of your list. This allows you to trial some bold ideas without the fear of doing major damage to all the hard work it took to acquire those subscribers.

     Discounts and limited time offers: to test out what type of sales work best without losing out on money you could test on a few hundred people. Once you figure out what type of offer works best you can offer it to others.


Tips to consider for optimal split testing

Be creative

Too push the boundaries of what’s possible you’ll need to think outside of the box. An idea that might sound a little risky might actually prove to provide great results and you’ll never know if you don’t test the idea. You’ll need to be original in order to stand out from the other mail lists that your audience might be signed up to. Use a media monitoring such as Meltwater to discover industry trends so your company can be at the front of the conversation and come up with new innovative ideas.

Take action

Once you receive the results of your A/B split testing take action right away. Don’t sit on the newly acquired information. Every day that you don’t optimise your e-mail campaign you’re essentially leaving money on the table. The process of constantly modifying your campaign might require constant work, but the payoff is worth it, as you’ll receive higher conversions.

Learn the tools

Autoresponders like Aweber and GetResponse provide a number of tools to improve your e-mail marketing campaign. They’re a great addition to your communication stack. For example, you can fine tune the sign up form to receive a higher number of subscriptions or make e-mails more attractive with graphics. Understanding the features of the tools gives you further opportunities to split test different elements.

Statistically significant

Try to use samples sizes that are statistically significant. What this means depends on a number of factors and might differ depending on what you’re A/B split testing and the total size of your list.

Use frequently sent e-mails

There is no point in A/B split testing for email’s that you send out once or twice a year. The time taken to test is too long and the results might be outdated by the time they roll in. It’s best to test on frequently sent newsletters, so that a week later you can put the newly found information into action.

Understand your results

Paying attention to the statistics regarding the results of your A/B split testing is vital for figuring out what content is resonating. However, you should also try to understand what’s happening behind the numbers. For instance, if one type of call to action works twice as well figure out why. Perhaps you can ask for feedback from the audience directly to get to the bottom of the answer.

Conclusion

By keeping in mind the principles above you can perfect the sales funnel of your e-mail marketing campaigns to produce record breaking sales. Keep in mind, that you have to be patient with A/B split testing to generate significant sample sizes.

It’s an ongoing process that should always be part of the email marketing campaign. Even some of the most successful companies in the world with impressive open rates and high sales number continue to optimise their campaigns via A/B split testing.

Brenda Savoie is a content marketer, private English tutor, and desperate dreamer. Writing her first romance novel. Seeking contentment through mindfulness. Check her last blog BestwritingClues Find her on Twitter and Facebook