TOONIMO'S BLOG
Read the latest on user experience, user training and onboarding from members, partners and experts.

Split Testing in Email Marketing: Tips and Tricks

Written by Toonimo | Jan 12, 2015


Email_marketing_large

 

For all aspects of digital marketing, split testing (aka A/B testing) is what of the most important methods of identifying success while running campaigns and conversions.

With all of the emphasis on micro content and social media engagement, the strongest and most direct digital marketing channel remains email marketing.

For companies to identify success in email marketing, the ideal method is to run some form of split test.

We've covered many of the basic aspects of A/B testing. Here are some of our tips and tricks to split test like a rockstar and increase conversion rates in email marketing.

Over Testing


Every online business has a list of subscribers. Treat the list with respect. Over segmenting the data sets will only reduce the success rates. Why? Each niche data set only provides information about that market, not about the entire list.

Segmenting is a good thing with a strong list but over testing or hyper-testing will either yield weak results or annoy the consumers.

How many split test variations should be sent out? That's a tough call. Use a software, like Optimizely to split test and understand the variations and what success metrics should look like.

Make the Right Call


As soon as the email campaign is launched, the natural reaction is to check the mail host repeatedly. Allow subscribers to mull over the content and let the campaign ride for more than a few hours before drawing initial conclusions.

A normal campaign should run for up to 48 hours before reviewing metrics and results.

Split Testing Fresh Data


Split testing is only successful if all variations are sent at the exact same time to the identified audiences. Otherwise, the data is skewed with many external forces impacting the results.


The rates of the test need to be fresh, relevant and identical.



Run the Marathon Not a Sprint


Every business sends various forms of emails. Some are purely promotional, while others are news or don't advertise a specific product. Testing email campaigns that are only timely for a specific time frame or are non-commerce driven, may offer interesting results but provide minimal value in the long term.

For the best split test results, compare campaigns that are critical to producing significant impact to a business. Sure, these tests are not as exciting in terms of raw data and take more time to produce results, but most businesses are build for long term, not short term success.

Permanence is the Enemy


Split tests are by default volatile and difficult to predict. Each email is its own beast. The entire aspect of testing is the wrench in the digital marketing industry because it is so difficult to pin down specific consumer behaviors.

Split test results should be viewed as fluid and should be repeated as necessary before coming to any conclusion. Just because the test worked once does not mean it will work again.

Patience is Split Testing's Best Friend


Email marketing is filled with variables that can be tested. Every test opportunity suggested by consultants, articles, influencers or friends may not be appropriate for a specific brand. Read articles and blogs with caution and review them for relevance to specific market segments.

While trying to create split tests that are campaign specific, create tests that will challenge overarching considerations that can drive future decisions. Patience is key when conducting split tests.