Marketers are always looking for new ways to improve the effectiveness of their eCommerce emails. They experiment with different types of emails, different verbiage, new CTAs, various pictures, email send times and more. Crafting the most effective email marketing campaign is an always-evolving, rarely-perfect science.
Science is half about making changes and half about remaining constant, however. Every experiment needs a control and far too often, marketers make changes on the fly, without bench-marking the potential for success. This is why A/B testing becomes so important, and it’s why your Magento or Shopify business needs to consistently utilize it.
Understanding A/B testing
A/B testing is a simple concept that has profound potential. It allows you to observe the impact of a single trait by sending two variations to two different groups. For example, Group A may receive the exact same sales email as Group B, however instead of a deal for 10% off, it might be for 15% off. Observing the reactions of both groups will be able to tell you how effective each email was by itself and when juxtaposed to the other.
For example, if the 10% off email to Group A got a 42% click rate and the 15% off email to Group B got a 45% click rate, you might conclude that the 10% coupon is a more worthwhile sale for your bottom line, due to the relative click through rate.
Simply put: if you want to measure the effectiveness of an email or part of an email, A/B testing allows you to do this.
What to measure via A/B testing?
There’s a virtually unlimited amount of variables that can be measured via A/B testing and just as many conclusions to be drawn. This being said, it’s important to isolate very specific variables to measure with each test—changes that are too broad between A/B groups might not yield actionable information. Let’s take a look at a few great examples of A/B variables:
Subject lines and email headers are great for measuring effective open rates.
Quantified variables can be changed to measure appeal. For example, the 10% vs. 15% example from above.
Calls to action and their effectiveness can be gauged via the click through rate.
Image appeal can be measured via the number of image clicks.
In changing these single variables, you can view data across a control group and an experiment group to see what performs better. If your data is conclusive, you can then apply that change moving forward in new A/B testing, changing a different variable to see if that increases results. It becomes an ever-evolving process that will shed enormous insight into different aspects of your email marketing and the behaviors of your subscribers.
Testing with a purpose
If you’re going to run an A/B test, make sure you have a reason behind it. Any good scientist approaches an experiment with a hypothesis: something they want to prove or disprove to further their knowledge. Looking at your existing data and formulating hypotheses about how certain changes may positively benefit your email’s effectiveness is critical. Stay true to your hypothesis when testing and make sure to control as many variables as you can!