A/B Testing – Why it matters in digital marketing

What is A/B Testing?

A/B testing is a practice where you take two or more different ideas and push them out into the market. This provides a test to study the effectiveness of each scenario in digital marketing. These ideas or assets could include social media ads, emails, print, and so-on. These assets can vary in design, message, and delivery. What A/B testing does is to help you identify which design, message, or scenario is most effective. Different versions of scenarios and assets are presented to similar audiences. They contain the same general information with slight variations in the design or delivery for example. You would then test each scenario and measure it’s effectiveness. This could be in regards to email click-open rate, more online visits, more sales, or better engagement statistics.

a/b testing

How It Improves Your Bottom Line

There’s a high potential for you to waste your resources on ineffective marketing tactics. With A/B testing, you can see what works and what doesn’t. This enables you to re-design your marketing plan and tactics to match with what a consumer needs and wants. For example, you may have one campaign that uses real people that doesn’t have as much attraction as the campaign that uses animations. By looking at the engagement and metrics of your A/B test, you can change your focus or revise it to be more effective. Testing and re-evaluation helps improve your ROI. Then you can optimizing with the most conversions from your campaigns.

How to Test?

First off, you can have more than two variants of a campaign to test. There are many variables that can change to effect the engagement of the marketing. Limited the variations you want to change will help you look at these elements more closely. The slight difference in one or two variables will help guide you. An example of a good A/B test was done by Electronic Arts. SimCity 5 is one of EA’s most popular games, selling 1.1 million in the first two weeks. Half their sales were from just digital downloads because of this A/B test.

THE ORIGINAL

Electronic Arts wanted to give more incentive for pre-orders. They released a promotional offer that displayed a banner on top of the pre-order page. However, the promotion wasn’t driving as many pre-orders as they had hoped. They changed the layout and design to see if that would change anything.

THE VARIATION

In this variation, Electronic Arts removed the promotional banner from the page. This variation with no promotional offer increased purchases by 43.4%. It seems like the promotional offer deterred customers and they just wanted to buy the game in general. In most cases, a promotional offer may be what customers want. An A/B test proves that you may not know. This A/B test gave insight about what customers didn’t want to see. Maximum revenue wouldn’t have been achieved if Electronic Arts didn’t test the promotion.

Keep in mind, promotions aren’t the only thing you can test. There are many other variables that are worth exploring in an A/B test.

  • Email campaigns
  • Landing pages
  • Graphic advertisements
  • Call to actions buttons
  • Call to action text
  • Images
  • Links
  • Headlines
  • Sub headlines
  • Infographics
  • Subject lines
  • Sending time and date
  • Length of emails
  • Length of paragraphs
  • Color of backgrounds, logos, etc

Timing is Everything

An A/B test takes time to get better results. When it comes to testing a promotion for example, you’ll need to let it run it’s course. It might take weeks before you see a key difference in campaigns. You’ll also want to only do one test at a time to get the most accurate results. Giving an insufficient amount of time for a test could lead to inaccurate or skewed results. In the digital market, you’ll need to account for errors or anomalies the longer you test. Test one variable at a time and be sure to give sufficient time for it. If you change more than one element, it’ll be more difficult to track one variable without any kind of variance.

Digital marketing has enabled retailers with more tools and accessibility that they didn’t have just ten years ago. Take advantage of this opportunity. Create these tests. Then you properly measure your success and failures. Your online presence and sales could improve. With more information, you have more power and control.