Recently, a routine eye exam got me thinking about A/B testing. The doctor swung a large steel contraption in front of my face. It’s the one you squish your face against to gaze through two tiny lenses. First, they start with the left eye. With a click between two settings, they ask you, “A or B?” If you’re like me, you answer as if your life depends on the correct answer.
I hesitate and get that awkward silence before they run down my two options again. “A or B?” This process gets repeated until we get down to the nitty-gritty. Instead of shifting between lenses, they’re spinning the lenses to narrow it down to the final prescription. Now the pressure gets intense. It’s like fine-tuning a stringed instrument. And we still have one eye left to go.
So why am I telling you about my trip to the eye doctor? Because as I was sitting there, it got me thinking about A/B testing in marketing, which is when we compare the performance of two versions of content to see which one resonates with audiences more. Like the optimist’s device, it helps you determine which choice presents a clearer vision.
A/B testing is the original field test
The origins of A/B testing date back to the 1920s, but not in a marketing boardroom or advertising firm. Instead, it started in a pair of farmer’s fields. A statistician and biologist named Ronald Fisher ran an experiment where he put more fertilizer on one area than the other and compared the results. It wasn’t until the 1960s and 1970s before A/B testing made its way into the hands of marketers. It began to flourish in the 90s during the digital revolution.
A/B testing can take a few different forms, including:
Website A/B testing, which divides traffic between two versions. That includes everything from the size of a button to the positioning of a search function. You measure which option yields the most conversions or visitors who performed the desired action. Some marketers will even use heatmaps to show where users are focusing their attention.
Email marking A/B testing (including subject line, design, CTA’s, etc.) divides recipients into two groups to determine which version of the email produces a greater open and conversion rate.
Select content (ads, emails, banners, newsletters, advertisements, etc.) to see which one receives more engagement.
Nearly every asset of a marketing campaign can undergo A/B testing. It shows what’s working and what isn’t. With various metric options, your results come quickly, giving you the ability to change things on the fly. What’s even better is that your audiences don’t even know they’re being tested. They act on their impulses and preferences and do not make decisions based on what they think others prefer.
So when is the best time to roll out A/B testing? There’s a joke among marketers that ‘A/B testing’ stands for ‘Always Be Testing.’ There’s no better time than starting right out of the gate. If you are launching a new website or email campaign and want to figure out your best approach, A/B testing lets you compare two concepts to figure out which one works best. Another scenario is if your campaign feels stalled and is underperforming. A/B testing can help isolate the problem and improve engagement.
A/B testing takes the guesswork out of your campaign. It lets you see the results of your hard work and gives you a better roadmap for future campaigns. Some results may surprise you.
Like a trip to the optometrist, the choice becomes clear. Let your users help create the best possible user experience.
Share
