Optimize Your Next Design Project with A/B Testing Nick Bowersox | November 9th, 2016
We often say that good design helps your business stand out. This isn’t just a hunch, it’s backed by research that has proven the influence that visual appearance has on the decision-making process of customers. According to a widely-cited study on visual appeal, it takes about 50 milliseconds for customers to form an opinion about your design, and in turn, your company. With .05 seconds to make a good impression, selecting the right design for your marketing is often the difference between gaining a new customer or increasing your bounce rate.
One of the major advantages of using crowdSPRING for design is the ability to receive designs from many different perspectives that will speak to your customers in different ways. Though this choice can be empowering for some, for others it can create a difficult decision. If design has such a major impact on the buying process for customers, how can you be sure you picked the right design when only going by your personal preference?
Many times, you can’t. But that’s okay! Using A/B testing, you can leverage multiple designs with your customers to help make sure you are using the perfect design to achieve your marketing goals.
Most companies with a strong digital marketing presence know not to rely on personal preference when making design decisions. Instead, they utilize A/B testing to deliver variations of designs to their customers and let customer response inform their decisions. The best part? With the right tools, A/B testing is completely seamless for the customers, so companies they can implement the test into their regular marketing activities without missing a beat.
What to Test?
To get started with your own A/B testing, you first need multiple designs. crowdSPRING Design Projects can be set up to award multiple creatives, giving you the opportunity to select as many designs for testing as you’d like.
Two of the best design projects for A/B testing are Email Templates and Landing Pages for a variety of reasons. First, the performance of emails and landing pages is easy to quantify: whereas a logo or illustration is up to interpretation, more clicks on an email or conversions from a landing page cannot be denied. Second, there are some great tools available that make A/B testing Email Templates and Landing Pages extremely easy.
Marketing Emails are the most effective way to communicate with your customers, and it’s not even close. According to a report by global management firm McKinsey and Company, “E-mail remains a significantly more effective way to acquire customers than social media—nearly 40 times that of Facebook and Twitter combined.”
Though more effective than social media, email requires good design to truly reach its potential for converting customers. Images, placement, and typography all play a role in the effectiveness of an email, so it is important to test what works and what doesn’t with multiple designs.
Most email marketing software allows you to run multiple types of A/B tests including subject lines and send times, but focus only on testing the email’s “content” and keep everything else the same. This will allow you to send 50% of your email list one Email Template, and 50% of your email list the other for an even test. A day after deploying the email, compare metrics like click-through rate (CTR) and conversions to see which template your customers preferred.
Landing pages allow you to create a custom-tailored experience for your customers based on current marketing campaigns or targeting. According to John Jantsch at Duct Tape Marketing, “Today, landing pages have simply become a required element in the marketing toolbox for every imaginable business, including local brick and mortar types.”
A/B testing on landing page is a little less straight-forward than splitting your email list in half and sending them two different designs. Landing pages randomly show a variant of the landing page to a customer, who continues to see the same variant of that page on subsequent visits due to cookie tracking. For a fair test, each variant should be given a 50% chance to be shown to customers. According to best practice, over 100 visits to each page is required before making any meaningful assumptions. Once you have enough traffic, compare metrics like click-through rate (CTR), conversions, and bounce rate to see which performed best.
Let The Data Decide
Image courtesy of Mailchimp
At the conclusion of an A/B test, hopefully you feel more confident about which design to continue using for your brand. With a design selected, you can now try A/B testing other elements of your emails and landing pages including messaging, promotions, and pricing to better optimize for your audience. Consider using 1on1 design projects with the winning creative to alter the design for A/B testing, and to create additional marketing materials that align with the winning work.
And just because you found a winner doesn’t mean you should settle for too long. Continue to test new designs to help your brand evolve, using crowdSPRING’s creatives to provide a fresh perspective on your brand and letting data guide your decisions.
You can’t A/B test without some options. Start a Design Project to enlist the help of thousands of designers that can help convey your brand’s message. Design projects get 110+ entries on average, making it easy to select the perfect options to help your brand stand out.