How universities and colleges can use split A/B testing

In a user-first world, digital marketers, designers and content creators need to make sure the journey they’re taking their audiences on is the best it can be.

Gone are the days when the design of a website, email, or marketing campaign can be based on assumptions and aesthetics; best practice now places data-derived decisions at the forefront.

And that’s thanks to a modern world of user testing methods now found in every savvy digital team’s toolbox.

There are a variety of tools out there—both qualitative and quantitative—to ensure your website is optimized for user experience, from surveys and audience interviews to Treejack testing and card sorting for navigation.

But one of the most popular and straightforward means of challenging assumptions and uncovering audience preferences is the A/B test, which is what we're diving into today.

What is A/B testing?

First, a quick-recap on what A/B testing actually is.

A/B testing is one of the key components of conversion rate optimization (CRO) and is intended to eliminate guesswork when it comes to web design.

It’s also known as split testing: you serve two variants of a page, app, email journey, or social ad to users to establish which performs best against a set goal.

The variation can be as subtle as a different colored button, or as dramatic as a completely new page layout or email design.

Whether you’re looking for your target audiences to continue their onward journey on your website or looking for an email sign-up, user engagement is measured so that you can make informed and data-driven decisions on which version has the best impact.

The benefits of A/B Testing

It’s no surprise that it’s among the most popular methods of user testing.

The A/B approach delivers clear evidence that can be used to improve the user experience and therefore engagement.

It allows you to optimize one small step at a time and isn’t necessarily a one-off—it can be used on a rolling basis to inform iterations of development activity and can work well for both small and large data sets.

A/B testing is also one of the least complex methods of evaluating page design and represents good value for money, as it saves digital marketing teams from wasting money on a fully-fledged implementation of poor new journeys that fail to meet user needs.

How higher education can use A/B testing

The application of A/B testing in the higher education digital marketing sector is varied and vast. It can test something as nuanced as a different email subject line or inform an entire website revamp.

There are so many applications for A/B testing across your university’s digital activity.

Here are some avenues to consider.

CTA buttons

Whether it’s on a web page or in an email, it’s surprising how much difference playing with the position, color, and wording can impact conversions.

You may need to optimize a specific page on your site or a specific email campaign, or something that has a more global impact across your digital activity.

Web forms

The number of fields, the way questions are worded, and the form design can all determine sign-up rates.

This could be a perfect place to start using A/B testing to optimize form submissions for open days, prospectus downloads, and recruitment inquiries.

Page content and layout

Whether it’s a headline or body copy content, the length, tone, format, and font can all impact engagement.

This can extend to imagery and video, too.

And more complex A/B tests can delve into layouts and designs.  

If your website is built using components with the flexibility to move blocks of content around, this can be a great way to check changes in the order of page content will result in positive user experience changes.

Emails

Subject lines are a common A/B test activity for email campaigns, with two variations played off against each other to determine which has the highest open rate.

And then there are tests within the email, on aspects like copy length, media, and colors, which can inform optimization for click-throughs and conversions.

Send times can also be a simple and effective A/B test for emails to determine which hour or day performs best.

All these factors can be tested to continuously improve your email open rates and engagement.

Social media adverts

Ads on social media are crucial to enticing prospective students to find out more about your institution, so A/B testing here can be a great fit.

Vary images, text, and copy length to gain an insight into what draws in your audience best.

Website navigation

A/B testing can help to establish the most user-friendly site structure, the layout of navigation that best works for the site, and the type of navigation (though this is best done alongside other testing methods such as Treejacking).

Creating a new navigation approach can be expensive from a development perspective, but even the order of navigation items can improve usability.

 

What timescales are appropriate for A/B testing?

For the higher education sector, A/B testing should ideally run over a minimum of a couple of weeks to capture a large sample of prospective students over the different days and times of the week, when behavior can vary.

For bigger, more complex tests, you may want to run this for longer.

How to run an A/B test

Before you test, be sure that you’ve gathered the analytics data which shows the need and pain points of audiences using your site.

This would be reflected in high bounce rates on a certain page, for example.

For iterations to an email journey, A/B testing can sometimes be undertaken in-house via your current email delivery software.

For more complex tests, you may need to turn to a paid web optimization service such as Optimizely, VWO, or Google Optimize.

The takeaway

A/B testing is a vital tool for university digital marketers if they’re to meet the needs of prospective students and maximize recruitment conversions.

It’s perhaps the best and most reliable means of testing incremental changes and tweaks to websites that are already well-designed.

But, if you’re starting out with a poorly designed website, it’s unlikely to get to the root of the issue, and that’s when A/B testing is best combined with a wider scope of testing tools.

Check out our guide to the best top 10 website optimization tools for more ideas on how to improve your site.

Has A/B testing been used successfully at your institution? We’d love to hear in the comments below or on social media.