Incrementally optimise your marketing assets with A/B testing. Gain valuable insights to inform future campaigns, boost conversions and improve your ROI.
We are consumed by huge amounts of data every single day.
Yes, you read that right: “We are consumed” rather than “We consume”. Every single day, we’re exposed to over 3,000 ads, each one demanding a portion of our time. Our attention is literally eaten away by data-driven media.
As a marketer, the challenge is finding ways to cut through all that noise so that your own ads capture the right attention from the right audience.
As with so many things, the trick is testing. Lots of testing. A/B testing.
Once your creative team has come up with the campaign, it’s A/B testing that allows you to refine and optimise the various components to maximise your potential results.
A/B testing can help you determine the profitable elements of your marketing campaign, boosting your ROI while providing valuable insights to improve future campaigns.
What is A/B testing?
Also known as split testing or bucket testing, A/B testing creates two versions of your campaign asset that are exactly the same except for one key difference. By serving one or other version to different people and monitoring the results, you’ll eventually determine if one performs better than the other.
And the reason for that difference in performance must be down to that one key difference you tested. Lock in the change and the asset should now perform better for everyone.
In email marketing, you might only have the opportunity to conduct one A/B test before the eDM needs to go out. You might test an alternative headline, banner image or call-to-action by sending two versions of the same eDM to a small subset of your list. Once you have your test results, you’ll know which version of the email to send to the rest of your list.
However, you can test some other assets again and again, continually refining them over time. For example, A/B testing an important web page – such as a campaign landing page or product page – can get quite granular. By themselves, each incremental change might seem trivial, such as whether the Buy Now button is on the left or right of the screen, or even if the button should be green or purple. But over time, these incremental improvements can add up to a significant change in your conversion rates.
Your pay-per-click adverts can also benefit from A/B testing. Slight changes in wording or messaging can have an impact on your click rate. Create two versions and assign a small budget to each. The version resulting in the most clicks gets more budget, while the other is switched off, to be replaced with another variation, another test.
For example: “Get 10% off your first purchase. Book online now!” versus “Get an additional 10% off when you book online now”. The first is a lot more direct, but could also be seen as off-putting and aggressive. The second is more diplomatic but might lack the drive required to close a sale. A/B testing can reveal which is the right option for your audience.
What should I A/B test?
Everything you can, basically.
Websites and landing pages
Experiment with wording, images and design, placement of CTAs, colours, features and content. Do you have a header image or is it a slider? Do you have featured blog posts on your home page or not?
Emails and newsletters
Most CRMs have inbuilt features that make it easy to set up A/B tests. Try testing out different subject lines, different days and times, different CTAs and different designs.
Ads
Depending on what kind of ads you’re running, you can experiment with the titles, body copy, whether to use an image or video, and so on.
Promotions
What’s more enticing to your customers: a discounted price, or the same price with a freebie thrown in? The only way to know for sure is to test!
Can’t I test everything at once?
If you’re thinking that A/B testing is a potentially slow, incremental process, you’d be right. But avoid the temptation to test multiple variables at once. Testing one variable at a time helps you pinpoint whether that variable had an impact or not.
Change too many things and you won’t know why that version worked better than the alternative – and which elements to retain or scrap.
Be methodical, testing each element with intent.
Wrap up
Over time, A/B testing will give you a much deeper and actionable understanding of how to optimise your marketing to appeal to your target audience – and extract the best results. You’ll be able to build more cost-effective marketing campaigns with greater and greater ROI.
