A/B Testing2024-10-18T17:43:59+13:00

A/B
testing

Elevate your performance

Did you know, that a successful A/B test can increase your conversion rate by 49%? At Traction we use A/B testing as a way to compare two different versions of the same ad, or content to see which performs better. You can A/B test nearly everything including Google Ads, emails and webpages, and most tests will isolate and compare specific elements, such as copy, colours and buttons.

A/B testing is an efficient and effective system for enhancing and optimising marketing over time. It helps you to develop high-performance digital assets that maximise the return on your marketing investment.

Using science to develop high performance marketing

You’re setting up a campaign and want to maximise ROI? First, create option A, the control, and then create option B, the variant, in which one element from the control is changed. This allows you to isolate how that element affects digital performance.

You randomly divide your target audience into two groups. One group is shown option A, the other is shown option B. You run these campaigns concurrently over a set period of time, and collect data on click-through rate, conversion rate, engagement and other KPIs.

Test complete, you compare this data to see which version of your ad performed better. You then make further changes based on these new insights and test again. You repeat the process over and over, improving your KPI results every time, until you’re satisfied with your campaigns performance.

a-b-test

Why do we focus on testing?

A/B testing is the ultimate way to enhance your marketing efforts. Using real people and hard numbers, it reveals what resonates most with your audience. It therefore allows you to generate a better return on your marketing investment.

A/B testing is unique to digital marketing. Traditional print, broadcast and outdoor advertising doesn’t offer the same positive feedback loop between brand and audience. Sure, focus groups can help to optimise your message pre-campaign, but there’s no easy way to test or enhance that message after the campaign begins.

This is what makes A/B testing so powerful: it reveals exactly what potential customers want from you and your marketing, and lets you make enhancements on the fly.

“A/B and multivariate tests underpin our marketing approach. Over the years we’ve refined our strategies to the point that we now outperform global benchmarks every single time. And we don’t limit it to digital; we encourage our clients to split test everything from sales playbooks to feature requests.”

– Chris Clarke

Have a sneaky peek at some of our foundational tests…

A/B Test Colours

Colours

Which combination of colours make your marketing pop? It might be a surprise to learn just how influential colour psychology is: according to HubSpot, up to 90% of an initial impression comes from colour, colour can increase brand awareness and recognition by 80%, and colour can influence 85% of customers’ purchasing decisions.

This is why Google A/B tested 41 different shades of blue to find the button that performed  best when the company introduced ads to Gmail. As a result they added an extra $200 million of ad revenue to their bottom line.

A/B Test CTA

Buttons

The entire point of most emails, ads and landing pages is to ‘convert’ the audience by getting them to click on a button. Your buttons are therefore one of the most beneficial places to focus your A/B testing efforts.

Between text, colour, size, shape and position, there are a surprising number of button elements to test and enhance. A few traits are universal, however: the design should be eye-catching, the copy should be compelling, and the button should be positioned prominently on the page.

A/B Test Tone of Voice

Tone of voice

Just as important as what you say is how you say it. Tone of voice is rather intangible and subjective, but A/B testing can help to quantify it, by directly comparing one version of copy against another.

The most effective tone of voice will be the one that resonates best with your target audience. You probably have an intuitive sense of whether your voice leans more funny or serious, casual or formal, irreverent or respectful, enthusiastic or matter-of-fact, but A/B testing can be a useful and methodical way to clarify those thoughts and write better copy.

A/B Testing FAQs

Why is A/B testing important?2024-04-23T22:16:51+13:00

A/B testing can seriously enhance the return you generate from your marketing investment by giving you the insights you need to craft better ads and content. At Traction our approach to A/B testing represents an almost foolproof way to increase your brand awareness, user engagement and conversion rates.

What is the difference between an A/B test and a multivariate test?2024-04-23T22:18:16+13:00

A/B tests are simple binary tests: option A, the control, is compared to option B, which features one changed variable, to see which performs better. Multivariate testing compares more than two options at a time by modifying multiple variables, e.g. changing both the font and the colour of a button, then comparing the performance of AA vs AB vs BA vs BB.

How do you run an A/B test on Facebook?2024-04-23T22:19:06+13:00

Facebook offers no fewer than five ways to create an A/B test: through the Ads Manager, the Experiments tool, when creating a new ad campaign, by duplicating an existing ad/campaign, or by clicking an automated prompt. Or you can leave it to our expert team at Traction, and our proven, proprietary approach.

How do you run an A/B test on LinkedIn?2024-04-23T22:19:40+13:00

LinkedIn allows advertisers to A/B test ads through its Campaign Manager tool. You’ll need to create a campaign, then create an exact replica of that campaign, but with one variable changed. It’s quite a manual process, but at Traction we have ways to make it more efficient and effective.

How do you run an A/B test on Mailchimp?2024-04-23T22:20:13+13:00

Email marketing can offer truly incredible ROI of $36 for every $1 spent, and you can dial this number up even further with A/B testing. Mailchimp makes it simple by prompting you to test when you create a new email campaign: just choose the variables, the send times and how many people will receive the test emails. Mailchimp can even send the best performing email to the rest of your contacts automatically.

How do you run an A/B test on Google?2024-04-23T22:21:08+13:00

There are a couple of ways to A/B test in Google Ads. You can manually copy a current campaign then edit whichever variable you want to test, or you can use the ‘Google Experiments’ tool, clicking the blue ‘+’ symbol under the ‘All experiments’ tab, then choosing your specific A/B test experiment from the pre-loaded (but very customisable) options.

What is A/B testing in SEO?2024-04-23T22:21:51+13:00

On top of ads, emails and landing pages, A/B testing can be used for SEO, it’s just the audience will be Google’s bots rather than your potential customers. A/B testing in SEO can be tricky – if you try to get Google to index two versions of the same webpage, you’ll be punished – but at Traction we’ve developed safe and effective SEO A/B testing techniques.

How do you split your audience for A/B testing?2024-04-23T22:22:34+13:00

In A/B testing it is critical that your audience is split equally in terms of size and traits, but randomly in terms of which group each individual audience member is placed in. At Traction we use automated tools that generate this equal, random split.

get in touch

Keen to get started?

Go to Top