What is A/B Testing?
Have you ever revisited a website and noticed a subtle difference to the design or content? You’re most likely being served a variation to see how you respond to it.
Perhaps you didn’t even notice the change – either way you are acting as a guinea pig for the company to gather smarter data for how they can generate more sales and/or create more leads from their website. The actions you take and journey through the website is monitored and assessed.
The digital marketing landscape has changed how sales and marketing practitioners work. With smarter data and greater insight, the digital world reveals the nitty gritty of where a campaign has been successful or where it might have flopped. There is no hiding. Everything is traceable and measurable, A/B testing is at the centre of this.
Lead generation and conversion optimisation has become a fine art for digital marketers. With a hundreds of analytical and campaigning monitoring pieces of software available testing variable elements, the opportunities are endless.
So what exactly is A/B testing?
A/B testing is simply testing two or more (A/B/C…) different elements of one action to determine which is the most effective.
This can be as simple as the colour of a button that you wish the user to click on, or the call to action in a piece of copy. There may even be multiple test variations. Contact form buttons are a regularly tested part of most company websites. Content and colour psychology can play an important part in the users conscious and creates that all important lead or sale for the business. A different user will be served a different button.
From several weeks of data we can then track which buttons converted the most visitors to the contact page. Marketers can learn which copy and what colour had the greatest impact.
A/B testing is often carried out for big website changes. You might have been given a Facebook update ahead of your friends. Facebook use the information on how you use this update to make small changes and amends before releasing it to all of their users.
Where else can you A/B test?
Almost anything digital with a call to action can be A/B tested.
This might be from the home page on a website to the call to action on a booking page. It might even be a Facebook advert or newsletter campaign with different variants. However, A/B testing is most commonly found within Ecommerce web teams.
Ecommerce
A/B testing has been hugely popular with big Ecommerce websites such as Amazon, Ebay and Etsy. Their potential for A/B testing is vast due to the range and combinations of testing by SKU and product categories. Their A/B testing can also includes product images, pricing, search results, product detail and even shopping basket designs and notification updates.
If you then look at the difference between mobile and desktop versions of these big online retailers and the A/B testing opens up a whole new dimension. As soon as the server knows what type of phone or mobile device the user is browsing on, different things start to happen. For example, a user browsing on a new expensive iPad Air is more likely to pay an extra 20% for an item than a user on a cheaper, low end mobile device.
Large organisations such as these invest huge resource into A/B testing in order to understand how best to push the customer to the conversion point of paying for those goods. Even the slightest changes in terms of colour or phrasing of copy can have a substantial impact on whether the customer reaches the conversion point. This is part of the conversion optimisation practice that digital teams undertake.
A/B Testing for Ads
We test multiple elements for our own work in newsletters and advertising with tools such as Facebook ads, retargeting and for our clients on Google Adwords. The test might be simple changes, but these small changes will create different levels of engagement:
Remember that when testing ads you can experiment with types of images to see how user react for both CTR and conversion.
For example, users are much more responsive to images that they can relate to; using an image with people rather than a generic image can be much more effective. You can then test image group types and find out which work best for which message you’re marketing.
Copy:
For a Google Adwords campaign you might run man different adverts but you may wish to test the phrasing of an ad to measure the CTR.
For a user searching ‘leeds digital agency’
Test A:
Digital Agency Leeds
www.companyurl.com
Leeds based creative digital agency.
Test B:
Digital Agency Leeds
www.companyurl.com
Talk to us, we make the web work.
It’s reported that users view their search results for less than four seconds before clicking through to a page, so testing this copy to see which is the most effective is incredibly useful.
Not only can you measure which of these two tests has the higher CTR, but you can measure which copy delivered the best leads and engagement on the website.
Why A/B test?
Done correctly, A/B testing will give you fantastic insights to what designs and calls to action best suits your digital campaign. From gathering data and learning what works and what doesn’t you can streamline your digital strategy and see a greater ROI.
If you’re new to digital marketing and are exploring digital tools we recommend some of these tools that will help you with your A/B testing:
Google Analytics – for reporting and understanding your digital strategy.
Google UTM Generator – for unique URL IDs
Campaign Monitor – for managing email newsletter, A/B testing and analytics
Adroll – for running smart retargeting campaigns
If you’re interested in making the web work harder for your business or would like to discuss your overall digital strategy, you can email us here.