A/B Testing: Why is it so important?
A/B testing, otherwise known as split testing, is an experimentation process in which two versions of a webpage are shown to users at the same time. Each user will see one of the variants – so you can compare how they perform based on how every user interacts with each version. It allows us to make careful changes to our user experience while collecting data on the results. In performing A/B tests, we can construct hypotheses and understand why certain elements of a user’s experience can impact how they behave on the website. We generate hypotheses before the test, as a way of devising what we’re actually going to test – then we generate another once we have results to understand the test. A/B testing is extremely useful to continually improve user experience throughout a website – it also helps us improve smaller, individual goals such as conversion rate.
What are the benefits of A/B testing?
There are quite a few benefits to A/B testing. It allows you to reduce abandonment, reduce risks involved in tech spending and development, increase conversion rates and effectively create content for your website. You use data to make improvements rather than opinions – avoiding debates and arguments over what two people think is best. Because you’re constantly iterating – making small, gradual changes as your testing progresses, you reduce investment risks and can see how your website is benefiting from the tests.
One of the biggest benefits of A/B testing is how quickly it can produce results versus ongoing debates, lengthy design and development. The ease of analysis with an A/B test means you can efficiently compare results and decide what changes to implement on your website. For example, running two versions of the same page at once gives us insight into how a user has behaved on both pages. When done successfully, this kind of test shows us what works well, what doesn’t and what needs changing on the web page in order to improve user engagement and drive conversions.
Almost everything on your website will be testable through A/B testing. Typically, clients think about design when it comes to A/B testing, but you can also test what products to promote, what offers are available and pricing/discounts to name a few. This means that even the smallest segment of your site can be tested. Are you having issues with abandonment in checkout? Running a test on your CTA, layout or coupon code box will allow you to see how user’s react to these slight changes. Analysing the results of these A/B tests will in turn give you key insights as to how you should be implementing changes going forward to reduce the risk of bouncing at checkout.
How do you A/B Test?
When starting your testing process, it’s best to focus on a smaller aspect of your website while you settle into the experimentation process, iterative cycles and accepting that some results will be negative. CTA, buttons and styling are good, basic points to begin your A/B testing as they’re relatively low risk, yet still bring valuable results to analyse. Depending on the expected improvement, traffic and number of variants, your A/B test could run from days to weeks. It’s important you have a good understanding of these points – they’ll give you an insight as to how long your test should run for to bring effective results. Statistical significance is the goal – there needs to be enough tests run to be certain that the difference between the variants is not just margin of error.
If you’re having trouble deciding exactly what you want to run an A/B test on, take a look across similar businesses, your competitors and agencies. Doing research will provide you with an overview of how other websites run things – and if they do it differently to you, it’s a perfect launchpad for your test. You might find that running an A/B test based on research shows you that your way is the best way – but this isn’t a waste of your time. The results from this will show you that you’re on the right track – and you can sit smugly knowing you’re doing better than your competitors.
Where to Test
The beautiful thing about A/B testing is that it can be done pretty much anywhere across your website. It’s useful for you to have an idea of what the pain points of your website are, as these are obviously always a great place to start! The most important places to optimise first when testing are the homepage, product detail pages, product listing pages and checkout.
It’s key to remember that you should keep your A/B tests on each section independent. If you want to run multiple tests at once, make sure you’re keeping track of what pages have tests on. Not doing this can cause a mix up in your A/B testing – which will muddy your results and create issues for you when analysing the results of the test. A/B testing works best when you keep it focused on one area at a time. Once you’ve closed the A/B test, analysed the results and begun to implement the changes – you can move on to another section.
If you don’t already – you should promote a culture of A/B experimentation across all areas of your business. A/B testing is a simple, effective way of optimising your website. There’s little excuse for not running A/B tests across your website – they generate really useful data for you to analyse.
Want to start running some A/B tests on your website? We can help.
If you’d like to find out how partnering with JH could move your business forward, give us a call on +44(0)115 7940060 or email us at breakthrough@wearejh.com.