- mobile - +44 7547 351 474
- email - firstname.lastname@example.org
3rd Mar 15
A/B testing for web design is quite simply what it says in the title – testing the design of version A (the current design) against version B (a new design or alteration). Splitting the site traffic to the separate layouts, then measuring the metrics you want to change, analyse which layout performs better and overall increases the UX.
A/B testing can be used when you feel it is required – there’s not really a right or wrong time and is more dependant on what you are looking to change (conversion rate, bounce rate etc). You could be testing for a full page redesign, testing the new layout against the current layout or something as small as the re-wording of a single call to action button – will users feel more obliged to click the button that explains ’sign up for a free account’ or the button that simply tells them to simply ‘click here’.
an A/B test UX flow diagram
A/B testing is not going to give you all the answers overnight. The test should be anything from a few days to weeks or even a month all dependent on the sites traffic.
Not giving a test enough time can give you inaccurate results, as you won’t really have enough of a visitor rate to give you meaning full statistics whereas running a test for too long of a period will give you inaccuracies due to there being more variables that are not within your control.
So making sure you have an understanding of your sites numbers and what it is you want to test is key making sure the best running time for your test.
Decide what it is you want/need to test (a call to action, an image, a redesign?) Set up the two designs you want to test Be bold and test big changes, it could be what your users are wanting Chose the right tool/apps to run your test(s) WHAT TOOLS AND SERVICES CAN I USE TO A/B TEST? There are plenty of websites and apps that you can choose from, some free, some not. It’s choosing the right solution for you, your website and your budget.