Our Process : A/B Testing
We have developed an A/B testing process that is scientific, rigorous and completely beneficial for your site.
We will create a testable, actionable hypothesis, run split tests in accordance with Google’s most recent guidelines, then learn from the results to further optimise your site.
1. CREATING A FALSIFIABLE HYPOTHESIS
Creating falsifiable hypotheses is an integral part of A/B testing. A falsifiable hypothesis is a prediction that can be demonstrated to be true or false with data gained through testing. This is what makes our A/B split testing a precise, scientific method.
2. RUNNING AN A/B TEST
Running a test successfully is one of the hardest parts of the process. We use VWO to make changes to a site, then to show either the original version or the test version randomly to similar visitors.
3. ANALYSING THE SPLIT TEST RESULTS
Once the test period has finished, we will analyse the results to discover if our hypothesis returned a positive result or not. We’ll be looking for statistically significant positive or negative results, not just minor fluctuations.
4. IMPLEMENTATION AND REVIEW
Whether the results are positive or negative, the most important thing is to understand why the outcome was that way and to implement those lessons in the future. This will allow the process to become more efficient and for similar gains to be made across the website.