Through A / B testing, it is possible to evaluate empirically the positive and negative impact of a redesign of a page. What to do? Testing different variants helps to experiment and optimize text and graphics to find the best solution.
Specifically, it is possible to create alternative layouts to identify:
- Which labels and texts are more effective?
- What is the graphic aspect of each element (buttons, text blocks, etc.) that captures users' attention?
- What is the correct positioning of the content and other elements within the page?
- What is the optimal number of fields that make up a form?
But is it better to change one thing at a time or proceed with the overall redesign? It depends.
The approach based on small progressive changes is certainly preferable because it allows each change to be evaluated punctually. Making radical changes to the appearance of a page does not make it possible to measure the effects of a single change: by changing several elements at once, it is not possible to understand who to attribute the positive or negative result of the change to.
The individual evidence emerged from the tests can however be useful to support a complete redesign of the user interface.