Continuing with this series of posts on marketing techniques for eCommerce, we will now seeing an eCommerce practice which should be performed more often: multivariate and A/B tests, which are very infrequently implemented in Spain, mainly due to client timing. This runs counter to the need for development based on effectiveness and conversion. Today we will be telling you about the twists and turns of both analysis, so that it’s crystal clear for you.
• Testing pages with high traffic
• Testing the distribution of whose contents is maintained
• Testing specific items (variables) in a page
• More time is required to obtain results
• The coding required to launch this test I more complex than for an A/B tes
The purpose of a multivariate test is testing out different variations of an element in a website.
The main goal of this kind of experiment is, in addition to finding the best combination of elements, improving conversion rates and establishing which specific elements in a webpage are responsible for increased conversion rates.
• Pages with low traffic
• Changing the distribution of contents in a webpage
• Changing the design or message in a webpage
• Performing an experiment. If the webpage on which the test is carried out requires many changes which are seen as basic (from the point of view of usability, accessibility, etc.), it may be a good idea to start by carrying out an A/B test and then refine using a multivariate test.
• It doesn’t specify which content or item gives rise to improved or worse conversion rates.
• An element may improve conversion rates and another element may worsen, it, so that the result is that there is no improvement with respect to the original version.
The purpose of an A/B test is trying out one or more versions of a full webpage with respect to the original version.
The main goal of this kind of experiment is thus to establish which version of a webpage has a higher conversion rate – where by “conversion” we mean executing a desirable action, such as buying a product, subscribing to a newsletter, etc.