eBrandz Blog

Understanding finer aspects of an A/B test variables

The important thing is to understand when to employ A/B or multivariate tests on basis of the end result that you are seeking. Trying out different test formats on different landing pages (or different offers) can be a good way to attain your goals. This obviously will need expert inputs.

As we have grasped in one of the previous posts, traffic volume is one of the key factors to make ac choice between an A/B test and a multivariate test apart from the page elements to test. You might consider the latter in case of high page traffic volume and if the page layout/ design are to be the same. An A/B test is preferable in case of a low traffic volume and you are looking to analyze different versions of your ad content in context of quality leads and conversion levels.

We shall now discuss more ideas to select the right type of conversion tracking experiment.  Multiple versions of promotional text often link up to the same landing page on a business site. A/B tests will offer a way to tag each of the promotional content versions even when all of them link up to the same landing page.

This will help you to find out which ones prove to be most effective. You can check data on clickthrough rates, average page depth, new leads, conversion figures, usage pattern, visitor loyalty, and revenue for each of the content versions.

A/B testing makes use of the variables that have been set in your tracking URLs for the purpose of comparing values. Specifically, this kind of testing will require the usage of utm_source , utm_medium, as well as utm_content. (It is not necessary for AdWords campaigns to tag the links in case you’re using auto-tagging. Through auto-tagging, variables are filled for AdWords campaigns automatically.)

In case you’ve multiple product types or multiple campaigns, the utm_campaign variable should also be used by you. For instance, if you’ve a ‘spring sale’ campaign plus a ‘brand XYZ’ promotion, you can opt to indicate the relevant campaign in your link. The example below will illustrate the usage of these variables. In it, the email newsletter is comprised of multiple links to a single page on your business site.

The first link here is a text link:
http://www.realexample.com/purchase_page?utm_source=newsletter&utm_medium=email&utm_content=buytext

On the other hand, the second link below is a hyperlinked image:
http://www.realxample.com/ purchase_page?utm_source=newsletter&utm_medium=email&utm_content=buypict

A URL builder tool of Google Analytics will create links, embedding the appropriate campaign information specified by you y. Using this tool will ensure that the links carry the correct syntax. Use consistent names for your campaign variable values. Select a code or name, for example, which denotes ‘cost-per-click’ and employ it consistently. utm_medium=cpc and utm_medium=cost_per_click are separate mediums as far as Analytics is concerned.