testing for conversion optimization

Test with Visual Website Optimizer

testing for conversion optimization

Example of an A/B test, where one variable is changed and tested.

Testing is the lifeblood of running a successful website. Without experimentation, there is no way to tell whether your online marketing efforts are paying off. Here are the two types of testing you’ll see referenced most often:

  • A/B Testing
    • Testing variations of a page
    • Especially helpful when comparing overall design or single elements within a design
  • Multivariate Testing (MVT)
    • A/B testing, while useful, has its limitations
    • MVT testing involves testing combinations of multiple elements
    • Dramatic results can be obtained by varying the copy, design, images, and other factors online
    • MVT will give you the winning combination you need to stay ahead of the pack

In addition to A/B tests where you send half the traffic to one page and half to a similar one that has specific variations, Visual Website Optimizer lets you track multiple items at one time through multivariate testing. This helps you determine the best collection of offerings that will help you achieve maximum conversion rates.

You can test headlines, images, offers, calls to action, and so on, and after the software has collected enough data, it will give you (after 100 conversions) statistically reliable reports as to what makes the site most effective at closing deals.

To run experiments, you’ll need to add the snippets of code on your site that will enable Visual Website Optimizer to vary your traffic to the different versions of your page.

Define and track your goals, and make adjustments on an ongoing basis. By using reporting software tools like Google Analytics and Visual Website Optimizer, you can make informed decisions with regard to adjusting your site and calls to action.

It is critical to understand which elements to test (the ones that will have the most impact) so you don’t do too many things at once and get lost in a sea of data. Start small and be certain which elements are making a difference before doing more complicated tests.

photo credit: fumi via photopin cc

0 replies

Leave a Comment!

Leave a Reply

Your email address will not be published. Required fields are marked *