Optimize your site with A/B testing

Big companies like Google and Amazon are spending huge amounts of time and resources continuously testing their sites. The reason is simple – the better your site is optimized for your target users, the better results the site can deliver.

Understanding the target users can be difficult, since preferences can change from country to country, from industry to industry, and even change over time. A/B testing is a great tool for getting better understanding of your users preferences, and for optimizing you site for specific goals.

In this post we will provide you with some tools that allows you to set goals for your A/B tests, and then set up A/B tests on your own site

A/B tests as concept

The basic concept behind A/B tests is pretty straightforward, you identify something on a page on your site that you want to improve, builds a new version of the page, and then randomly serves the two versions of the page to your users.

After some time you can compare the user behavior on the two versions of the page and select the best performing.

A/B tests are primarily suitable for very specific tests with quantitative results, that can be monitored straight from your site statistics.

If you’re looking for qualitative tests, without any specific or well defined area, there are better  tests available – e.g usability tests or focus groups.


Before a A/B test can be designed, you need to have a goal with the test – what do you wish to improve on your site?

Some of the typical goals are:

  • signup rate for newsletters
  • conversion rate in shopping carts
  • click through to product pages
  • click to partner sites
  • click on articles
  • etc.

Valuable sources for target setting for the test, can be the internal statistics from your site or best practices.

Best practices

Before setting up a test, you need some hypothesis to test. These hypothesises can either be based on your own theories, or from best practices from other sites. If you need inspiration for new tests, the site ABtests.com is worth a visit. On ABtests.com you can see test results and best practices within the fields:

  • email newsletter sign-up page
  • home page
  • landing page
  • launch reminder page
  • pricing page
  • product page
  • sign-up page

Based on these best practices, you can start working with some hypothesises on what you want to optimize on your page.

Getting started

Now that you know what you want to test, it’s time to implement the A/B testing on your website. In this article we’ll base our guide on the free Website Optimizer tool from Google. The tool is free to use, and you can use your normal Gmail or Google Analytics login to log in to the tool.

Setting up a test requires the following preparation before you start:

  1. Select the page(s) you want to test
  2. Create an alternative version of the page, including the hypothesis you want to test
  3. Identify your conversion page. A conversion page can be a shopping basket, a newsletter signup page etc.

When setting up your experiment, you need to give your test a name, and specify the original page, the new page, and the conversion page. To avoid errors, the url’s are instantly checked when you have typed them.

Next step is to copy javascript into all the files specified in the previous step. When this is done, you can validate the implementation and put the test live.


To get a valid result, you need at least 100 conversions per page variation, so depending on your site traffic, getting the results might take some time. A/B testing is often a iterative process where you test -> implement -> test -> implement… The greater importance of the part of your site you’re testing, the more often you should test it.

Happy testing :-)

Share and Enjoy

  • Facebook
  • Twitter
  • Delicious
  • LinkedIn
  • StumbleUpon
  • Add to favorites
  • Email
  • RSS
fold-left fold-right
About the author
Jørgen Nicolaisen has been passionately interested in everything online since 1995. His experience is based on working with small hobby projects as well as high volume websites. Jørgen is currently focused on the PHP based programming framework - Codeigniter, and WordPress naturally

Comments are closed.