Split Testing
Split Testing is undoubtedly the best way to increase your conversions. This is how it works: you set up variations of what visitors will see when they reach your website, and you add code to your “thank you” or “receipt” page to indicate to your software that a conversion was completed. You use software to track conversions, and the software lets you know which variation is earning you more money. You can either run simple A/B testing that tests just a few variations, or utilize multivariate testing that allows you to test many variations of elements on your website.
There are many tools out there that will track the testing for you. Google offers a tool that is great, and it’s FREE! Simply set up an account Here, and their videos will lead you through the entire process very simply.
Here are the basics:
1. Choose which type of testing you want to perform, A/B or Multivariate. A/B testing will render faster results, so if you’re looking to increase conversions quickly, this may be the best bet for you. If you set up multiv
2. Pick a Page to test. The page must be an “action page” where customers have to perform some action, like adding something to their cart, clicking on a specific product, etc. This may be your home page or a product page. Make sure it is a page that gets a great deal of traffic, so you can see test results quickly.
3. Decide what your conversion goal represents. It may be adding a product to your cart, or purchasing a product, or even filling out a form. Whatever you want customers to do, that is your conversion goal. If you want results fast, choose a conversion goal that already happens frequently.
4. Create Variables. Make variations of the content on the page you wish to test, and then make sure each of the variable make sense when placed together. The more variables you test, the longer the results will take. Try a new headline, a new offer, or a new image. You know your business and customer best, use this knowledge.
5. Analyze the Results. Use these to determine your next steps:
Estimated conversion rate range ? This will show you how well your original content is performing against the new variations.
Chance to Beat Original ? This will tell you how likely the new variations are to out perform the original in the long term.
Improvement ? Shows improvement in percentages.
Conversions/Visitors ? Shows how many conversions per visitors the new variations resulted in.
Relevance Rating ? How relevant was the page you selected for testing?
We can’t wait to hear how testing increased your profits!
Tags: a/b testing, increasing conversions, multivariate testing, split testing


Wed, Nov 25, 2009
Advanced Techniques, Conversion, Ecommerce