The problem is that, for the small business owner, large scale direct mail testing is too expensive and not practical. The solution? Use your website as a means to quickly and easily test page elements, finding the combination that maximizes conversion rate.

We are going to look at two basic types of web testing; A/B split testing and multivariate testing.

A/B Split Testing

This is the most basic type of test you can perform. The way it works is by splitting your web traffic evenly across two different versions of a page. Let's say you have a landing page for the sale of your ebook. You get decent sales from the page but you want to find a way to improve those conversions.

With an A/B split test, you would test an alternate version of your sales page against the current one. A program works in the background to deliver the current version of the page to 50% of visitors and the new test page to the other 50%. After a set period of time, you see which page is converting better and use that as your new landing page (as well as the page to beat in future A/B split tests.)

One disadvantage of A/B split testing is that you can really only test one item at a time. For example, suppose you test a page by changing both the heading and the image. After the test, you find the page has a 25% higher conversion rate than the original. What you don't know is whether it was the new heading, the new image, or the combination of the two that made the difference.

Because of this, A/B split testing is typically done by changing only one variable at a time. This means that you cannot test variable interactions and combinations. So, A/B split testing is best for when you only want to test a single page element. But what if you need to test multiple factors? That's a job for multivariate testing.

Multivariate Testing

Multivariate testing allows you to simultaneously test several variables on a page. This allows for variable interactive testing, something that just can't be done with A/B Split testing.

For example, with multivariate testing, you could test a page's headline, images, pricing and guarantees all at the same time. Each of your visitors would see a different combination of elements and, given enough time, you would see the combination that produced the highest conversion rate.

You would then change out your existing sales page with the one the produced the highest conversion rate from your multivariate test. Because you are testing multiple elements, you need to have a good amount of traffic coming to your site to complete the test in a reasonable time. If you don't have at least 200 unique visitors per day, then stick with A/B split testing until your traffic increases.

The Google Website Optimizer

The Google Website Optimizer is a free program allows you to run both A/B split and multivariate tests on your website. You've just have to give credit to the folks at Google. They have made it possible for the small business owner to use sophisticated conversion rate tools without the need to purchase expensive software or on-site consultants.

Check out the Google Website Optimizer homepage today for instructions on setting up your tests. Make no mistake, testing your landing pages is the number one way to increase your conversion rates. Start testing your way to higher profits today.