How To Use Google Website Optimizer To A/B Test and Multivariate Test
Website owners seeking to improve their profits have to focus on how many of their visitors become customers. A visitor is not the same as a customer. Too many website owners think this is the case and focus all their time and resources on search engine optimization (SEO) and other traffic generating techniques. Whilst this is very important generating traffic should not be where 100% of your time is spent; it’s just as important that you make the most of that valuable traffic once it reaches your site.
A high conversion rate can keep a website afloat even in periods of low traffic volume. If a high number of visitors are being converted into customers, then profits will be maintained across low and high traffic times. Variation in the profit level will be expected, of course, but the salient point is that profitability is constant with a high conversion rate.
Testing To Improve Conversion Rates
To improve the conversion rate, website owners can utilize two widely popular testing methods: A/B and multivariate testing. A/B or split testing takes a single item on a web page and creates two versions. One version is left unaltered, but the other version is changed in some way. The two versions are shown to an equal number of visitors. Whichever has the higher conversion rate wins.
Multivariate testing is much more complex in that it can test multiple combinations of items at once. It takes the basic structure of A/B testing, multiplies it and is able to test many different elements simultaneously. Multivariate testing takes far less time to complete a larger set of tests. You can find more information on A/B and multivariate also called split testing over at Maxymiser.
Using Google Website Optimizer
The website owner may be asking himself how he can implement these two testing methods and reap the rewards. A free service called Google Website Optimizer provides them both. Google Website Optimizer allows a user to design a test using its unique interface that is web-based. Users upload the content, whether it’s text, images, headers or some other components and then design alternatives. The website runs the test by showing the variations to visitors, recording which combination yields the highest conversion rate.
Hints On Usage
Despite the ease of use and sophistication of Google Website Optimizer, some commonsense tips may be in order to help ensure the experience of using the website is a positive one. Firstly, run tests on pages that receive a lot of traffic. This is a basic principle of multivariate testing that helps ensure the reliability of the results. High Internet traffic numbers help avoid distortions and biases that can show up in the data. Additionally, pages with a lot of traffic can be optimized much faster.
While multivariate testing may be faster, the more page sections that are added to the test, the more traffic or time will be needed to obtain satisfactory results. Google Website Optimizer can test a few elements or many of them at once, but the larger tests will likely take longer simply due to the nature of the test. This is the reason why web pages with a lot of visitor traffic can be tested faster.
When selecting the goal of the site, it is important to pick something relatively common. The goal is the action the website owner wants the visitor to perform so he will become a customer. Many people pick order completion as the specified action, but since this does not happen as often as adding an item to a shopping cart, for instance, the results may be skewed. The effectiveness of different combinations is most important; the conversion rate follows from it.