07/30/2012 03:53 pm ET | Updated Sep 29, 2012

5 Website Tests All Companies Must Run

Unfortunately, too many companies fall into the trap of believing that -- because they like the way their websites look -- they're as effective as possible when it comes to converting visitors into customers.

Since pretty websites don't necessarily equal effective websites, we need to rely on scientific testing tools to determine which website elements are working -- and which ones should be replaced with more lucrative alternatives. The specific process that most digital marketing experts use is called "split testing," or, the process of serving up two or more website variations at random to determine how visitors interact with these pages.

For a complete guide on how to use split testing on your website, take a look at Smashing Magazine's "Ultimate Guide to A/B Split Testing." Then, use the information to set up your own split testing program that incorporates the test mentioned below.

Test #1 -- Your website's headline

Although past studies confirmed that new website visitors spend an average of seven seconds on a new website before deciding whether to engage further with the site's content or click the "Back" button, I believe this initial window has become even shorter.

As audience's become increasingly tech-savvy, they tend to know what they want and whether or not they're seeing it at first glance. For this reason, I believe a more accurate estimate falls in the range of two to four seconds.

Since you don't have much time to engage new visitors, your website's headline -- the first, most prominent piece of text these readers will see -- has to be as effective as possible. And while you can model your chosen headline off of formulas that have been proven to work in the past, an even better approach to making this content work effectively is to split test multiple options until you have a conclusive winner.

Test #2 -- Your email newsletter opt-in offer

If your website runs an email newsletter (and when you consider the benefits of being able to follow up with past customers, it really should!), you likely have an opt-in form with a statement of benefits somewhere on your site to encourage readers to enroll.

Since effective email newsletters can contribute substantially to your website revenue, it's important to make sure this site element is operating effectively. To do this, I recommend starting with the specific benefit you're offering readers. Try testing different perks related to receiving your newsletter (for example, "Free 20 Percent Off Coupon Code" versus "Free Shipping Coupon Code") to see which offer results in the most new subscribers.

Test #3 -- Your "Buy Now" buttons

If your website sells objects directly from its pages, you likely use some type of "Buy Now" button to entice visitors into purchasing.

Every element of this button should be split tested -- from the color of your button to the text to includes, and to any external graphic elements that are used to make the button more appealing. Even the smallest of changes made to these important images can result in major conversion rate improvements, as seen in the following case study:


In the above example, blogger Glen Allsop (who sells a blog training product titled "Cloud Blogging") was able to increase the conversion rate from 10 percent to 13.9 percent (39 percent) by split testing a few slight changes to his "Buy Now" button.

Test #4 -- Location of action objects

Both the opt-in form described in Test #2 and the "Buy Now" button showcased in Test #3 can be considered "action objects" -- that is, their primary function is to entice website visitors into taking some sort of action. For ideas on where to place action objects, use heat map technology (which shows how visitors' eyeballs track across website's page) to pinpoint optimal locations for your content. In my testing, I've found that "action objects" often convert best in the upper right hand corner of web pages.

Test #5 -- Your calls to action

Finally, any time you're asking a website visitor to take a particular action (whether enrolling in an email newsletter, buying a product or engaging in any other type of activity), you're using a call to action statement.

Call to action statements represent another great opportunity to improve your website results through split testing. Since few web masters are expert copywriters, they may lack the training needed to make the statements as compelling as possible. Instead of guessing, use A/B split testing to determine which call to actions are the most effective.