Increasing Website Conversion – The Missing Step

Time and time again I see business owners, product managers, marketing directors or namely anyone responsible to the profit and loss of their product, division or company miss the boat when it comes to Website conversion. They either don’t know, forget or don’t understand the value of testing.

When I mention the word “testing” to some people I see their eyes roll back, or I get a blank stare. Some have an idea what testing is, but don’t want to spend the money. Others have no idea what I’m referring to and don’t want to admit it.

Let’s start with what testing means. First, of all it doesn’t mean that all your links, forms and functionality work. That’s a programmer’s view point on testing. What I mean about testing is testing your message, text, graphic and layout. What’s the point? Well, it can mean a huge difference in increasing Website conversion. If you’re not sure about the value of Website conversion read how a 1% increase in Website conversion could mean a 160% increase in profits. Without testing you are most likely losing a boat load of money.

Just about everything should be tested, but when talking about Web conversion and testing you are usually dealing with page elements such as your headline, call to action, page layout, price the offer, etc. There are two types of testing we usually conduct and I recommend both. We utilize A/B split testing and multi-variate testing with live target audience tests and Website traffic tests.

A/B split testing means that you are testing two different page elements or pages. Site visitors or testers see either version A or version B. Multi-variate testing means that you are testing, as the name implies, multiple variations. This test could include the header, call to actions, graphics, offers, etc. and these are tested all at once.

One of the fastest ways to get some immediate feedback on your page elements is by conducting a live test with your target audience. The live target audience test consists of testing several members of your target audience one at a time. This can be done in-person or online. We use to use this method exclusively for usability (how easy is the site to use) but with each test we found valuable marketing/conversion information. Unless we test strictly for usability, we refer to this test as live test.

With online or in-person testing you first develop a set of questions. These questions include how the target audience perceives elements of the landing page. With tools like GoToMeeting and Camtasia you can test and record the subject’s moves and voice. Unless you have sophisticated tools, with this method you lack seeing the person’s body language including their facial expressions. However, even without this information online tests can yield valuable information.

The second type of testing is using a tool like Google Website Optimizer. Of course there are expensive programs out there that will yield more information, but don’t over this free tool. Let’s say you think your home page, landing page or shopping cart page is not performing as well as you like. Perhaps your designer wants to use graphic A and you think you should use graphic B. Instead of long discussion about which one to use you let the audience decide. So you run both on your site. One half of your audience will see version A and the other half will see version B. You count which one had the most conversion and that one wins.

With Google Website Optimizer you can also test the headline, call to action, text, layout, price and other page elements. This test will help you put together a super page which can have tremendous effects upon your Website conversion.

Before you spend another dime on advertising you should have your marketing team test. Pay-per-click is getting more and more expensive and search engine optimization takes long and can be costly too. First test your page elements and then turn on the advertising. Test again and advertise more. With the increased conversion rate you’ll be making more money which you can invest in more traffic.

In testing there is one rule to follow: test everything and always be testing.