GPWA Times Magazine - Issue 24 - April 2013

W hen the word “marketing” comes to mind the last thing most people think of is conducting an experiment, yet testing a valid hypothesis in order to make informed decisions is one of the greatest marketing weapons you can wield. A complex chemical equation is not what you’re looking for here; it’s something much more straightforward and effective. What is a split test? A split test (or A/B test) is when you compare one page against an altered page that contains one major difference. It is a tried and true method through which to test the effectiveness of varying el- ements on your landing page. The point of performing such a test is that it gives you the ability to make decisions based on valid data rather than intuition or guesswork. When you run a split test you must follow the basic principles of the scientific method: 1. Ask a question. 2. Do background research. 3. Construct a hypothesis. 4. Test your hypothesis by doing an experiment. 5. Analyze your data and draw a conclusion. 6. Communicate your results. So, for example, let’s say that there is an office dispute over what color the “Join Now” button should be on your website. Half of your staff believes that the color green is the most appropriate due to its relation to a traffic light. The other half of your team, how- ever, is adamant that red is eye-catching and attractive and will therefore cause more people to click. The question then is simple: “Will a green button produce more conversions than a red one?” Now you have a basis on which to conduct a split test. Create two identical landing pages, one with the green button and one with the red. Randomize and split your current site traffic so that half is sent to Page A and the other half to Page B. After a set amount of time, visits or conversions, compare the results to see which was more effective. These tests need not be limited to design choices, either; they can be used to test changes in layout, struc- ture and functionality. The results you gather from split testing, provided you test correctly, are invaluable in helping make deci- sions as to what elements are important for your audience. Getting started Now, you may find it hard to know where to start, and the most important thing in an experiment is to begin with a solid test- able foundation. This is known as the “control.” Your basic land- ing page needs to be clear and concise, detailing the facts about your product or service so that the features and benefits are obvious to the visitor. Now you need to ask yourself what makes your product or service unique from the rest of the market. Is it that you offer better prices, service or quality? Split Testing by Tabitha Monroe 51 Split Testing