After spending 6 months or so trying to perfect your new e-commerce website, the last thing you probably want to hear is that your development, design, and overall perfection process is just beginning.
I need to make an addendum to the piece I wrote back in October 2013 explaining the five things you need to do on the day you launch your website. I need to add a sixth item to that list: Begin Usability Testing.
Over the last 6 months, I've read a lot of newly written reports about usability testing, and although the concepts are not new, I feel that the public conscious awareness has grown considerably recently.
Usability Testing refers to the process of evaluating the effectiveness of your website by observing the actions of real users. There are sophisticated facilities that uses multiple video cameras and screen recorders to track user eye movement relating to their mouse movement while observers watch, listen, ask questions, and take notes. Those facilities are expensive, but that's not the only way to perform usability testing today.
Through Usability Testing, you can discover where your website is non-intuitive, difficult to navigate, and frustrates your users. The results will help you redesign your website one small piece at a time.
Years ago my own company used to spend months trying to design the best website for every customer. The process was tedious and was usually based on the likes and desires of our customers. Even though we disagreed with some of the designs we implemented, the customer thought they knew best. Invariably we redesigned those websites over again within 12 months once we had data showing how bad the designs were for their bottom line.
The lessons learned through that process was that no one should simply trust their own website design ideas. It's better to put up the website fast, then immediately activate some type of usability testing so you can find out your failures quickly.
Let me pause for a moment on the topic of Usability Testing, and talk about A/B Testing. They are different types of tests.
With an A/B test you are establishing a specific test surrounding on change on your website. This change could be a background color, an image, or even a change in how the navigation works. The point is that someone in your organization needs to identify some aspect that needs testing, and then set up the test.
Google Analytics has A/B testing built in now; they call it Experiments and you can find them in the Behavior section of your Analytics account.
Google Analytics has the ability to track website Goals through their Goal Flow Reports, but those reports only help you understand how people move through your website, and if they are getting to the right place that you want them to get to.
So you see, although A/B Testing and Goal Flow Reporting are easy to set up, they only give you answers if you know the questions to ask. You might not know what questions to ask, or worse, you could be asking the wrong questions.
Through Usability Testing and watching people use your site you will discover what's wrong without knowing the questions to ask ahead of time.
I explained the expensive Usability Testing process above that requires multiple cameras and an observer, but there are much less expensive methods.
Although I haven't tested GhostRec, I see that their pricing structure is the same as SessionCam, which I didn't particularly like. You could try their demo before paying them, but you should also read my previous review before evaluating any of these companies.
For those of you not yet convinced that Usability Testing is important, let me present an analogy. In your retail store you will occasionally reorganize your showcases based on daily interactions with customers. What you're doing there is real life changes based on the usability you experience with customers. Through website usability Testing you can extend the same process online and get the guidance you need to continually redesign and improve your website.
Your website is never complete, and your e-commerce site will never reach its full potential without this type of testing.
Fair warning though, this is not easy, and it takes a while to gather, analyze, and comprehend the results.