A Matter of Opinion
At a recent meetup, I had the chance to discuss a web project with a web marketer. She lamented that her CEO had recently attended an SEO seminar and favored keyword-saturated language for the corporate website. Her option was human-friendly copy because she was sure that would produce more conversions. But they were at an impasse. Without quantifiable data on which style would convert visitors better, it boiled down to a matter of opinion. We see opinion guiding many websites. Not just in the language but in the layout, design, typography, calls to action, etc. This leads to high bounce rates, low conversions, and ultimately frustration for the site owner because the website doesn’t deliver the business they desire.
One of the tricks we use at Tarfoot Consulting is A/B experimentation. Basically, two versions of a page are used in conjunction with conversion goals to determine which version of the page performs better. The experiment runs for a period of time or until results are statistically significant. The results then speak for themselves and the content or design that performs better is put into place as the default page. Here’s an example, let’s say our web marketeer ran an experiment to determine which language style led to more conversions on her website. Two versions of the home page could have two distinct styles: keyword-focused and human-focused. Each time a visitor lands on the home page of the site, they’re shown a random version. Half the visitors will get version “A” and the other half will get version “B”. Then a call to action, let’s say this is completing a sales contact form, is put in place as a conversion goal. If more people to version “B” of the home page fill out the sales contact form, then version “B” has been statistically proven to perform better than version “A”.
What else can I do?
A/B experimentation isn’t the only way to quantifiably gauge performance on a page. Heat maps are also a very useful way to see where users click on a page. Various link styles and positions leading to the same link can be placed on a page and data is then gathered on how many clicks come through each link. Heat maps are very useful visual representations of the data that make it easy to see what users are doing. Many times, we like to sit with users and do focused, user testing to learn more about how users interact with a website. We record the users eye movements, facial expressions, and voice as they navigate a set list of tasks within an web application or website. We also record everything that happens on the screen. then we review this information with our visionaries and project managers to determine what changes and/or enhancements need to happen in order to create a better experience for the user. Web strategy doesn’t work when it’s based on opinion alone. While it’s true that years of experience with the web and user interaction is very useful in the initial stages of development, continual user feedback creates a compelling website that converts visitors to customers or users and ultimately brings in revenue.