Would this button be more effective if it was bigger? How will using this font influence user behavior? Why aren’t people clicking this link?
Basically, how do you choose what to include on your site?
One way to solve this dilemma is to use data (a.k.a. facts) to guide design decisions. If you have an analytics platform installed (e.g. Google Analytics), there’s a wealth of usage information readily available about the visitors to your site…for free!
Monitoring the average time on page, bounce rate, number of click events, pages viewed, and so on can unveil valuable opportunities for a site. If you dig a little deeper into the data, you could wind up with golden information for improvements.
Get ready to learn more about two techniques to start analyzing and testing on your website to discover new ways to meet your website marketing goals.
With all that data you’ve (hopefully) amassed from website tracking, where do you begin to find these “actionable insights?”
To be honest, there are several types of customizable analyses that can hone in on very granular aspects of a website.
You can learn more about which pages are loading slowly, whether or not users are interacting with sections on-page, where users are leaving the site and much more. Choosing a proper analysis just depends on which specific goals need to be influenced in the solving the problem at hand.
To start things off, be sure to cross-reference the goals and objectives of the site with the monitored KPIs (key performance indicators) to ensure that a sound strategy is in place. These will point out the most important areas that are performing well, or need improvement.
Once the strategy is squared away, it’s time to nail down the specific types of analyses that are going to solve the issues at hand.
Do you need to know what country the visitors are coming from? Are your product pages performing well? What about the navigation pages: are they being visited? Are there any pages where users aren’t finding what they’re looking for?
Make sure to ask questions that cover every base, find the right metrics to determine the answers to these questions, then identify the areas that deserve more in-depth research.
Some quick, common groupings of pages to get started with analyses are:
Once the data has been pulled that’s relevant to each analysis, look for patterns in the data. These patterns will help determine what is, and what isn’t, working on each group of pages. Be sure to take a good look at these pages to determine what the users may be deterred by.
No design should ever be set in stone on a website. With the rapid changes that happen online, it’s best to stay on your toes and be flexible.
Users do not always react to certain elements in an intended way. To really find the best experience for the users, look for sections to test.
The testing can be as advanced as completely switching layouts and hierarchies, or as simple as changing the font of a form. Any change that brings statistically significant increases to conversion behavior from a large enough sample could be implemented for improvements.
Really, test everything on a website. Even the most seemingly insignificant sections can influence user behavior. These seemingly small changes can add up to a noticeably better experience over time through aggregating marginal gains.
The most frequently tested portions of a site that can influence conversion rates are:
A control and a test group should be setup to ensure each change is isolated and is statistically significant. Try this very simple calculator for A/B significance testing by simply dropping in your visits and conversions.
With that result, set two new plans: one to implemented the winner, and one for the next element to test on-site.
By continually identifying better working components of a page, the increases in effectiveness can start to stack (and so will the amount of desired user actions).
Collecting and using website data to your advantage can be as simple or advanced as you want, depending on your needs. By segmenting and analyzing this data, as well as collecting more info through A/B testing, various sections on-site can be improved on a continual and scalable basis.
For basic analyses, the data is readily available. Simply using your website analytics platform of choice, the standard metrics can be sifted through in order to discover user behaviors and opportunities.
For more advanced changes, in-depth analyses can be performed. Have a specific goal in mind and examine all of the variables that could influence it. Then, using statistically significant data through detailed analyses and A/B testing, reinforce your design choices in confidence.
The best part of these projects is that they show real, data-driven approaches to improve a website. This ensures that the facts support your decisions, leaving you, your users, and your boss happier.
What ideas are you excited to try for your website? Need some other ideas for website maintenance tasks? If so, then let us know in the comments section below!
By Sean Doggendorf, Moosylvania
Contact Moosylvania at: email@example.com