A website overhaul requires careful planning and detailed execution. Marketers and designers must not only craft a visually compelling and easily navigable website, but also ensure that the site aligns with the company's long-term strategy.
As they embark on and carry out a website project, marketers should ask themselves and their teams six probing questions about the eventual website's design and performance. Using research and findings from user testing, they can address these questions, optimize the new website and guarantee a successful launch.
How do users understand my content?
Card sorting can help marketers group elements of the website and structure content in a way that makes sense for the end user. For this test, marketers should write out all of a website's pages, categories, and pieces of content on index cards or Post-It notes and ask a variety of users to group these cards in the most sensible way. If external users are not an option, you may gather a diverse group - from marketing leaders to technical experts - within your company as an acceptable tester proxy.
Using card sorting before creating a new navigation structure helps marketers organize a large amount of information and group various parts of the website into user-friendly categories.
At our digital agency, we recently overhauled our own website navigation structure, otherwise known as a services taxonomy. As part of this process, we asked leaders across different business units to card sort our existing services, activities and deliverables. This process helped us understand how we should group and present our services to current and potential clients.
While it took some upfront work to write out all the terms and recruit employees from various departments for the activity, this was an easy, inexpensive test that will optimize the user experience and change the way we go to market.
Does my navigation work?
Navigation and taxonomy testing helps marketers determine whether users can find what they are looking for on a website. While marketers can use this test at any time, it is especially valuable before building a new navigation structure. The results of this test help marketers restructure the website's navigation in a way that makes the most sense for the end user.
One example of successfully using navigation and taxonomy testing was when working on a website redesign for a university. Ten users were assigned to test a new navigation. The first five testers found several areas for improvement, allowing for labeling and structural changes in line with their feedback. The last five testers then validated the changes to the website and confirmed that visitors would be able to easily navigate the site and find sought-after information.
Navigation and taxonomy testing is generally easy for marketers to complete on their own, but tools such as Optimal Suite's Treejack make setting up and running tests easy, quick and relatively inexpensive.
What impression is my site leaving?
Moderated user testing is a methodology - executed by interviewing users with a task-based script - that provides conversational, qualitative data on how end users feel about a website. Although marketers can implement this test at any time, it is especially useful to test design and UX directions before building the website.
Moderated user testing is helpful when marketers are considering multiple design options and want to determine which option will best resonate with end users. This test was used on an NFL.com club site to gather fans' preferences, which allowed us to confidently decide on a visual design and content strategy for the homepage before designing the interior pages.
This test can be as simple as informal, one-on-one conversations with end users. Marketers can also formalize this test by inviting users to a testing lab to observe large-scale user interactions.
Does my site get the job done?
Another methodology, unmoderated task-based testing, allows marketers to ensure users can easily find information on the website by having them complete various tasks - such as finding a certain practice area or the "contact us" page - and collecting data to analyze how they accomplished these tasks. Marketers should use this test to identify the most pressing issues to correct before a website launch.
Unmoderated task-based testing can be done remotely, making it an easier test for marketers to complete in tandem with a busy workload.
Though it requires significant planning - including recruiting test participants and finding an (often expensive) testing tool to analyze the data - unmoderated task-based testing is one of the most useful and common types of testing because it provides marketers with rich, qualitative information, and at scale can also provide quantitative data.
Is my site ready to launch?
An important question marketers must ask before launching their website is if it's truly ready for prime time. Marketers can overlook bugs or errors that could have been easily fixed before launch. Marketers should use fit and finish testing to clean up any lingering errors and ensure the website is ready to go live.
When One North redesigned Plante Moran's website, we wanted to confirm that users could actually complete the top 10 objectives we designed the site to accomplish.
Using fit and finish testing before the launch, we found a handful of important, low-cost issues to fix. In addition, we identified several higher-cost changes that we plotted on a value vs. cost matrix, which helped us develop a plan for future enhancements.
Fit and finish testing can be done informally. For example, we completed the test for Plante Moran's website with 10 of its own employees. Yet it still yields valuable information that can help marketers identify and fix problems and fine-tune website design, leading to a data-backed sense of confidence prior to launching a new site.
Can I optimize the user experience?
Marketers use A/B testing to gather data and make sure they are using the best version of the website. With this methodology, marketers can compare two design elements - e.g., two different sets of fonts or two different images on a page - to see which version performs better. Marketers can carry out A/B testing by asking website visitors to compare two items or using content management systems to launch two pages at the same time and analyze the resulting data.
A/B testing is done after a website's launch to test the efficacy of different options on a live site. Tracking actual metrics - including which label users click on more or which page layout leads to more conversions - allows marketers to select the option that will better help the company accomplish its goals.
This test ranges from very simple, such as changing the color of a website's call to action, to much larger, such as comparing two different landing page structures. Even minor tests can have a large impact and allow for more valuable interactions with users.
Whether a marketer is just starting the planning process for a redesign or seeking to confirm that a recently launched website is performing optimally, user testing can provide marketers with valuable, actionable data. Using the appropriate tests allows marketers and designers to diagnose and fix any problems and ensure their digital platforms drive business success.
