One of the biggest challenges we face in web development isn’t necessarily how our different clients’ markets and technologies work – it’s how humans work. What users say versus what users do are two completely different things, and the only way to uncover the truth is to test. User testing is more than just a check list of product requirements – it’s the most convincing support for your design and development decisions.
Since you’re going to optimize the results of your usability testing, you’ll need to consider the following:
- Identify Goals: What are you testing for? Why are you testing? You should always know what you want to get out of your usability research.
- Know Your Audience: Who are your targeted users? What distinguishes them? Which users are currently a top priority for you, and why?
- Goal Metrics: How do / will you measure performance and success of the website? Do you have data that needs explanation, or that highlights possible issues within the current user experience?
- Competition: Who are your closest competitors, and how do you compare? What behaviors or expectations might your users bring to your online presence based on their experiences with the websites of your competition?
- Timing & Scope: What time frame are you working with for collecting your data? When is it due?
- Impact to Business: What will you do with the research results?
Understanding the Methods
Once you have a plan in place, you’ll need to decide which type of test(s) will be the most beneficial. The more concrete your usability goals are, the easier the selection process will be. From a web design and development perspective, there are a variety of tests you can choose from. But here a few we use most frequently:
- Online User Tests (scripted): Online user testing is the most controlled test, and is usually recommended for testing specific aspects or use cases. An example may include whether or not a user can find or access a certain feature on the website and how long it takes to do so. These are generally scripted tests, meaning participants are given a set of questions asking them to complete certain tasks. Their sessions are recorded as they do so, and they’re encouraged to verbally express their thoughts. Recruiting is necessary in order to execute a scripted test. It’s important to have a game plan for gathering participants.
- User Recordings (unscripted): Recordings reveal how people use your site in the most natural setting, even if they browse multiple pages. Tools are available to record page visits, mouse movements, mouse clicks, scroll movements, key strokes (excludes passwords and numbers that resemble credit cards), and browser viewport size changes.
- Heat Maps: Heat mapping will help you determine where your visitors are clicking or tapping (if on mobile). They quickly help uncover issues with your page – for example, are visitors clicking your links? Or are they clicking on areas that aren’t links?
Heat maps can also track mouse movement and scroll depth on individual pages.
- Card Sorting: A simple, yet powerful approach, this is accomplished by writing different elements of your site on note cards or Post-It notes, then having participants organize them in a way that makes the most sense to them. If you’d like to go paperless, you can use a testing tool like OptimalSort for quick analysis of common groupings. Card sorting mostly deals with issues of navigation, organization, labeling, and grouping. This is an excellent method to understanding your site’s information architecture. Card sorting also requires recruitment.
- A/B Testing: Also known as split testing, this method compares two versions of a web page or a specific element on a page to see which performs better. The variations are served up randomly to users who enter the site, and their engagement with each experience is measured and collected. It is essentially an experiment to determine which variation performs better for a given conversion or goal on the website, and then implementing the winning variation.
- Exit Polls: Exit polls are a great way to find out if a user’s visit to your site was successful. We recommend asking questions such as what was the purpose of their visit, if they were able to accomplish the goal of their visit, and if not, why not.
- User Survey: What’s the difference between a survey and a poll? Surveys allow you to ask more questions that are in depth about what made a visitor choose you, what they like/dislike about your site, who they really are, etc. Surveys typically require more time from your users than a quick exit poll, so it’s important to set that expectation when inviting them to participate.
- Form Testing: If forms are a big concern on your site, consider looking into a tool that enables you to track form fields and their completions. This type of user testing lets you analyze your forms to see exactly where visitors are dropping off, and identifies any fields that are confusing or a nuisance to visitors.
Of course, many of these types of user tests can, and should be, used in tandem. This is a great way to collect a plethora of insights regarding numerous aspects of your website, including forms, design, navigation, and more.
Usability testing helps you see the bottom line of whether your design works or doesn’t. Most large organizations (like Amazon) are constantly employing user testing to optimize their sites for their users. Test early, and test often. Every organization is different, so there is no magic usability test that will tell you everything you need to know. So define your hypothesis, try a variety of test methods, and get ready to learn about your website visitors!