There are so many assumptions made during digital strategy workshops and brainstorms. The problem is assumptions are just that; assumptions.
When you set up your digital strategy, you have to make choices.
How should we organise our content and structure to help people find information? Should the mobile menu be a hamburger, or appear on swipe? Do we need to have an ‘About Us’ section on the landing page, or does it detract from the offer?
Everyone on your team likely has a gut instinct about what the answers to these questions are, but getting it wrong could have a big impact on the success of your campaign or website.
If you’re investing in your website or campaign, do you really want to let your (or your boss’) gut decide the big questions?
User testing answers those tricky questions that your gut can’t be trusted to figure out.
User testing takes your gut instinct and puts it to the test with your real-life target audience. It validates and confirms business decisions, and shows you what might be going wrong with your existing performance. You can understand the needs, motivations and behaviours of your users by physically observing them use your website. It’s super important to design your new site with your target audiences in mind, and user testing can show you exactly what they want.
User data takes the guesswork out of business decisions. But how does it work?
User testing is all about gaining an insight into the mental models of the user.
We need to find out what the user believes about the system they are using – believe being the operative word here. Their knowledge of your product is, at this point, based on beliefs, not fact. A user’s mental model is always changing, and can be influenced by a series of external factors, such as their environment, coworkers, or even applying lessons they’ve learned from using other systems. Our goal is to gain insight into the user’s mental models while accessing the site and learn how we can best solve their problems.
Quantitative and qualitative data: The perfect couple
When collecting user data, we always need a good mix of quantitative and qualitative feedback. Quantitative feedback will give you performance metrics on task achievement rates, task completion times and other useful numerical data. Qualitative and verbatim feedback will provide a deeper insight into the subjective, emotional-based response of the user; the way your website is making them feel.
User testing can be done in person or via phone or video or online link. While you can still uncover valuable insights remotely, it’s best to do user testing in person to get a real understanding of the user’s mindset while attempting a task. The type of testing to be conducted will all depend on the time and budget you have available, as well as your objectives.
A few things to prepare ahead of a user testing session:
- Recruit the right participants.
- The participants need to reflect the website’s target audience. Testing with the wrong people can send you down the wrong path, which can mean a huge waste of time and money.
- Create the test script.
- Outline the approach to your testing, including objectives, scenarios and questions for participants. Use your script as a guide to prompt investigation and get an understanding of how often and how common a problem may be.
- Prepare your facilities.
- Secure a quiet room away from noise and distractions. Set up the tech hardware that you’ll need, such as a laptop or mobile devices, with prototypes ready to go, as well as some recording and note-taking software to record the session for analysis later.
Here are five user testing techniques that give you actionable insights to improve your site.
Card sorting is a method that asks users to organise items into groups and categories. There is no right or wrong answer to these tests; it’s all about what makes sense to the user. This method helps to create or refine information architecture. Card sorting gives us insight into the users’ mental models when accessing website content, and helps us to understand how the various buyer personas expect to see website content organised.
Tree jack testing, or reverse card sorting
A reverse card sort, or a tree jack test, helps us to test a proposed site architecture. We ask users to navigate through a hierarchy of categorised cards to find the card that best suits the task. This type of test helps us to determine the effectiveness of labelling taxonomies and the proposed grouping of information. You can test a few variants of proposed information architecture to get insight into which is best for the user.
A product walkthrough is done through observations and interviews in the user’s’ home or workplace where they are most likely to interact with the product or service. We can watch them use the product or service from first opening, setting up and onboarding all the way through to ongoing use and reuse. This type of test gives us a great understanding of the user’s behaviours, needs, expectations and frustrations in their real-world environment.
Concept testing with low fidelity prototypes
A low fidelity concept test is used to present a rough outline of a prototype to a user. This helps to validate the concept early on before we get too far down the development path. Using this test, we can sense check the logic and flow, as well as the proposed features and content in a design. The results will be high-level feedback and discussion on user expectations.
Usability testing with high fidelity prototypes
A high fidelity prototype more closely resembles a finished product. A high-fidelity prototype would include clickable links or buttons, a realistic visual presentation with graphics, spacing and layout, and may also include any content that would appear in the final design. This type of test shows you how usable a system is by observing users’ actual behaviour and mental models. Observations you take in this test are more likely to be accurate behavioural feedback, rather than verbal feedback on how a person feels.
A/B or split testing
User testing doesn’t end once you have a finished product. A/B or split testing is a way to compare two versions of a single variable by splitting an audience between them and measuring which version performs better. Instead of using testing to decide how something should be built, you use it to test hypotheses about what might work better. It means you can test and optimise a variety of options on an ongoing basis, rather than setting and forgetting.
User testing helps you create an experience that really understands your audience.
User testing helps identify website features that are both usable and useful for your target audiences. When building a new site, we need to remember that we are building for a real, human audience, with real, human emotions; leaving them out of the equation won’t help.
Understanding what the user needs and wants out of your website saves money and time. You’ll avoid accidentally developing features or content that are frustrating, unreliable or unhelpful. The data you uncover helps you and your team to make decisions based on evidence, not gut instinct. Not only will you validate your assumptions, you’ll also be creating the user experience your customers demand. And a happy customer is a valuable customer.