Observe the user in real conditions
User testing allows you to evaluate the user experience of a digital product (website, mobile app, business solution) by observing how users accomplish tasks. It is part of the UX design process.
UX design: allowing users to easily find what they need
User experience, also known as UX design, is the way a user experiences their interaction with a website. In the context of a web project with Drupal, a successful user experience implies a precise understanding of the users' needs.
For example, what does a user visiting their regional transport network’s website want? To calculate a route? To get information about subscription prices? To find out about tomorrow’s strikes? All of the above at once? These are all features that will need to be integrated intelligently. Because beyond listing users’ needs and making those features available, much of the UX designer’s job consists of creating an intuitive website interface: easy, enjoyable and fast to use.
If the features are available but require filling out a ten-step registration form and the “Pay my subscription” button is nowhere to be found, chances are people will choose to take their car instead. And the disappointments users experience on a site are often opportunities for its competitors to create satisfaction.
In practice, the problems are usually more subtle than this, but you get the idea.
Giving a voice to those concerned
The UX design field develops numerous tools and methodologies. A lengthy process characterized by the central role of the user in all decisions. Among these methods, one can use user research (user interviews, questionnaires, focus groups). It is also recommended to conduct user testing to verify the decisions and trade-offs made during the design phase.
A user test consists of putting several users in the real context of website use (in front of prototypes). Then, observing them as they perform specific tasks in order to identify blocking points. In addition, the test aims to encourage users to verbalize their thoughts. The idea is simple: the more feedback you get, the more you can adjust the site to improve user journeys.
These tests generally take place when UX designers already have a good idea of the future layout of the site: they have identified user needs and produced mockups (wireframes). User testing allows them to evaluate their work and make the necessary adjustments.
A roadmap for conducting user tests
Upstream of the tests, a UX designer will write a test scenario, meaning a program outlining how the interview will unfold. They set the objectives of the interview, the questions to ask, the tasks to complete, and the success criteria.
Generally, the steps of a user test are as follows:
- Welcoming the participant
- Explaining the stakes of the user test
- Presenting the outline of the test
- Questions about the participant’s profile: age, place of residence, occupation, connection to the tested site or brand, etc.
- Page usability: does the user understand the site’s value proposition, the meaning of tabs and buttons? Can they identify the goals to be achieved on each page?
- Performance test of one or more user journeys: the user is encouraged to complete a series of tasks, for example, “you want to add an item to your cart and find the nearest point of sale to your location.”
- Debrief
At the end of this process, the UX designer prepares a user test report. It highlights user feedback on usability, tasks completed successfully, those that encountered more difficulties and why... This deliverable will be handed over to the client and will make it possible to make adjustments to the site’s mockups.
For more information about our methodology, click here.
Moderated testing and unmoderated testing
There are two types of user tests: moderated and unmoderated tests.
Moderated tests can take place via videoconference or face-to-face, although videoconferencing is often preferred because it is less restrictive. They involve the presence of a moderator who guides and observes users as they interact with the website.
Unmoderated tests, on the other hand, take place without the direct presence of a moderator. Users receive tasks to complete and are asked to provide feedback, often through questionnaires, video recordings, or written comments.
To help you choose, here’s a quick overview of how these tests work as well as the advantages and disadvantages of both methods.
Moderated tests
Moderated testing is an evaluation method led by a moderator who guides participants without influencing their choices. This approach is preferred for several reasons:
- Generation of precise feedback and opinions:
The presence of a moderator allows for asking questions and obtaining a lot of feedback. For this, semi-open questions are generally used, offering users several response options to answer more precisely: feasible, somewhat feasible, complicated, not feasible. Open-ended questions are also used, such as: “What do you prefer?” This type of question gives participants the opportunity to express themselves spontaneously, using their own vocabulary and providing a more natural response. - Cost: Moderated tests can be less expensive than more in-depth or complex tests. This saves time and resources.
- Time: They take less time to conduct, which is crucial when quick results are needed.
- Simplicity: They can be simpler to implement, requiring fewer technical resources or specialized knowledge.
But these tests also have their limitations. Depending on the needs, time constraints, resources, and required precision, choosing moderated tests can be wise or, conversely, limiting.
- Potential biases: There is a risk of cognitive bias, as users may want to respond favorably in the presence of the UX Designer. Users may sometimes behave differently when a moderator is present, which can alter how they interact with the product or interface.
- Limited depth: They might not explore specific issues or areas in depth, leaving gaps or blind spots in the evaluation.
- Availability constraints: Moderated tests require the user’s live participation, which can make it challenging to find time slots that work for everyone.
Unmoderated tests
Unmoderated tests are not to be outdone! They offer:
- Great time flexibility: Users can take the test at their own pace, which broadens participant availability.
- Low cost: They require less human resources for direct supervision.
- Greater freedom of expression: Users feel more comfortable expressing their opinions without the pressure of being watched live.
- No moderator influence: User feedback can be more spontaneous and less biased due to the absence of a moderator.
However, these tests also have disadvantages, including:
- Less control: It is difficult to ensure that users follow instructions exactly or clearly express their thoughts without direct guidance.
- Difficulty interpreting some feedback: Without a moderator present to ask follow-up questions, user responses can sometimes be vague, making them harder to interpret and increasing the risk of bias in the results.
The little surprises of a user testing session
When a moderated user testing session begins, it's crucial for participants to share their screen so we can follow their actions on the tested site. This allows us to understand their journey, the buttons they use, and whether they backtrack. But sometimes, they find themselves unsure about how to present their screen. This can waste time during the interview, or even abruptly interrupt the test.
To avoid these pitfalls, we recommend sending a brief guide by email before the session, explaining how to share the screen and check browser compatibility.
Of course, in the grand adventure of user testing, other mishaps can occur: audio that suddenly goes silent, the mic on strike, or even the participant simply forgetting their test session. That's the magic of the unexpected!
If you’re interested in support with UX design, contact us to discuss your needs!
Keep reading:
- Design Sprint: 5 Days to Create the UX of Your Project - 11/29/2023
- User Research During a Drupal Project Redesign - 06/16/2023
- User Testing... Why and How? - 02/19/2019