Research Research
Usability Test: How To Prepare And Conduct One?
Content list
Research Research

Usability Test: How To Prepare And Conduct One?

Capa de Artigo Teste de Usabilidade

UX Designers share a common understanding that they should not assume that the user has the same behavior as they do when interacting with a product. After all, you are not your user. And that's why it's so important to conduct a usability test to collect information about how the user interacts with your product.

It might sound simple, but planning, preparing, and conducting a test is a complex process that requires attention. You will only see results if the tests are done properly.

This article will show you how to prepare for a usability test and which techniques you can use to avoid any biases!

What is a usability test?

The usability test is a common research method in UX Design. Basically, it is used to understand how users behave while interacting with your product.

Thus, usability tests have 3 main goals:

  • Identify problems in the UI and product design;
  • Discover opportunities to improve the product;
  • To learn more about the user, their behavior, and preferences.

Why should you run a usability test?

There are some techniques for evaluating a UI without necessarily testing it with users. Heuristic evaluation is one of these techniques and is quite common.

However, there are situations where you cannot predict how the user will behave when interacting with your product.

A usability test is the most efficient way to understand and identify users' behavior toward your product or UI.

There are key moments during the tests; for example, when users try to perform a certain task, this is an excellent opportunity for the mediators to see how they can improve their products and optimize the user experience.

Furthermore, there are 3 basic reasons why you should conduct usability tests:

  1. You are not your user: that is, there is no way you can predict exactly what the user's behavior will be on the interface without a usability test;
  2. It's a cheap method: of course, there are several other research methods, but the usability test is certainly the cheapest and quickest and can be carried out in just a few days;
  3. It is a basis for argumentation: once performed, the usability test provides data and facts that can be used to convince stakeholders of the best course of action for the project.

Reading tip: Heurísticas de Nielsen – 10 Dicas para melhorar a Usabilidade de sua Interface

How to prepare for a usability test?

Although quick and low-cost, usability tests should be properly executed in order to fulfill their objectives.

To do this, it is essential to be well-prepared before applying it. Getting ready beforehand is essential for the success of your test.

We will show you a short step-by-step guide to help you prepare for your usability test!

1) Define the test objectives

First of all, it is important to know why you want to perform a usability test. What is the goal?

To define the goal, you need to evaluate the current state of your UX design project:

  • If you do not yet have a UI, you may want to perform the test with competing products. This way, you want to evaluate what features and other design issues work with similar products;
  • If you already have a UI, it may be time for you to test your product to gain insights. Then you will have a direction on what development improvements are needed;
  • Or, you may already have a released product and want to re-evaluate some features and understand what needs to be updated or not.

Next, you need to understand what exactly you need to be tested:

  • What information do you consider important to collect;
  • What problems do you want to address?

With this information, you can understand exactly what will be tested and have a few hypotheses, so you can build your scenarios, tasks, and scripts more efficiently.

2) Choose which type of usability test you will use

There are different types of usability testing, each with its own characteristics, objectives, pros, and cons.

Qualitative or Quantitative

The first choice that must be made is between qualitative and quantitative tests:

  • Qualitative: intended to find information about how users use and behave with your product. The qualitative type of testing is well suited to find problems during the User Journey;
  • Quantitative: to collect quantitative information about the user experience. Time and success rate are some examples of observable metrics.

Remote or In person

Imagem ilustrando teste de usabilidade remoto

Another choice regarding the type of usability test is the geographical location of the participants.

The tests can happen in person or remotely.

In-person testing requires the moderator and the user to be physically in the same place. On the other hand, remote testing is done with the mediator and the user in different locations.

Although it may seem like little difference, face-to-face tests tend to collect much more information than remote tests. This happens because the observation of the mediator is extremely important. They can analyze the user's body expression and behavior, which is a bit more challenging to capture in a remote session.

Reading tip: Remote Work: What is the Outlook During and After the Pandemic?

Moderated or unmoderated

A moderated usability test is administered by one person, a researcher or a UX Designer, who can answer and ask questions to the user. In addition, they introduce the test to the participant and give explanations of how it will work.

On the other hand, an unmoderated test is one that does not have a person to supervise and follow the user. Generally, in this type of test, the user will perform the tasks without asking or being asked any questions during the session.

It is important to say that a moderated test brings more insights and information to the research. Because the researcher-participant interaction allows deeper exploration of the user's questions and doubts, on the other hand, it costs more than an unmoderated session.

The types of usability testing can be combined. You can opt for a qualitative, face-to-face, moderated test; or a remote, non-moderated qualitative test.

3) Choose participants

The next step is about finding the participants to apply the usability test.

This phase can be a bit more complicated and time-consuming. You want to be careful and assertive when choosing the participants.

It is simpler to recruit participants if your usability test is about an existing and already released product. You can simply call users of your product.

But, if it is for a product launch, you need to be certain of who your users are and have an established user persona.

The most important thing is that the participant reflects your real users or the best representation of them. This way, the results, information, and data collected will be as assertive as possible.

How many participants?

At a first glance, we may believe that the more participants in the test, the better.

But the truth is not quite like that.

The basic behaviors of your users tend to be similar. So a larger number of participants will only bring repetition of results and no more relevant data than if it was done with fewer people.

Therefore, the efficient number to apply the usability test is 5 people! More than that and you will only waste time and money.

But remember that the choice and the definition of your user profile or persona is the key to a successful result.

Reading tip: How To Create Valuable User Personas That Will Actually Matter?

4) Prepare a test script

The script is a crucial part of usability testing. With it, you will be able to guide the user through the interaction, narrating the journey, requesting actions, and gathering feedback.

Let's go through some tips for preparing a great usability test script!

Tips for preparing a usability test script

1) Plan ahead

To create a usability test script, it is important to plan ahead. Remember the goal of your test and keep that in mind when creating your script.

How will the questions in the script help achieve the test objective?

Then define who the participants will be, where the test will take place, and what kind of test it will be, as mentioned above.

From there, you can start writing the questions and user actions that will be part of your script.

2) Record

Imagem ilustrando gravação de teste

We also recommend that you record your test for further analysis. Of course, you will note insights during the test, but it is not always possible to capture everything at the moment.

But remember to have the participant's consent to record the usability test. Don't make it sound like an interrogation with the camera pointing at their face. Otherwise, it can be difficult to get them to behave naturally. After all, the main purpose of a usability test is to understand how the user would normally act with the interface.

3) Make the participant feel comfortable and encourage them to provide feedback

It is important to try to make the participant as comfortable as possible, for the reason we have just pointed out. However, it is common for the user to get the feeling that they are being tested and not the product.

Therefore, do not ask questions that might make them feel uncomfortable and unable to do the usability test.

Instead, encourage them to ask questions and ask them to narrate their thoughts during the navigation tasks. Knowing what the user is thinking is important to help understand their behavior in front of the UI.

Reading tip: The Importance of Psychology in UX Design

4) Be clear and concise

Write your test questions or tasks clearly and in a way that is not open to misinterpretation or doubt.

That way, you can ensure that the user understands exactly what they have to do or what is being asked.

For example, if a login is required before accessing the platform. Make this task clear: "Log in with your Google/Facebook account to access the platform."

Before applying the usability test, it is advisable to ask someone from your team to review the questions and tasks to understand if they are clear or not.

5) Create detailed questions with no bias

Okay, this tip seems a bit complicated, but let's explain it properly and clarify it!

To ensure good information from the user's answers, it is necessary to create more detailed questions, for example:

  • Fewer details: how often do you visit this site?
  • More details: How often have you visited this site in the last 6 months?

However, it is important to remember that the participant will always think there is a right way or a right answer to be given. Thus, be cautious when choosing the words to avoid as much bias or induction as possible. For example:

  • "Now, try to find a way to carry out a more detailed survey."
  • "Now, do the same search but with more details, using filters."

Notice that the first sentence implies that the task is difficult to do. So think about the best use of words when writing questions and tasks.

6) Break a complex task into simple tasks

When proposing a long and complex task, the participant may get lost and forget the action that needs to be done.

Instead, break the complex action into several other simpler actions. For example:

"Create a project, invite your team members, and schedule the stakeholder meetings."

The above action is quite complex and extensive, so try to break it down into smaller tasks to avoid overwhelming the user:

  • Create a project;
  • Invite your team;
  • Schedule alignment meetings.

This way, you ensure that the user will not get lost and will complete each of the tasks and actions proposed by your usability test.

7) Create realistic tasks

Have the participant take actions that they would normally take in their daily life. For example:

On the website, find and buy a black sweatshirt, size M.

By indicating this action to the user, you completely disregard their reality and make them just do the task without interacting with the site.

Instead, the task could be as follows:

On the website, search for a product that you like and that fits your budget.

This way, you give your participant the freedom to interact with the site and associate this task with their reality.

8) Create Actionable Tasks

Actionable tasks are those that tell the user what to do instead of asking how they would do it.

Whenever the user is asked HOW they would do something, they tend to answer and not act. And most of the time, people do not detail in words all their actions.

So concentrate on writing tasks that indicate what should be done rather than how they should be done. For example:

  • Avoid: Enter the website and show me how you would schedule a dinner party for a Saturday night;
  • Do: Go to the website, find a restaurant of your choice, and reserve a table for the weekend.

9) Don't give hints or too many descriptions of the task

When writing an assignment, avoid describing how the user should do it. As that could influence the participant by clues hidden in the speech. For example:

"Log in, click the menu button, then 'add friend' and look for your friend in the contact list."

With the task description above, you have pretty much said what you expect the user to do. Instead, be more succinct:

"Log in and add one of your best friends to the platform."

This way, you can observe the user's behavior when performing the task.

So be careful when writing tasks. Do not influence the participant's actions. Otherwise, you might hinder the usability test.

Reading tip: Outcomes and Outputs: Know the Difference

Conducting the test

Imagem ilustrando condução, um maestro e sua batuta

After preparing the script comes the application of the usability test.

But, although this may seem simple once you have the script ready, you still need to pay attention when conducting the test.

If you are not prepared to apply the test, the user can put you in difficult situations with unexpected questions.

In addition, it is quite common to become anxious or nervous, to talk more than observe, or to remain completely silent.

All three situations are not favorable for your test and can influence the results. But there are some techniques to avoid these scenarios during the usability test application.

Echo

This technique consists of repeating the participant's question to prompt them to bring more information or to rephrase the question.

Using the same words as the participant removes any bias you may have when interacting with the user for more information. Here are two examples of this technique:

  • User: Those buttons are confusing, I'm not sure….
  • Mediator: Are the buttons confusing?

However, pay attention to the tone at the time of your question so that you don't intimidate or give the user the feeling of judgment.

Boomerang

The boomerang technique consists of returning the question to the user. Questions like "what do you think?" or "what would you normally do?" are good examples of this technique.

Here are some examples:

  • User: Do I have to register now?
  • Mediator: What do you think? What would you do at home?

Again, be careful with the tone of your question so as not to give the feeling of judgment.

Columbo

This technique consists in getting more information out of the user by only saying a few words or incomplete sentences in order to make him better explain what he means.

See some examples:

  • User: If I click here, do I lose my job?
  • Mediator: Hm, are you thinking that… if you…?
  • User: I'm not sure which button I should click, I don't think I can tell the difference between these 3 buttons.

Notice how you have not influenced the user with your question; instead, you have prompted them to further develop their own question.

The columbo method is more sophisticated and difficult to apply. It takes a lot of training and practice to be able to use this technique naturally.

Be careful! A/B test is not a usability test!

It is quite common to confuse usability testing with A/B testing. But don't fall for it! The two types of testing have different objectives, and one does not replace the other.

While usability testing will observe user behavior when using your product, A/B testing is a tool to verify which version of a function creates more conversions.

Because of this, A/B testing cannot be considered usability testing because it only collects binary information. If something works better than another. That's it.

With A/B testing, you cannot recognize other difficulties and problems that the user may encounter during the interaction and the journey. Moreover, you won't know what motivated the user to choose one over the other, and often this information is indispensable.

Reading tips and final thoughts

A usability test is very useful and requires a lot of study and practice. Because of this, we recommend a read for you to improve your knowledge on the subject: Rocket Surgery Made Easy: The Do-It-Yourself Guide to Finding and Fixing Usability Problems, by Steve Krug.

Don't forget to like and share if you enjoyed this content! This small gesture helps us a lot! Feel free to continue browsing, and if you'd like to stay up-to-date, sign up for our newsletter!


We are proud to have people from our community hired every month by great companies, in countries such as
Brazil, USA, UK, Ireland, Germany, Netherlands, Spain, Portugal, Austria, Czech Rep., New Zealand, and Canada.

Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro
Parceiro Parceiro