Usability Testing: What It Is and How to Do It

If you're looking for one thing that can make your product more popular, desirable, and financially successful, it could be usability testing. If you’re not doing usability studies, you’re missing out on a lot.

Usability Testing: What It Is and How to Do It

Want to know why? Let us explain.

What Is Usability Testing?

Usability testing is a technique used for evaluating and subsequently improving your product's design by observing how real users interact with it.

In order to conduct usability testing, let's first establish what usability is. According to the Nielsen Norman Group, there are five main components of usability:

  1. Learnability: how quickly users can learn to use your product
  2. Efficiency: how quickly users can perform their tasks using your product, i.e. the ease of use
  3. Memorability: how easily people can get back to using your product after not using it for a while
  4. Errors: how many errors people make and how easily they can continue their work after those errors
  5. Satisfaction: how satisfying the user experience is

Although there are many research methods for evaluating usability, like A/B testing and heuristic usability evaluation, the most basic and common one is usability testing.

During usability studies, you observe how end-users from your target audience perform a series of tasks associated with your product in order to uncover usability problems within your designs.

Keep in mind that you don't have to test a finished product with users. You can start test sessions even at the prototyping stage of your UX design process.

You can test your projects either in a lab or in remote usability testing conducted online, such as that offered through the Teston user testing platform.

But before we dive into the process behind usability testing, let's talk about some of the benefits of usability testing.

The Benefits of Usability Testing

Observing how real people interact with your designs is a very insightful and often enlightening process.

Often product developers spend a lot of time using their own products. They get so used to the user interface, features, and menus, that they start neglecting certain problems associated with their design.

Performing certain tasks may come easily to the creators, but new users may be confused and even dissatisfied when trying to complete tasks. This is where usability testing comes into play.

Well-crafted usability testing allows us to see:

  • The problems with our current user interface and how we can fix them
  • The key areas for product improvement
  • The users' satisfaction with our products

We're long past times when people had only a handful of web pages and services they could use on the internet to achieve their goals.

With plenty of options these days, 79% of users will search for another site to complete the task if the content on yours is not optimised. Additionally, 88% of consumers online are less likely to return to a website after a bad experience.

If we look at mobile apps, 90% of users stop using an app due to poor performance.

Improved usability also makes a strong case from a business perspective. The improvement in usability metrics increases website traffic and conversion rate.

From a business perspective, improved usability benefits include increased employee productivity and decreased support costs.

Overall, products with greater usability are more profitable. Put simply, usability testing makes you more money and saves you more money.

A successful and productive usability testing session with users requires a well-thought out test plan, a properly facilitated testing session, and a bias-free interpretation of the results. We'll talk in detail about each of these in the following sections.

How To Effectively Plan a Usability Testing Session

If you want your usability studies to be productive, you need to be prepared. You can't just invite people to do something with your product. An efficient usability test plan includes:

  • Scope: Define what you are going to test — a home page, an e-commerce website, a prototype, or a specific section of your user interface.
  • Goal: What are the main questions you want to answer with this test session? For example, "Why don't people use this new feature?" or "Why are there so many support calls regarding our checkout process?"

You don't have to be very specific, but it's important to have an overall goal for your usability studies.

  • Target users description: You need to recruit participants for your studies that represent the end-users of your product. Otherwise, some usability findings may be irrelevant.

Consider creating user personas at this stage to better understand what kind of test participants will be of the most value to you.

For example, at Teston you can recruit target users for your usability testing based on a wide range of criteria, such as gender, occupation, age, etc.

You can recruit users from different cities and countries, and even get test results in specific language in case you’re developing products for a defined local market.

  • Tasks: Design real-world tasks that participants will complete during the testing session. It's an art in and of itself to construct tasks that invoke natural behaviour from participants as they follow goals, so we'll talk more about that in a bit.
  • Success Metrics: In order to compare participant's results, analyse testing results, and track usability improvements in subsequent testing sessions, you need to decide what metrics you'll be following while conducting a usability study.

The most common usability metrics include task success rate, task required time, and users' subjective satisfaction.

At Teston, we simplify the whole process of creating a test plan for your studies. We help you design test tasks to gain deep insights into how people interact with your designs, and recruit the most appropriate target audience for your tests.

The final product of your usability testing preparation stage will be a usability testing script. It consists of a set of tasks that you ask users to perform within a given interface.

A well-designed task is realistic, actionable, and doesn’t hint at any specific steps the user should take in order to complete it.

Think about tasks in the context of user goals. Remember, people don't come to your website to complete tasks — they come to solve specific problems and achieve certain goals.

For example, a user goal might be to buy a new washing machine on a low budget.

An example of a poorly designed task for this goal is, "Can you go to our website and put a 'Samsung' washing machine in your cart that costs less than $500 using our price filter?"

A well-designed task might read, "You have a low budget. Use our website to buy a washing machine."

When designing your test tasks to reflect real-world user problems you'll see the entire user journey from the very beginning (entering your website) to the very end (purchasing a washing machine), and you won't give any hints on how to navigate your UI, use filters, or manage a shopping cart.

If there are any usability problems with any of those sections, the test will be able to reveal them.

At Teston, we provide you with hints and templates to help you design test tasks that will allow you to gain deep insights into how people interact with your product so you can find areas for improvement.

After you created a script with a set of tasks, it's a good idea to pilot test it.

Try it out with at least one participant or a colleague, or at least perform a walkthrough yourself. That way, you'll be able to test the clarity of your script and how well users understand the tasks.

This is especially valuable if you perform unmoderated usability testing when no facilitator is present while users are performing their tasks.

With Teston you’re free to run a pilot testing with just one participant in order to polish things out before running studies at scale. After creating and testing the script, the last step before the testing process starts is to recruit participants for your studies.

You can recruit participants yourself or use online testing services like Teston that conduct the recruiting for you. In the latter case, all you need to do is define the target audience criteria, including gender, age, occupation, income, city, and more.

The number of participants you should conduct user testing with usually varies from three to eight, with five users being the most optimal number, according to the Nielsen Norman Group, one of the pioneers of usability testing.

You designed a testing script. You recruited participants that represent real users. It's time to dive into the action and perform the usability testing itself.

How To Conduct a Usability Testing Session

When conducting a usability survey, there are three main rules to follow. We'll detail each one below.

1. Don't Listen to What People Say — Pay Attention to What They Do

There's an old usability testing rule that when you test your designs with users, there is a big difference between what people say or think about your product and what they actually do.

The information that comes from participants is usually biased for several reasons.

Participants may try to please the facilitator with their answers or try guessing the expected behaviour from their past experiences with somewhat similar projects, but this data is never reliable and has nothing to do with how they will actually use your product.

If you want to gather verbal feedback from your users, do that after they perform tasks, not before or during the tasks.

A great way to reduce the number of user biases is to thoroughly screen test participants before your testing session. Online testing services like Teston perform recruitment screening for you.

2. Don't Ask Leading Questions

Sometimes people who conduct usability studies are prone to asking leading questions. These are questions that already contain the answer within them and often force test participants to act in a certain way.

An example of a leading question is, "I noticed this task was difficult for you, was it?" In this question, the facilitator, noticing that the test user took a long time to complete the task, already implied that the task was difficult.

However, the real reason could be that the participant simply observed other options or misunderstood the task. Even experienced usability experts are sometimes inclined to ask leading questions to validate their own theories, so watch out for that.

3. Ask Open-Ended Questions

If you want to get richer data from participants, instead of asking leading questions or closed-ended questions that can be answered with a simple "Yes" or "No", ask open-ended questions.

Here are some examples of open-ended questions: "How useful do you find this feature?" or "How would this feature fit into your workflow?"

Those questions don't imply anything or force a specific response from a participant, and instead, allow you to gain additional insight into what people think about your product.

After you're through with the usability session, it's time to analyse and present the results.

How To Analyse and Present Usability Testing Results

The usability testing session has less value if you can't present its results to the key stakeholders, team members, or clients in a simple and efficient manner.

In order to effectively present the results of your usability studies, you have to structure all the insights and observations you've made.

The very first distinction you can make between different findings is how severe the problems you revealed are. Typically, there are three levels of severity of usability issues:

  1. Critical issues: those have to be fixed ASAP, otherwise users won't be able to complete their tasks
  2. Serious Issues: issues that, if not fixed, will make users feel frustrated, which may negatively affect conversion and drop off rates
  3. Minor issues: low-priority issues that can annoy users but won't affect their completion rates as much and can be fixed later

Another way to structure your usability findings is to apply specific usability metrics to the tasks that participants completed during the test session.

Those metrics include success rates, task time, error rates, and satisfaction questionnaire ratings if you added questionnaires after the studies.

All those parameters can be measured for every participant. That is why it's especially effective to record usability studies with users — given they provide consent.

Finally, if your observations during the studies are not of a quantitative nature, but rather behavioural, make sure to document those concisely and in detail in your report. Don’t be vague when you report about usability problems.

An example of a vague report statement might be, "couldn't finish the checkout process." A specific and concise report might say, "tried entering card number into the name field."

When you're done analysing and structuring the information you've obtained during usability testing, finish your report with a short summary about the testing methodology you've used, the goal of the usability study, and whether you achieved it.

You can also include information about the participants, and when and where the test was held. You'll need this information for future reference if you're going to repeat usability studies or track your overall UX enhancing progress.

It's Time To Perform Usability Testing

Now that you know why usability testing is important, what its benefits are, and how to conduct usability studies, it's time to put it all into action.

Whether you decide to perform usability studies on-site or use online usability remote testing services, it's crucial to put your product in front of your target users as soon as possible, even if all you have is a raw prototype with a handful of core features.

If you decide to perform unmoderated remote testing with Teston, our group of testers is eager to help make your product better. Time to test!