How to Conduct Usability Testing
So you’ve made your unit and functional tests for your product. You then made the code and voila! The test passes, your code coverage is high, and now your code is perfect, right?
Not quite.
Remember that we as developers are building the products for our customers. If they don’t enjoy using our product, then no matter how many automated tests you made for it, they simply wouldn’t accept it.
For this reason, we need to test our product further, this time focusing on how our customers will use it. This is because the tests we made before lack “humanity” so to speak. It cannot test the satisfaction and emotion felt by our customers when using our product.
This process is called usability testing, the process of testing the usability of our product by the user.
What is Usability Testing
Usability testing is the act of testing how simple or usable the design and functionality of the app is to the users. The test is conducted with a group of users who fit the target market or persona of the product. These tests are conducted repeatedly, from the earliest development of the product, until the final release.
By conducting usability testing, we can detect flaws in our design which we might have overlooked. When we pay attention to how the users react and feel about our product, we can gather insight into how well-designed our product is.
With this new knowledge, we can make further improvements to our product to increase its quality, and by doing so we can increase customer satisfaction.
Usability Testing Objectives
To ensure that your usability test is fruitful, you should always aim to achieve these five objectives during the test:
- Find out if testers can complete a task without external aid.
- Note their behavior, expression, or mental state as they attempt to complete tasks.
- Pay extra attention to whether or not the users enjoy using your product.
- Find problems or issues as the user uses the product, and determine the problem’s severity.
- Search for a solution to the problems you find (if any).
By following these objectives, we would be able to gain useful knowledge about issues on our product. We can then improve the product based on the insight we’ve gathered, and guarantee the product (almost) problem-free during the final release.
Usability Testing Methods
There are several methods to conduct usability testing. The method you choose would depend on the product you are working on, and where you currently are in the design process.
In-Person
This is the most formal of the methods. The testing is done live, and requires an adept, empathetic moderator to provide limited guidance to the users, and take note of their experiences when using the product.
Remote
Users are tested in their own preferred location, at their home for example. This is done to get the user’s reaction and behavior in their own environment, in order to produce a more accurate insight.
Guerilla
This involves you or your team going out onto the streets and testing your product on pedestrians or colleagues who are passing by. This process is relatively simpler compared to the other two, as there is no need for a recruitment process. However, it's quite risky since the data gathered might be slightly inaccurate.
Usability Testing Process
Here’s a guideline for the process of your usability testing to attain the best result:
Plan
You and your team need to define what is it that you wish the users to test. It might be a particular feature, component, or the general overall design of your product.
In any case, you then need to make a hypothesis from everything you want to test. This hypothesis will help you to decide the specific aspect you with the users to test.
Afterward, you can decide the method of conducting the test. You should determine if the tests are done live on the scene, remote via an online medium, or by testing your product on passers-by.
Remember to stick to the scope of what you want to test from the beginning to the end, and test each aspect individually. This will aid you and your team to get a better understanding of the overall quality of your design.
Set Tasks
Determine the tasks that the testers need to do throughout the test. Prioritize the most important tasks first, and make sure that you meet the usability test objectives. Define the goals of each task clearly, and realistically.
Limit the instructions you give to the users, and let them play around and grasp an understanding of your product on their own. This is necessary to measure how well-designed your product is, and how easy it is for users to understand it.
Recruit Testers
Gather users to test your product according to your product’s persona. To do this, you can first screen your recruits by conducting a short survey to filter the candidates and select only those who fit your target audience.
To find testers, you can advertise online or offline, and offer incentives to attract people. You can also ask relatives or acquaintances to participate as testers, and so on.
The number of testers is entirely up to you and your team. Generally however, the more testers you have, the more problems you can uncover.
Facilitate and Moderate Tests
Have the test done in a suitable, comfortable environment for the users. Pay attention to the users, and the issues they face. See if they make mistakes when trying to accomplish a task, if they missed something, or if they misunderstood the rules and does the wrong thing.
You can also record the usability test session, so that you and your team can better detect the number of times the user became confused or disgruntled. You should also ask the users to think their thoughts aloud, and talk about how they feel about the product as they go through the test.
Metrics of Measurement
There are two metrics to assess the user’s behavior:
Quantitative
Anything that can be counted. The range of time needed for users to accomplish a task (if succeeded), the rate of successes and failures, number of clicks, number of times the user became confused, and so on.
Qualitative
Everything that cannot be measured quantitatively. The user’s facial reactions, body language, and other emotional or physical responses to stress while undergoing the test. You can also give a post-test survey to measure the user’s satisfaction, and their opinion on the level of effort needed or difficulty faced when using the product.
Rapid Usability Testing
With tight deadlines and many features or user stories to work on, you might be compelled to think that usability testing is of less importance, and can be held off until later.
This is a common mindset due to the length of time it requires to plan, recruit, and facilitate the test. This issue can be fixed however, by the use of rapid usability testing.
Rapid usability testing is done remotely, to save time on organizing the testing facility for the test environment. The number of testers does not need to be many, 5 testers are enough to uncover most of the design or usability issues in the product.
All you need to conduct the test is a flat design, an online testing tool, a research method, and the research analysis. If all these requirements are fulfilled, we can gather insight into our product within minutes, even if your design team is relatively small.
Below is a couple of research methods for rapid usability testing:
Question Test
Provide a questionnaire after the test completes. The questions can be used to gather both quantitative and qualitative data. Ask for feedback on which content in your product is best, and which content is worst. Also ask what the users would do to improve the current design, if any.
5-Second Test
This test is done by showing the testers your design four 5 seconds, and afterwards evaluating the simplicity and readability of your product’s design. Ask the users what element they most focus on, you could also list a couple of elements in your UI and ask if the users missed them or not.
Navigation Test
For this test, our aim is to determine whether or not the user can successfully transition between screens or pages. The tasks for this test can either be single or multi-stepped scenarios.
You can also ask follow up questions such as why the user clicked a certain button or link, or how the user imagines the steps they need to take in order to complete the task.
Click Test
In a click test, the user will be measured on how they can complete a certain task, and through the task we will look for design feedbacks or certain criteria needed so that the user can do the task successfully.
After upholding a click test, we can gain insight such as the success rate for users, the time it takes to complete the task, and qualitative feedback.
Preference Test
These tests help us decide which design users prefer, it can be useful if the team is deciding between two or more design styles. The test is used to determine which design the app should have, and to compare and validate between one the design idea with another.
A/B Test
In an A/B test, the users are divided into two or more groups. Each group will be given a different design of the same product. Then, the results of the groups are compared. We can then see which group succeeds more, is more satisfied, and is less confused by the design.
The final design of the product is the design of the group with the best test results.
Test Results
After the testers have completed the test, we can begin analyzing the result and creating a report. We can start by looking back at video recordings of the test, and review the data gathered throughout the test. This process should be done collectively, involving the entire team.
The test report should include:
- Title of Usability Test
- Research Method
- Research Assumptions or Hypothesis
- Test Results (Success or Fail)
- Design Recommendation
Conclusion
Usability testing is the best way to measure how the users would feel and behave when using your product. It’s a necessary step to take if we wish to create software that satisfies our customers, since after all, creating software for our customers is our goal from the very beginning.
By conducting usability testing throughout your development cycle, you are guaranteed to make a well-designed and enjoyable software, ensuring that your users are happy, and satisfied.
