Welcome to the sixth article in our user experience series. In our previous post, we discussed some strategies for the design ideation process and how to create interactive prototypes.
This post will look at the next step in the design process, which is usability testing.
In case you missed them, you can find the other articles in our user experience series below:
After creating some interactive prototypes of varying degrees of fidelity based on a team’s ideation process, how do you know whether or not your designs meet the needs of your users? It is now time to test!
Usability testing is about evaluating whether your product or system meets the needs of the users for whom it was originally intended. Remember our users that we were trying to collect feedback from during the user research phase? It is now time to get real people to try out your design and observe whether they can use the product and whether they are satisfied with the end result.
A common misconception regarding usability testing is that a large number of participants are needed to have statistically significant data regarding usability issues. Fortunately, you don’t need to recruit hundreds of participants for this. Studies have shown that only about 5-15 participants are required to identify the vast majority of usability issues. The reason is that the goal of usability testing is behavior-based insights, not statistical validity. For example, if 5 out of 10 participants testing a mobile application could not figure out how to use a crucial feature, that gives your team valuable insight into some problems with your initial design and that there is room for improvement. In fact, the law of diminishing returns indicates that with every subsequent participant past a certain threshold, the value of insights gained from that participant is minimal.
When setting up a usability test, here are some important considerations:
- Pick participants that represent the user base of the product. If an application is intended for high school students, scuba divers, or single parents, — find them.! It is important to inform your participants that you are testing the product, not their intellect or abilities. If participants struggle to complete a task or understand something in the application, reassure them that they are doing a good job and they are not in the wrong. Insights gained are valuable feedback and not intended to judge how capable a participant is.
- Come up with a set of tasks for the participant to perform. Try to keep the task description fairly broad and observe how they achieve their goal. For example, “using the application, find information about a college major that you are interested in and what the tuition and fees are.” When running the test, allow the participants to familiarize themselves with the application before asking them to complete tasks. That way, any biases introduced by the “novelty effect” can be minimized.
- Keep in mind that at this point, your “application” is most likely still a prototype. Therefore, participants may be clicking through some static screens or even using a paper prototype and you may have to manually shuffle screens in front of them to mimic an interaction; This is entirely okay. Again, the goal is to identify glaring usability issues in the product rather than wowing the participant with how advanced your system is.
- Observe a participant’s progress and keep track of any important metrics. If they are struggling to complete a task, it is okay to stop them after a certain amount of time and have them move on to something else. There are a number of metrics (both qualitative and quantitative that can be gathered from a usability test. Some of them include task success or error rate, time of task completion, NASA-TLX (for measuring how demanding a task is), a System Usability Scale (for measuring ease of use and satisfaction), and interviews to gather any other feedback before or after a usability test. Gathering metrics helps stakeholders and the product team as they seek to iterate on a design and re-measure to quantify improvements.
As alluded to above, usability testing aims to identify the problem areas of a proposed design or prototype application and iterate upon it. No design is perfect the first time, and usability testing allows us to get quick feedback and improve the product. For example, if participants could not understand the wording of a certain action or could not figure out where in the application to click, the next design iteration can seek to improve upon that.
In conclusion, usability testing is a powerful methodology that can be used to quickly find issues with a design before any time is spent fully developing a product or releasing a feature into production that users may find unusable and frustrating.
For more information about usability testing, we recommend reading:
(Images in this posting are able to be used and shared commercially via “google advanced search”.)
Author: Sam Cheng