Prototype & Testing
We completed our final designs as a result of extensive iteration and review as a team. The level of iteration we achieved is a testament to the team’s commitment to making this app as user friendly as possible (Fig 1). The card used in our final design developed according to the importance of information discovered in our research. They evolved in an attempt to present large amounts of information in an easy-to-read way, therefore, not adding to our users’ cognitive workload’ (Krug, 2014) (Fig 2). As a result of these iterations, we landed on our final design for testing (See Appendix U).
To conduct our usability testing with an interactive prototype on Maze (see Appendix V). We recruited nine participants with a mix of male and female users. The age range was 27 to 61, with five frequent users and four infrequent users. Although we had a mix of people our primary age bracket was 40’s, in future testing, we must ensure we are recruiting users evenly in each age bracket. However, we had restricted access to participants due to the pandemic. Testing with older users who are not as comfortable with technology can highlight many areas that need refinement, solving all users’ problems.
We developed a script (see Appendix W) with several tasks to create consistency in our testing, asking participants to sign consent forms ensuring they were happy with the procedure (see Appendix X). We distributed a SUS questionnaire (See Appendix Y) to obtain quantitative feedback. The results of testing our tasks are as follows (See Appendix Z):
1. Onboarding: Uncertainty about what onboarding is with users suggesting they would generally skip it.
2. Search: Users found this process easy to complete with it on average taking 15.1 seconds. They also noted the distance to the bus stop and the available seating being great additions. However, some users thought the accessible space meant there was only one space open.
3. Adding favourites: This was a very well-performing task; users navigated it quickly and easily.
4. Support: Users found the help section in just 15 seconds. This area generated much positivity with testers that are aware of how active Dublin Bus is on Twitter.
5. Payment: This was our worst-performing task requiring, on average, 20 seconds to complete. However, it is difficult to say if this was because the prototype was on a desktop screen. Users stated they might have found where to add payment if the prototype was on their phone.
As a result of our redesign, our users on average marked 4.78 in the confidence of using the application. In future, it would be beneficial to conduct a SUS questionnaire along with our initial casual interviews & observations to assess how much we increased usability.
To improve our onboarding process, we need to review our calls to action. We did not contextualize this task in our script, so it is hard to tell if this caused confusion or the task itself. Our search feature was very successful. However, users suggested adding the length of time expected to be on the bus and directions from the bus stop to the final destination. We should also include the recognized disability icon beside the accessible space. Finally, we must design a more straightforward way of adding a payment method with users suggesting it reflects apple wallet (Fig 3).
There are plenty of opportunities to develop the Dublin Bus application further and design amendments to address. Overall, our testing successfully highlighted that our app is efficient, engaging and easy to learn and allowed us ‘to see how successful users are with their tasks’ (Harsha, 2019).
References
Krug, S. (2014). Don’t Make Me Think, Revisited (p. 15). Pearson Education.
Harsha, S. (2019). Usability testing in design — why is it important?. Medium. Retrieved 21 January 2021, from https://uxdesign.cc/usability-testing-in-design-and-why-is-it-important-cfddfbbdaac9.
