Doctar - UX Testing
First test
The analysis of quantitative results was the basis for developing the solution and the first prototype in low definition.
Test objectives
General: Find out how easy or not to use the functionalities included in the prototype.
Specific: Observe the success/time/failure metrics in the completeness of tasks, validate what was acceptable, and identify failures and consequent adjustments.
Applied methodology
This step was performed using qualitative and quantitative techniques, both supported by the same script with HEART metrics.
- The quantitative was made exclusively by measuring the execution of tasks using Maze.
- The qualitative consisted of remote interviews monitored by Google Meet, shared screens, applying the same Maze test, and in a second moment sharing the prototype link for free exploration, using the “think loud” technique, in which we asked users to think loudly during the execution of tasks and comments.
Learning
The above results are from the third version of the test. It was very clear that usability tests are delicate in their purpose and execution, and tasks must be drafted very clearly and objectively, otherwise users may become confused and the results will be skewed or will not have their relevance and objectives achieved properly.
Even with objectivity and clarity in tasks and on how to perform satisfactorily, some users do not pay attention to tasks, are “exploring” the interface and easily stray from what was requested, or interrupt the test with some distraction such as answering phone etc.
Second test
With the application of the same objectives and methodologies of the first test, here we expected an increase in the success of the tasks, based on the adjustments made based on the learnings made.
In the general picture there was improvement in the task success rate, and discarding the dropouts the rate was 100% for all who took the test until the end.
Below we can see some sequences of screens with the heatmaps showing the execution of some of the main tasks.
Patients
Search for an endocrinologist and make an appointment with Dr. Patrícia Fernandes.
Cancel the appointment with Dr. Patricia Fernandes.
Professionals
Cancel Alex Mathias’ appointment scheduled for 5/17 at 09:00
On the 25/05/2021 agenda, make an appointment for the patient Dario Cavalcanti at 14:00.
Start the teleservice of the consultation of the patient Luiz Souza Lopes, scheduled for 17h of the day 17/05/2021.
Additionally
The prints of the report screens below show the “Usability Score” of 71 and 78, respectively for patients and professionals, meaning that ease of use has been set at the threshold between medium and high for both cases, according to maze metrics.
Final qualitative tests and their learning
Again, qualitative tests were performed with 6 possible patients and 6 professionals, as follows:
a) Performing maze tasks, thinking aloud, with as little assistance as possible to complete the tasks.
b) After the completion of the test, a brief exploration by the prototype directly shared by Figma, and a conversation about feelings and perceptions of the applications.
c) critical points and suggestions for improvements.
All participants successfully completed the tasks, but the execution time was well varied.
Additionally, some users of professional qualitative research revealed, even completing the tests with praise:
“I would like the icon name to appear when pointing the mouse”
I think a scroll bar for the time grid would help a lot”
The first quote above is professional referring to tooltips with the name of each section on the main menu buttons, which is not possible to implement in the current version of Figma while maintaining navigation, but was planned for development.
In the second it was clear that the “detail of detail” in the construction of the prototype was missed, not to display the scroll bar that, of course, will exist.
As for the qualitative comments of patients:
“I’m not in the habit of setting up more than one reminder for an event””
I wish I could do a doctor’s research from the date”
Highlight: the participants of the qualitative stage did not previously know the interfaces.
Final conclusions
It was clear that the “exploratory” tendency, in contrast to what was requested, is an inescapable human factor and with all the challenges that usability tests can mean, the results were excellent, showing an evolution in the success indicators between tests, meaning that the initial learnings helped create a superior product for the high definition prototype.
Mission Accomplished!
Related articles:
