When we design we can fall into the presumption of things that are not true (and not realize until very late!). Even more, if we don’t have a lot of time (we never have it) maybe we think that the best is to skip the research and become our own reference (Won’t someone please think of the users?!). But discovering and empathizing is essential to start any creative process.
In this article, I will focus on the importance of user research during a recent design process of a mobile wellness app. The project was an exercise from the UX-UI Ironhack Bootcamp. The time and resources were limited: An individual project carried out during 3 class days. I had to identify an opportunity for an app to improve people’s day-to-day lives and deliver a proposal in the form of a wired click-through prototype.
My first step was delimiting the agenda, I didn’t have a lot of time: I decided that more than a half of the time I spent on discovery and define the problem, cutting the development and delivery time.
We had to choose from a list of “7 wellness to have a balanced life”, I selected the topic of Physical Wellness. I was interested in focusing the app on the personal management of Stress. I started diving on the web looking for professional information.
Stress is “the evil” of the XXI century, more than 60% of the United States citizen ever had at least one problem due to stress. There are certain commons symptoms as sleepiness and anxiety, but some people have more serious symptoms: mental problems, heart disease, depression or an eating disorder. Because of cortisol and adrenaline that the people receive during a stress situation, someones get hooked and look for more. The modern societies along, with smartphones too, push up to stress addictions. In our society, busyness seems to equate with importance.
The reasons why people get stressed can be grouped into 7 the reasons, but basically, it is a psychosomatic response to being unbalanced. But for some people, it is difficult to balance their life again and gets stuck in stress. I needed to understand them better. I began to develop interviews.
The interviews usually give more qualitative information than quantitative but are good to begin to understand the whole picture. I interviewed people who had to deal with stress recently, regardless of whether they had any health complication from that. It was important to generate trust, I needed them to tell things that perhaps were sensitive issues. Among the interviewees, I was able to meet people who had panic attacks and anxiety problems. Likewise, asking the interviewees, I realized that the definition of stress was very broad.
I split the interviewees into two groups:
—Some get the stress symptoms under a pressure situation. Before an exam, or during organizing the Christmas dinner. Then the stress goes.
— Others have stress during a period. Different small situation triggered the symptoms daily. The stress continues after the pressure situation, needing a more radical change in their lives to cope with it. This second group has a tendency to addiction or major health diseases.
“When the pressure was overcome and the stress remains the health problems appears.”
I started to map on canvas all the information I got related to stress.
I am not a psychologist or doctor or an expert on the subject. I know that I being doing assumptions. But my application should help to solve a problem and I needed to develop a hypothesis:
“From all the people who suffer stress, there is a group that has health complications because they don’t know how to cope with daily stress, extending it over time.This group has a different behavior of the other stressed people, and hurts itself.”
I needed to know how these people behave, what is wrong, and how can I help. This requires a lot of information, and the best way to obtain it was a survey.
I identified two groups of stressed users. So, using lean survey canvas, I designed a questionnaire with a screener question that would allow me to differentiate the results. I called people to anonymously answer.
I had the weekend to collect as many answers as possible. At the moment of analyzing the data, of 116 unique visits, 87 felt comfortable sending their response and only 41 people (47%) were from the target group (people with chronic stress). The results data allowed me to validate and enrich the information I already had:
- 75% Anxiety and 70% Bad Temper
- 50% Didn’t sleep well. But cares sleep doesn’t help with stress
Main Recommended Solution to deal with stress
- Have Fun (music, exercises, video, others)
- Dealing with the source of the stress
- Ditch toxic persons
Common suggested solution (in the free text input):
- “Solved one problem at a time”
The analysis also revealed another important fact, the hypothesis that I had raised was wrong. There is no different behavior between both two groups. They deal with stress in the same way. Comparing the answers, I noticed how they responded in the same way. One with more difficult than other, but both groups de-stress when feel that their problem is being solved.
The User Personas is the creation of a fictional character that represents real users. In my case, I called her Martha, a “full time worker mother of two children”. Martha has panic attacks, and anxiety problems. Martha thinks a lot about her stress, and she always complains about her daily problems. But she doesn’t ask for help. Despite being happy with her work and family, the stress doesn’t allow her to enjoy it. She says he needs a radical change, like a vacation to India. But what she really wants is cope with their stress, to complete her routine, and enjoy her life. Thats is her goal.
Having my user Personas defined, I can write the features that the application must have. Must be features that really help users like Martha. How I can help Martha?
During the research I taked note of advices to deal with the stress. Advices from the interview, the survey and the Internet articles. I wrote them in post-its and started to arranged it on an affinity diagram: I stacked the similar advices, and then I wrote a title to each column. I ordered the columns as if they were steps. I was able to realize the steps the user needed to deal with stress:
1st. Calm down 😌
(to relax the symptoms)
2nd Fun 😄
( to forget her problems)
3rd Solve the problem 🧐
(focus 1 by 1)
Solving the problem the users Feels that all is ok. ✔️
I fell I were designing a system more than an app. A three-step system to feel better. Calm — Fun — Solve. And getting the user to complete the three tasks from his phone is not so easy. So I drew a graph with the flow of the desired application.
With the guerrilla test, I can improve the flow. To do this, I made some prototypes on paper that helped the testers to imagine how will works. With this usability test, I validate and detect some frictions to improve.
For example, the onboarding was long, to introduce slowly all the process to the user. But this long presentation results exhausting for users like Martha. I shorten it and let the user learns it on the way.
Another drawback during the test was the screen where the user must solve the problem that stresses her. A really important screen, that gives sense to the whole app. But the texts were not entirely clear and users didn’t know what to do now the onboarding was short. This screen was redesigned for the final prototype.
At that moment, I didn’t have enough time remaining to build the prototype to deliver. And the scope of the project began to be large. There were additional screens, such as a history of problems, a reminder, and the configuration of the app. But I decided to shorten the scope, in order to concentrate the efforts on the calm-fun-solved flow. Some screens were left outside.
I had a lot of helpful information from all the process, so I could design wireframes using a computer design program (Sketch) really fast. The wireframes were imported to Marvel, which allowed me to add gestures to interact over the screen. This prototype was tested with real users from a phone. Some changes were made, and I could reach the deadline with a good proposal:
The research’s insights based the desitions. The result was quite solid for being carried out individually only in three days! Didn’t matters that in the middle the hypothesis change, the main goal and the person remains. The information obtained had helped me to design empathizing with the user. It’s good keep in mind that a big scope is difficult to test. Creating Martha, and have all the research that supports it, helped me to understand and focus on what really gives value to the users.