Mychart by epic
Enhancing patient experiences by fixing usability issues and reducing frustration.
CLIENT
Epic Systems Corporation
PRODUCT DESCRIPTION
MyChart is a patient portal developed by Epic Systems Corporation, a leading healthcare software company. MyChart serves over 165 million patients globally, reflecting its massive reach and adoption in the healthcare industry
MISSION
By addressing usability challenges and identifying root causes of dissatisfaction, we aim to enhance the patient experience and eliminate frustration.
PLATFORM
Mobile App
Introduction
One morning, you rushed to the hospital, completed forms, saw your doctor for 20 minutes, and had blood samples taken. By noon, everything was done. Two days later, you received your test results but felt frustrated—not because of bad news from your checkup, but due to issues with the app delivering the results. Navigating the app felt like being lost in a jungle due to its quirky interface, overwhelming menu options, and scattered button placement. Recognizing that millions of patients rely on MyChart, I saw this as an opportunity for improvement. To address these challenges, I set out to identify the root causes behind user dissatisfaction with MyChart.
Discover Stage
Collecting feedback from internet
To identify problems with MyChart, I analyzed user reviews on the Apple App Store and Google Play. These platforms provide valuable insights because customers often share honest feedback about their experiences with the app. By focusing on 1 and 2-star reviews, I uncovered recurring issues and frustrations that users were eager to voice. This method allowed me to pinpoint several common pain points and gather actionable insights for improvement.
Define Stage
"How Might We" Framework
After reviewing feedback from 1- and 2-star ratings, I identified patterns in user complaints and grouped them into distinct categories of issues. To dig deeper, I used the “How Might We” question framework to turn these challenges into opportunities for improvement. By framing constructive, open-ended questions, I aimed to uncover innovative ways to address existing problems within MyChart while maintaining a positive and solution-oriented mindset.
Below is the design process for the 'How Might We' framework
Online feedback regarding print-related issues
Result in HMW
Online feedback regarding message system
Result in HMW
Online feedback regarding UI & Animation issue
Result in HMW
Online feedbacks regarding Schedule and Appointment
Result in HMW
Key takeaways for “How Might We” Questions
Frame questions broadly enough to focus on uncovering opportunities rather than solutions.
Avoid making questions too narrow, as this can limit the exploration of potential possibilities.
Use positive language to emphasize enhancing the user experience rather than discouraging app usage.
Evaluation Stage
Impact and Effort Matrix
After the define stage, we identified several HMW (How Might We) questions. The next step is to evaluate which issues are truly worth solving. Are these problems impactful enough to address? Will users value the updates or changes? Will the solution create meaningful value in the market? This matrix serves as a reminder to stay focused on solving real problems rather than proposing solutions without purpose.
My goal is to prioritize solutions that deliver the highest impact with the least effort. By using the Impact and Effort Matrix, I can better assess and measure issues, ensuring that my efforts are directed toward meaningful and efficient solutions. This helps me make informed decisions about where to invest my time and resources for the greatest user and business benefit.
Final Result
Low Effort, High Impact
User Interface Issue
How might we enhance the user experience as seamlessly and ease the faction for users while using this app
Low Effort, High Impact
Communication, Message System Issue
How might we make it easier for users to get timely responses from doctors.
These two issues hold significant potential for creating value if resolved. They address critical user pain points and have the capacity to improve the overall experience and satisfaction.
Conclusion
The greatest challenge in this case study was identifying problems that are truly worth solving. In reality, many solutions are built without addressing actual market needs. Overcoming this challenge required active listening—turning user feedback into valuable insights instead of dismissing it as complaints. This process taught me the importance of focusing on solving problems that matter to users, ensuring every solution adds real value. Moving forward, I will focus on addressing the two key “How Might We” questions: improving the user interface for a seamless experience and enhancing the messaging system for timely responses. I plan to develop specific solutions for these challenges, starting with actionable designs and eventually progressing to high-fidelity prototypes.