1.5 min video recap
Summary
As part of a larger project aimed at increasing prescription refills through research, data analysis, and rapid experimentation, we uncovered a quick solution that boosted first-time refill numbers by over 60 per month, resulting in approximately $60,000 in additional revenue each month. This initiative laid the foundation for continued improvements in patient activation and retention.
Skills
+Design systems
+User research
+Growth experiments
+Prototyping
+User testing
Role
Intermediate product designer
Team
Myself, 1 product manager, 2 Developers
Context
PurposeMed is a multi-treatment platform that connects patients from underserved communities to clinicians so that they can get specialized care from their homes. Their treatments include HIV prevention & treatment and ADHD & mental health care. Their pharmacy Affirming Care ships the medication.
Problem
44% of Freddie patients don’t end up refilling their first prescription.
After patients refill their first prescription, 80% will continue refilling. Freddie makes money from prescription sales. Increasing the first refill rate will increase company revenue.
Goal
How might we increase the percentage of patients refilling their prescription for the first time?
Discovery
We started by understanding the refill process through discussions with clinicians, reviewing touchpoints with marketing, testing client scenarios, observing sessions, and conducting user interviews with Freddie patients.
Discovery
We found that the in-app refill process is confusing, causing a 33% of patients to drop off before initiating a refill.
Other insights: Patients' lifestyles have changed, making daily dosing harder. Many don’t remember how to start a refill. Reminder emails are mistimed due to delays, missed prescriptions, or skipped doses. There were many valuable learnings from discovery, but we couldn’t tackle them all at once. We did a prioritization session and chose the lowest effort solutions that would have the highest impact.
Solution validation
We hypothesized the drop-off from login to the refill process was due to a cluttered dashboard. To test this, we ran a two week experiment. Here are the changes we made:
Before
Solution validation
Simplified dashboard experiment
Experiment
Solution validation
Experiment findings • Feb 9th - 23rd 2024
+30
Additional prescriptions
+$30,000
Additional revenue
The experiment also resulted in a reduction in time to login -> starting the refill process and an increase in completion of the refill questionairre.
Solution
The simplified dashboard experiment increased first time refills by reducing friction and cognitive overload. As a result, permanent changes were made to the dashboard using the new design system.
Before implementation, the designs were usability tested in Maze with 10 users. They performed well and didn’t require further refinements.
Improvements:
Revised copy to improve expectation setting.
Deprioritized features that weren't used frequently.
Simplified the card design to clearly show the next step in the process.
Utilized progressive disclosure to guide patients through each step.
Solution
The refill experience: A simplified, guided process built on a scalable design system for future growth, utilizing progressive disclosure to deliver information at the right time, ensuring users are never overwhelmed while maintaining clarity throughout the journey.
Solution
The simple improvements made it easier to complete the refill process.
+60 / month
Additional prescriptions per month
+$60,000 / month
Additional revenue per month
+23
Additional users downloaded a lab requisition
-6 s
Decrease in time to complete the refill process
Lessons learned
In this project, I truly learned the power of lightweight experiments. We ran a simple test that proved highly effective, so we decided to "turn it on" while the new designs were still being developed. By doing this, we were able to see a direct impact on client retention in real time. This not only validated our approach but also gave us the confidence to continue improving the experience, knowing that our direction was solid. It was a great example of how quick experiments can deliver immediate value while providing the foundation for more long-term enhancements.
Next steps