Improving GLP-1 medication user experience with Research and Design
Overview
Starting GLP-1 medication is a planned decision, yet users still hesitate, drop out of the funnel and quit the product. A pharmaceutical company focused on GLP-1 needed to know where the user experience lost people and why. The average user stayed for 3 months, so both the start and the end of the journey mattered.
This UX research case study covers how I ran ethical, remote user research for a medical product. I combined interviews with four surveys that reached 20,000 users in total. The findings shaped new designs and a 30% increase in purchase conversion, achieved through multiple A/B tests.
My role
UX research
UX Design
Date
Client
Bolt Pharmacy
The challenge
The company needed answers to three questions:
What causes users to drop off in the funnel?
Which reasons do users give when they hesitate?
Which reasons are at play when users quit the product?
Four factors made this harder.
A medical context. GLP-1 is a medical treatment, so the research had to be ethically sound. Participants needed to feel safe and in control of the interview.
Different user groups. People who want to start, people who are on medication and people who quit do not share the same concerns. One set of insights would not cover all three.
A short retention window. The average user stayed for 3 months. I needed to understand behaviour over time, not only at sign-up.
Two kinds of evidence. Interviews could explain why users behave as they do. Quantitative data was needed to show how widely those reasons applied.
The result
The result
Running multiple A/B tests increased purchase conversion by 30%. The retention proposals were also implemented, but no outcome was shared with me, so I can’t report on their impact.
What I learned
How to use correlational research well
How much impact good research can have on the success of a product
How to combine behaviour metrics with A/B test metrics
The process
Plan ethical user interviews for a medical product
I built a research plan for 6 remote interviews. Because the topic is medical, I designed the plan around ethical practice:
Every participant consented to the interview and to the recording.
I did not ask questions that could lead to medical advice or imply it.
I told participants they did not have to answer anything that felt uncomfortable.
Participants could pause or quit the interview at any time.
Use the theory of planned behaviour
GLP-1 treatment is something people plan. This made the theory of planned behaviour a good fit, so I used it as the framework for the research. I then tested the theory through the A/B tests.
Combine user interviews and surveys to study retention
For retention, I used both qualitative and quantitative data. I interviewed another 5 users who were on medication and ran 4 surveys. In total, 20,000 users responded.
Find the gap between three user groups
The data showed a real gap between people who wanted to start, people who were on medication and people who quit.
Identify stigma as the driving factor
Stigma was the driving factor across all user groups. It showed up in several forms:
Stigma about using the medication
Stigma about who to tell
Fear of the pain of needles
Side effects
One interviewee described how long the decision took:
It took me a long time to decide to inject myself. In the sense, it’s an extreme method in my eyes, but I tried dieting and exercising for years.
Redesign the quiz, checkout and treatment pages
The findings led to design proposals for both acquisition and retention. On the acquisition side, I designed 30 new designs. The changes included:
Steps in the quiz where people find out if they are eligible
Motivational pages
More questions about preferences and expectations
An improved checkout
Improved treatment overview pages
A/B testing to increase purchase conversion
We validated the designs with multiple A/B tests. Together, the tests led to a 30% increase in conversion on purchase.
Improving GLP-1 medication user experience with Research and Design
Overview
Starting GLP-1 medication is a planned decision, yet users still hesitate, drop out of the funnel and quit the product. A pharmaceutical company focused on GLP-1 needed to know where the user experience lost people and why. The average user stayed for 3 months, so both the start and the end of the journey mattered.
This UX research case study covers how I ran ethical, remote user research for a medical product. I combined interviews with four surveys that reached 20,000 users in total. The findings shaped new designs and a 30% increase in purchase conversion, achieved through multiple A/B tests.
My role
UX research
UX Design
Date
Client
Bolt Pharmacy
The challenge
The company needed answers to three questions:
What causes users to drop off in the funnel?
Which reasons do users give when they hesitate?
Which reasons are at play when users quit the product?
Four factors made this harder.
A medical context. GLP-1 is a medical treatment, so the research had to be ethically sound. Participants needed to feel safe and in control of the interview.
Different user groups. People who want to start, people who are on medication and people who quit do not share the same concerns. One set of insights would not cover all three.
A short retention window. The average user stayed for 3 months. I needed to understand behaviour over time, not only at sign-up.
Two kinds of evidence. Interviews could explain why users behave as they do. Quantitative data was needed to show how widely those reasons applied.
The result
The result
Running multiple A/B tests increased purchase conversion by 30%. The retention proposals were also implemented, but no outcome was shared with me, so I can’t report on their impact.
What I learned
How to use correlational research well
How much impact good research can have on the success of a product
How to combine behaviour metrics with A/B test metrics
The process
Plan ethical user interviews for a medical product
I built a research plan for 6 remote interviews. Because the topic is medical, I designed the plan around ethical practice:
Every participant consented to the interview and to the recording.
I did not ask questions that could lead to medical advice or imply it.
I told participants they did not have to answer anything that felt uncomfortable.
Participants could pause or quit the interview at any time.
Use the theory of planned behaviour
GLP-1 treatment is something people plan. This made the theory of planned behaviour a good fit, so I used it as the framework for the research. I then tested the theory through the A/B tests.
Combine user interviews and surveys to study retention
For retention, I used both qualitative and quantitative data. I interviewed another 5 users who were on medication and ran 4 surveys. In total, 20,000 users responded.
Find the gap between three user groups
The data showed a real gap between people who wanted to start, people who were on medication and people who quit.
Identify stigma as the driving factor
Stigma was the driving factor across all user groups. It showed up in several forms:
Stigma about using the medication
Stigma about who to tell
Fear of the pain of needles
Side effects
One interviewee described how long the decision took:
It took me a long time to decide to inject myself. In the sense, it’s an extreme method in my eyes, but I tried dieting and exercising for years.
Redesign the quiz, checkout and treatment pages
The findings led to design proposals for both acquisition and retention. On the acquisition side, I designed 30 new designs. The changes included:
Steps in the quiz where people find out if they are eligible
Motivational pages
More questions about preferences and expectations
An improved checkout
Improved treatment overview pages
A/B testing to increase purchase conversion
We validated the designs with multiple A/B tests. Together, the tests led to a 30% increase in conversion on purchase.
More works
Project
©Nathan Boomsma 2026. All rights reserved.
Project
©Nathan Boomsma 2026. All rights reserved.
Project
©Nathan Boomsma 2026. All rights reserved.