Skin health app and web flow
Industry
MedTech
Activities
UI/UX, UX Research, User Testing, User Interviews
​
Task
Increase the number of consultations via teledermatology portal
​
Result
4x increase in monthly number of consultations
Overview
Problem
Miiskin is a teledermatology platform that enhances healthcare accessibility in the US by allowing users to securely submit information for dermatologists to review. Originally, it functioned as a marketplace that enabled small clinics and doctors to establish their own online practices.
Miiskin's marketplace model faced low conversion rates because small clinics typically attract less traffic. In contrast, Miiskin.com drew many visitors with its professional team. Therefore, the company wanted to transition to a centralized model to improve conversions.
Business goal
Increase the number of patient consultation requests.
Build an MVP flow
Design a cost-efficient and quick solution, which would allow us get first real users from miiskin.com.
Conduct user research
Find out who our users are and what they expect from our service in order to create a better product for them.
Iterate and evolve
Gradually improve the conversion flow and the product with consideration to the resource limitations.
01. Discover
Desk research
​
A cheap way to get initial insights into users and market.
I like to start work on a new project by studying unfiltered user discourse across platforms like Reddit or social media groups.
This method allows me to:
-
Reveal market competitors and understand user sentiments towards them.
-
Gain basic insights into user motivations and pain points.
-
Identify emerging trends and potential opportunities within the target market landscape.
-
Find relevant areas of study for the future user interviews.
Competitors research
​
Understanding the market proposition.
To kick off the project, I analyzed both primary and secondary competitors. I created a Points of Parity and Points of Difference table to grasp the essential functionalities necessary for our product's basic market viability, as well as the distinguishing features that set our competitors apart.
Points of Difference
can be used as ideas to improve our offering
-
Easy sign up (e.g. with Google)
-
Mobile app
-
Care for minors
-
Home delivery of medications
-
Pharmacy pickup of medications
-
Insurance support
-
Membership model
-
Requesting medications
-
Postponed photo upload
-
Saving incomplete requests
-
Skin quiz
-
Subscription RX model
-
Order tracking
-
Photo testimonials
Points of Parity
a must-have basic functionality
-
Web-based consultation flow
-
Asynchronous or video consultation with a healthcare provider
-
FAQ
-
Chat support with a provider
-
Self-pay payment option
MVP flow
A band-aid solution to test the channel and get to know the users.
Real processes don't always adhere strictly to the ideal double diamond model. Initially, we didn’t have any real users engaging with the new channel, nor did we have the development resources to create a comprehensive end-to-end web flow. To explore this new avenue, we developed a simplified version of the flow.
IMPACT
In the initial month, we acquired 30+ converted users and a significant number of registered users who abandoned the flow. This presented an opportunity for user recruitment for research. Despite its minimal viability, the flow underwent multiple iterations and consistently yielded more conversions each month.
Survey
​
Leveraging survey for effective research participants recruitment.
After numerous unsuccessful attempts to recruit users for in-depth interviews, we decided to implement open-ended surveys with a small incentive for our users. After filtering the results, I selected the responses that appeared most genuine and informative and reached out to offer participation in an interview. Remarkably, we received a 100% positive response rate.
IMPACT
In addition to providing research participants, the survey yielded valuable insights, which I integrated into the affinity map created after the interviews.
User interviews
I conducted multiple interview sessions with both users who completed the conversion process and those who abandoned it.
​
“The big draw of teledermatology is convenience and quick access to care, but making a user download the app to get care is a huge blocker. When I'm stressed about a skin issue, I don't want to think about creating new accounts or downloading things.”
Lindy
“I had one more question, which I had sent to the doctor via email, but she never responded. And what she prescribed was not strong enough. If I could message her, I would be able to tell her that and maybe get a stronger prescription of tretinoin. That doesn’t seem to be available.”
Katie
“I filled out the information about me, and then it asked for my credit card, but at that point I still hadn't seen the information about the doctor. I didn’t know what would happen after I pay.”
Eileen
After conducting interviews with six users and analyzing the survey data, I organized the results into an affinity diagram.
Main user insights
Users don't want to search for the information.
Many simply follow the flow once they started it without visiting other pages / screens for information. We should consider adding the info directly into the flow.
Many users drop off when prompted to download the app.
According to the analytics from Mixpanel, 74% of users would drop off when prompted to download an app. We should look into creating a seamless solution.
Communication with dermatologists feels cumbersome.
Lack of embedded messaging forces users to seek other ways of communication with a derm - through calls or email, which makes the communication feel fragmented and clunky.
02. Define
Personas​
​
Defining our audience
At this step I started to know our users quite well, so I could define several personas using Miiskin:
Prescription John: This user is clear about his needs and utilizes our service to promptly obtain prescriptions for solutions he's researched on forums and social media. He values confidentiality and swiftness.
​
Tretinoin Jane: This user is well-educated on skincare topics. She has tried many over-the-counter products, but social media influencers and dermatologists that the real effects come from the science-based prescription medication.
Urgent Kelly: When faced with sudden issues like rashes, this user seeks immediate answers, driven by concern about their condition worsening before their next physical appointment.
Hesitant Mary: This user is inclined to abandon the process, preferring in-person dermatologist consultations over online routes due to lower trust.
BUSINESS PERSPECTIVE:
From a financial standpoint, it's prudent to focus on personas that repeatedly engage with our service with minimal effort. Prescription Johns often return for medication refills, having simple issues that experienced dermatologists can quickly review. Thus, our initial focus is on catering to prescription-seekers.
CJM​
Based on insights, I constructed user journey maps to visualize potential pain points experienced by users.
​
Themes​
​
Best ideas emerge at the intersections of these 3 pillars: user needs, business goals, and competitive insights.
​
I facilitated a mini workshop with the product owner to prioritize areas for improvement collaboratively. Data points were represented on sticky notes, categorized into three groups:
-
Business goals (blue)
-
Competitive insights (green)
-
User needs (yellow)
We then clustered these data points into themes, identifying those that intersected across all three pillars. For each theme, we formulated "How might we" statements to guide ideation sessions.
03. Develop
Prioritization​
​
Using the lean prioritization to sort out the ideas.​
Prototyping & Testing
​
Conducting moderated testing sessions enabled me to refine the flow of the web app solution.
I conducted 5 moderated testing sessions specifically targeting the web app flow, with users tasked to request a consultation with a doctor. Across all sessions, users achieved the task successfully without requiring guidance, indicating a seamless flow. However, there were minor interface nuances highlighted in their feedback, guiding further refinements.
Some of the user testing insights:
-
A user couldn’t find the reviews - we should test adding some right into the flow
-
A user still had unanswered questions before the payment step - we should look into summarizing everything before payment
-
A user wanted to review their answers before payment - we should test allowing users to double-check their questionnaire
-
A user didn’t understand why she had to insert her address (“is medication going to be delivered?”) and phone number (“are they going to call me?”) - look into explaining/eliminating these fields
04. Deliver
The solution
​
Seamless web onboarding.
​​
Details
Registration
​
No need to download the app to get started = low commitment from a user.
​
No need to create a password; a user can log in with the code sent to their email.
​
Using the "Progressive Disclosure" principle by providing information gradually.
Questionnaire
​
Utilizing multiple problem-specific questionnaires instead of a one-size-fits-all generic solution from the past.
​
Implementing the "One thing per screen" principle to reduce cognitive load.
​
Utilizing a progress bar to assist in managing user expectations.
​
Allowing users to specify their particular treatment preferences (e.g., topical only, most affordable, etc.).
Pharmacy selection
​
Allowing users to automatically find pharmacies in their area ensures that the experience is not interrupted.
Photo upload
​
Allowing users to take additional photos and both take and upload pictures provides more flexibility.
​
The photo upload screen is automatically tailored to the described problem
Payment and submitting the request
​
Reiterating what's included in the payment and suggesting easier payment methods like Apple Pay.
​
Allowing users to continue using the web version to view the doctor's response but encouraging them to download the app, which enhances user retention. More than 70% of users ultimately downloaded the app.
Messaging
Addressing users' issues such as lack of guidance and information by implementing a Miiskin Bot to explain to patients what to expect and providing access to medically reviewed educational resources.
​
Allowing users to communicate with the doctor via chat within a set timeframe.
Retention
​
Since Miiskin is a newer company, it's important to share some success stories. To obtain before/after photos and testimonials, we offer patients who have completed a consultation the opportunity to share their story in exchange for a gift card.
​
Treatment sometimes necessitates follow-up consultations, and Miiskin reminds patients to schedule those when it's time.
​
To save users time, Miiskin remembers and auto-fills information, but users can still edit it in case something has changed.
Components library
​
Optimizing the maintenance and implementation for the design and development teams.
I've developed a UI kit encompassing all reusable visual design elements, adhering to the principles of Atomic design. I've proactively organized meetings with both the CTO and a front-end developer to explore optimal strategies for enhancing their workflows. Armed with their insights, I've meticulously crafted a comprehensive library of components.
In a bid to streamline my own workflow, I've ensured that all elements are fully customizable and flexible, leveraging my expertise in utilizing Figma's capabilities.
Results
​
Achieved a 4x increase in conversions over a period of 6 months.
All changes outlined in the task were implemented gradually. We would develop a segment of the flow, release it, conduct A/B testing, and then iteratively improve upon it. As of the time of writing, the entire web flow has not yet reached 100% completion according to the design, primarily due to resource constraints. However, we have observed a steady increase in the number of patient consultations. The data presented here is extracted from Mixpanel, but specific numbers have been withheld in accordance with agreements with my supervisor.