Medosome Biotec

Giving children a better chance at a long and healthy life.
3D rendering of Clinical Trials Dashboard on a smartphone

Designs and data have been modified and do not represent real products or systems.


Summary

I transformed a fragmented, spreadsheet‑driven workflow into a unified, real‑time clinical trial oversight platform that gives biotech leadership instant clarity, anytime, anywhere.

0%
Financial loss from missed or late information
15%
Reduction in analysis time
30%
Shorter leadership meetings
Instant
Risk awareness through real‑time alerts


Clinical Trials Dashboard

A mobile‑first command center preventing costly delays before they happen, was needed by Metosome Biotec, a small biotech company that specializes the health challenges faced by children around the world.

My Role


As a Product & UX Designer I led the project from discovery through delivery, including research, strategy, IA, wireframing, prototyping, usability testing, and UI design.

Platform

Mobile‑first responsive web application

UX Methods

Stakeholder Interviews, Personas, Journey Mapping, Competitive Benchmarking, Information Architecture, Wireframing, Prototyping, Usability Testing

Tools

Figma, Lunacy, Google Suite

A High‑Stakes Problem

Assumption

There are multiple off the shelf solutions for Biotech and the modern workplace offers multiple modes of integration and data management to make executive decisions easier than ever.

Truth

A biotech CEO was losing money, not because trials were failing, but because information was failing. Critical updates were buried in spreadsheets, emails, and PDFs. By the time leadership learned about delays, the damage was already done: missed milestones, shaken investor confidence, and costly setbacks.


Photo puzzle

Conflict

Medosome Biotec is a small company in a saturated space, so custom software would have to be produced and a minimal cost, can we produce a solution better than the multiple products on the market?

Clinical trials move fast, and fall apart even faster when leadership doesn’t have the right information at the right time.

Rich

The CEO

Rich Photo

Age 62

"I need to see what’s at risk before it becomes a problem; not after."

Description

Rich leads a mid-sized biotech firm with several active clinical trials. A former scientist turned executive, he’s strategic, analytical, and time-starved. He travels often to pitch investors, meet with regulators, and manage high-stakes partnerships.

Objectives

Stay informed on all trial statuses at a glance

Identify red flags without wading through emails or spreadsheets

Report confidently to investors and the board

Motivations

Being proactive, not reactive

Leading with data-backed confidence

Protecting financial and scientific investments

Frustrations

Data is scattered

Updates are too detailed or technical

Important risks are buried or missed entirely

What is Leadership Experiencing?

Observation and having research and leadership walk me through their process I uncovered forces driving the crisis:


Photo of data on a screen

Fragmented Information Across Spreadsheets

Trial data lived in: Multiple Excel Sheets, Email Chains and Texts; nothing was centralized. Nothing was real‑time. Everything required digging. This fragmentation meant the CEO often learned about issues only after they had already escalated.

Overwhelming, Dense Reporting

"I don’t have time to read through rows of data to find what matters."

Important insights were buried in noise. Risks were easy to miss. It’s not a lack of ability; it’s a lack of translation.

Delayed Awareness of Problems

By the time leadership discovered a delay: slow enrollment, site activation issues, missed milestones, it was already expensive.

The cost wasn’t limited to time lost; it eroded confidence with investors and partners.


photo of data on an unfocused screen

Communication Bottlenecks

Operations managers and project leads were flagging issues, but through channels that weren’t built for urgency. Leadership wasn’t ignoring problems; they simply weren’t seeing them in time.

Overarching Driver

Biotech leadership needs real‑time clarity, not after‑the‑fact reporting.

The problems wasn’t our trials were failing. The problem was our information was failing.



Photos of a calendar, a stressed man, and a data sheet

Root Cause

Biotech leadership needs real‑time clarity, not after‑the‑fact reporting. The company wasn't facing crises because trials were failing, it was losing money because information was failing.

Problem Statement

How Might We enable a Life Sciences executive a single place for clinical trial data (as opposed to multiple computers or multiple screens) so that they can finally see every clinical trial clearly, catch risks before they become crises, and make confident, company‑shaping decisions anywhere, anytime?

Design Process

A Lean, Iterative Approach Grounded in Real Workflows

Lean UX allowed me to rapidly prototype and test solutions. It provided a good balance between developing a solid research base, and quickly test options, with attention given to Section 508 Accessibility concerns.

A diagram of the Double Diamond process of the Lean UX Process

The Human-Centered design process involved an iterative cycle of gaining options, followed by refinement.

  • Think: Conducted interviews, synthesized research, mapped journeys
  • Make: Built low‑fidelity prototypes, redesigned IA, created wireframes
  • Check: Ran multiple rounds of usability testing and workshops

This cycle repeated across major features, ensuring every decision was grounded in testing data.

The design time line of A Mobile‑First Clinical Trials Command Center

TLDR: AI Use For UX

I tried a few AI tools, taking this project for research to production;

Once the research was processed, and I had worked out the major pain points with AI, I discussed solutions and different user flows. Once I had the best concepts I used AI to design wireframes, that I refined by hand. There was no single prompt that provided the final designs, the process was an ongoing conversation with iterative improvement.

I would estimate I saved 20 hours of combined research and research processing time. Approximately 10 hours were saved in the design phase by using AI, and additional more hours were saved by having AI generate clickable prototypes with HTML, CSS and JavaScript.

Based on my experience and best practices I have outlined how to use AI in UX, followed by more a more detailed description of how I used it for this project.

I would recommend:

1. AI‑Research (Claude + Perplexity + Custom Agents)

AI to scale research, not to replace human insight. Claude and Perplexity to scan research documents, interview responses, survey data and the like. Created a custom agent to categorize thousands of community posts into themes. Use AI to find contradictions, outliers, and underserved segments.

Example prompts

"Analyze this research data. Cluster recurring pain points. Highlight contradictions or unmet needs mentioned indirectly." "Summarize this document into actionable insights for UX design."

    AI Outputs
  • Theme clusters
  • Early hypotheses to validate in interviews.
  • A synthesized baseline of the user experience.
    Refine manually
  • Removed false positives and emotionally inaccurate interpretations.
  • Re‑clustered themes based on real interview nuance.
  • Ensured insights reflected lived experience, not AI generalizations.

    • Time saved
    ~20–30 hours of manual reading and coding.

2. Visual Exploration & Concept Divergence (Midjourney + Figma Plugins)

    Use AI to explore more directions, faster, without dictating the final design.

    What to do
  • Used Midjourney and Figma AI plugins to generate early UI concepts and layout variations.
  • Iterated through multiple prompts to explore hierarchy, tone, and visual direction.
Example prompts

"Generate UI concepts for a solution for our user (Be specific based on your project). Prioritize clarity, trust, and accessibility. Include simple navigation, and (Describe the color palette needed).”

"Explore variations of a resource directory page with strong hierarchy and minimal cognitive load."

    AI Outputs
  • 20+ layout variations
  • Color palette explorations
  • Early visual metaphors (e.g., "guided path," "mission‑driven community")
    Refine manually
  • Rebuilt all layouts in Figma to meet accessibility and UX standards.
  • Adjusted spacing, contrast, and hierarchy based on user priorities.
  • Removed decorative elements that didn’t support usability.

    • Time saved
    ~6–8 hours of early‑stage visual exploration.

3. Prototype Development (GitHub Copilot + Figma Dev Mode)

    AI accelerates the technical scaffolding of interactive prototypes.

    What to do
  • Used GitHub Copilot to generate starter code for interactive components.
  • Asked AI to convert Figma flows into lightweight HTML/CSS prototypes for testing.

    • Example prompt
    "Generate accessible HTML/CSS for a prototype with WCAG‑compliant keyboard navigation."
    AI Outputs
  • Initial code.
  • Quick prototypes to validate navigation and content density.

    • Refine manually
  • Rewrote code to meet accessibility, performance, and security standards.
  • Adjusted interactions based on usability testing feedback.
  • Ensured the prototype matched the intended UX, not AI’s assumptions.

    • Time saved
    ~4–6 hours per prototype cycle.

4. Decision‑Making: Where AI Helped, and Where It Didn’t

    Be deliberate on choices about when to use AI and when not to.

    AI is ideal for
  • Scaling research
  • Exploring divergent concepts
  • Speeding up repetitive prototyping tasks
  • Summarizing long documents
  • Generating edge cases and failure scenarios

    • AI is not ideal for
  • Emotional insight
  • Prioritization
  • Interaction design decisions
  • Accessibility judgment
  • Final UI design
  • Anything requiring lived experience or nuance

    • AI as a multiplier, not a crutch.

User Research

Understanding the User's Experience

AI Assisted Research Icon

Even on projects with constrained scope, I leverage AI to maximize research depth and minimize administrative overhead. This ensures every hour spent on discovery yields maximum insight.

Competitive Landscape Analysis

  • Used AI to instantly synthesize data from competitors like Bloomberg, Apple Health, and Asana.
  • Spotted cross-industry patterns and differentiation opportunities in minutes, not hours.
  • Filtered out noise to focus strictly on features that mattered to our stakeholders.

Script Refinement & Error Checking

  • Tested interview scripts with AI to catch biases and logical gaps before the first meeting.
  • Validated question flow for clarity, ensuring high-yield conversations with the CEO and leadership.

Solution Scouting

  • Queried AI to quickly compare off-the-shelf solutions and our workflows needs.
  • Quickly ruled out non-viable options, allowing the team to focus build efforts only where custom solutions added unique value.

A Microphone Icon

Discovery Research

1:1 interviews with the CEO, project leads, and operations manager

Mapping of current workflows and bottlenecks

Competitive benchmarking (pharma dashboards, Bloomberg Terminal Mobile, Apple Health, Asana)

A Key Icon

Key Insights

Leadership needed clarity, not more data

Project leads needed a way to communicate without admin bloat

Operations needed a way to escalate risks earlier

A portfolio view was more valuable than trial‑by‑trial dashboards

Journey Mapping

The CEO’s real‑world workflow revealed:

  • Fragmented data; no single source of truth
  • Critical insights buried in noise
  • Teams members join meetings miss-informed
  • No real‑time alerts; late discovery
  • Outdated data; slow communication loops
A Journey Map of the User Journey of the Medosome Biotec CEO

Industry Analysis

Existing platforms failed because they did not fit our process, not because they were bad.

Competitive Benchmarking Revealed:

  • Pharma dashboards were too dense
  • Productivity apps lacked scientific context
  • Financial tools had strong visual hierarchies but weren’t trial‑specific

The Opportunity:

Combine the clarity of productivity tools with the rigor of clinical trial data, in a mobile‑first format.

Program Goal

Design a mobile‑first dashboard that enables a biotech CEO to monitor real‑time progress across all clinical trials, quickly, clearly, and in the palm of their hand.

AI Assisted Design Icon

I integrated AI tools to accelerate the transition from concept to code, ensuring rapid iteration.

Realistic Content Generation

  • Provided contextual prompts to generate realistic placeholder content for wireframes, replacing generic "lorem ipsum" with data that reflected actual user scenarios.
  • Enabled stakeholders to visualize the final product's tone and volume early in the process, leading to more accurate feedback.

From Inspiration to Usability

  • Used AI UI generators to rapidly explore visual styles and layout patterns for inspiration.
  • Applied human judgment to refine these concepts, ensuring accessibility, brand alignment, and functional usability before moving to high-fidelity prototypes.
  • Transformed raw AI concepts into polished, production-ready designs through iterative manual refinement.

Accelerated Code Implementation

  • Leveraged generative AI (ChatGPT, Claude) to draft boilerplate code and component structures for interactive prototypes.
  • Reviewed and optimized all generated code for performance, accessibility, and maintainability.
  • Reduced development time significantly, allowing more focus on user testing and interaction logic.

Prototype & Validate

Testing Early Concepts Against Real‑World Needs

Testing Approach

  • Low‑fidelity wireframes
  • Rounds of usability testing
  • Validation of micro‑interactions (long‑press, risk identification flows)
  • A/B testing of portfolio vs. single‑trial dashboards

Key Findings

  • Early prototypes tried to do too much
  • Executives wanted only what matters now
  • Portfolio view outperformed trial‑specific dashboards
  • Progressive disclosure reduced cognitive load
  • Mobile‑first design was essential for travel and meetings

The Big U-Turn

We had tried a few off-the-shelf solutions, that were either too large, or too specialized, both failed to deliver.

We had wasted time on several trail versions of applications, and we did not have what we needed.

Going over my notes I noticed that most decisions could be made based on a few key data points, that could be presented in a simple dashboard. A quick meeting with engineering let me know that a custom application would quick, easy and it was badly needed. After an MVP was working, leadership was amazed.

a mockup of a early prototype of the Mobile‑First Clinical Trials Command Center

Above: Early prototypes proved too ambitions, and did not test much better than commercially available products.

List of Failures Lessons Learned
DO NOT READ


Set a reasonable MVP

Don’t get carried away. It can be exciting to design solutions for 100% of needs, and another 90% of future needs, but with each level of capability comes a new level of complexity, preventing a streamlined workflow.


If it is not good enough for an MVP don’t buy it.

Off the Shelf Options were a costly mistake, Biotech specific applications proved ungainly and too large and too complex for our needs, while smaller applications for general project management did not hold some of the abilities our MVP required.


Involve the future users early

Telling a research team you are going to be adding software to their workflow may not be met with enthusiasm. Work together to map out their current pain points with spreadsheets, and information bottlenecks, and you will get a better product and better cooperation.

Technial Constraints

Balancing Scope With What Mattered Most

A first place ribbon icon

Prioritized for MVP

  • Portfolio dashboard
  • Risk alerts
  • Drill‑down detail views
  • Notes and annotations
  • Real‑time notifications
A second place ribbon icon

Deferred

  • Advanced analytics
  • Video recording
  • Complex data visualizations

Solution

A Mobile‑First Clinical Trials Command Center

Unified Portfolio Dashboard

User Need

See all trials at a glance.

Pain Points

Data scattered across spreadsheets and emails.

Solution

A scrollable portfolio of trial "cards" showing phase, progress, risk status, and upcoming milestones, all in under 30 seconds.

A mock up of the dashboard

Real‑Time Risk Alerts

User Need

Know what’s at risk before it becomes a problem.

Pain Points

Delays surfaced too late.

Solution

Color‑coded risk indicators and push notifications for enrollment issues, site delays, and milestone changes.

A mock up of the notifications screen

Drill‑Down Detail Views

User Need

Understand the full picture when needed.

Pain Points

Dense reports buried key insights.

Solution

  • Enrollment progress
  • Active sites
  • Mini geographic maps
  • Milestone timelines
  • Risk summaries
  • Notes for leadership annotations

A mock up of the trial detail view

A mock up of the risk detail screen

Progressive Disclosure for Cognitive Ease

User Need

Clarity without being overwhelming.

Pain Points

Executives don’t have time for data dumps.

Solution

  • High‑level KPIs first
  • Details only when tapped
  • Plain‑language labels
  • No jargon


A mock up of the export summary screen

Notes & Leadership Annotations

User Need

Capture insights on the go.

Pain Points

Important context lived in memory or email.

Solution

Voice or keypad notes attached directly to each trial.

Removing Extraneous Information

Removing extraneous information, and functions to make the simplest, yet fullest user experience, gave me a new starting point.

the user flow chart

Final Design

A Dashboard Built for Fast, Confident Decision‑Making

Dashboard View

  • Scrollable trial cards
  • Phase, progress, risk status
  • Days to next milestone

Detail View

  • Enrollment
  • Active sites
  • Milestone timeline
  • Risks panel
  • Notes

Risk Alerts

  • Color‑coded indicators
  • Push notifications
  • Tap‑to‑expand risk summaries

Snapshot Export

  • One‑tap PDF or email summary

The result:
Executives can understand the health of all trials in under 30 seconds.

all the screens of the application

Impact

The Difference was Obvious

Key benefits were:

  • Centralized hub; no digging; instant clarity across all trials.
  • Risks surfaced immediately; simplified visuals; actionable insights without noise.
  • Real‑time data; shorter meetings; early intervention prevents costly delays.


A Journey Map of the User Journey of the Medosome Biotec CEO

How Each Pain Point Was Solved

Pain Point: Data scattered across spreadsheets, emails, Excel sheets.

What Wasn't Working: No single source of truth; everything requires manual digging.

How the Application Solves It: Centralizes all trial data into a unified, mobile‑first dashboard with instant visibility.

Pain Point: Risks surface too late; delays discovered after they cause financial damage.

What Wasn't Working: No real‑time alerts; risks buried in dense reports.

How the Application Solves It: Real‑time push notifications and color‑coded risk indicators surface issues instantly.

Pain Point: Meetings start with "catch‑up" instead of decisions.

What Wasn't Working: Everyone arrives with different versions of the truth; spreadsheets slow everything down.

How the Application Solves It: Provides a single, real‑time source of truth that aligns the entire team before meetings begin.

Pain Point: Dense reports bury key insights; too much data shown at once.

What Wasn't Working: No progressive disclosure; high cognitive load.

How the Application Solves It: Uses progressive disclosure, high‑level KPIs first, deeper details only when tapped.

Old User Experience


1. Cab Ride: Preparing to Meet Investors

Trying to get the status of clinical trials, by searching multiple excel documents

Frustrated & Anxious
AI Image of CEO looking at screen, fustrated.
Pain Points
  • Outdated Data
  • Delayed Awareness of Risks
2. Mid-Flight: Decision Making

Trying to conduct analysis to decide the best course of action

Overwhelmed & Confused
AI Image of CEO trying to understand spreadsheets on a smartphone, while in an airplane
Pain Points
  • Difficulty Comparing Data Sets
  • Fragmented Sources
3. Presenting to Venture Capitalists

Presenting, knowing some data is missing, and may be out of date

Unsure & Stressed
Pain Points
  • Being Unprepared
  • Unable to Answer Questions

New User Experience


1. Cab Ride: Preparing to Meet Investors

Getting the status of all trials takes seconds on a phone.

AI image of CEO in an airplane looking at smartphone
Solution
  • Application consolidates current data.
  • Notification alerts notify CEO of all vital information regarding delays or risks delays.
Screenshot of notification screen of application, that shows rescent changes to trials
2. Mid-Flight: Decision Making

CEO can compare trials, or or examine each trial in detail.

AI image of CEO in an airplane looking happy
Solution
  • Key data on all trials is posted in easily scannable and comparable formats.
Screenshot of dashboard screen of application that shows status of all trials
3. Presenting to Venture Capitalists

CEO knows all data is current, and knows they’ll be alerted to any changes.

AI image of CEO content, knowing status of all trials
Solution
  • In depth key data is available for all trials, CEO can leave additional notes on each trial.
Screenshot of detailed view of trials from application
A 3D rendering of 3 pages of the Clinical Trials Command Center

What Did We Deliver?

A mobile‑first dashboard that enables a biotech CEO to monitor real‑time progress across all clinical trials, quickly, clearly, and in the palm of their hand.

Solutions for:

Seeing all trials at a glance

Problem Solved:
Data scattered across spreadsheets and emails


Knowing what's at risk before it becomes a problem

Problem Solved:
Delays surfaced too late


Understanding the full picture when needed

Problem Solved:
Dense reports buried key insights


Clarity without overwhelming

Problem Solved:
Executives don’t have time for data dumps


Capturing insights on the go

Problem Solved:
Important context lived in memory or email


Real-World Outcomes

0%
Financial Loss

Preventable losses due to late or missing information have ended.

15%
Faster Decision‑Making

Leadership can intervene early with investor updates, partnerships, or trial pivots.

30%
Shorter Meetings

Teams arrive aligned, informed, and ready to act.

Instant
Risk awareness through real‑time alerts

Qualitative Feedback

"I need to see what’s at risk before it becomes a problem."

"I don’t have time to dig through spreadsheets on a plane."

"This finally gives me clarity."



The dashboard didn’t just organize information, it transformed how the company operated.

A 3D rendering of 2 pages of the Clinical Trials Command Center

Lessons Learned

Key Takeaways


Intentional simplicity drives adoption

The most powerful version of the product was the simplest one.


Don't take user requests at face value

Users asked for "everything," but testing revealed they needed far less.


Design for the platform, not the trend

A mobile‑first web app was more sustainable than a native iPhone app.


Progressive disclosure is essential for executives

Show what matters now; let them dig deeper only when needed.



Case Studies

Veteran Pathfinder Logo and screen shot

Veteran Pathfinder

Problem Statement:
How might we support veterans in lifting one another up through accessible online resources that ease the emotional, social, and practical challenges of transitioning to civilian life?

Success Metrics:
Veterans using the platform found solutions over 70% of the time as compared to 53% using regular means. Veterans fund their solutions in under 2 minutes as compared to 1 day when using regular means.

footer
About Me | Case Studies | Contact Me
Vets-Path.com © 2026

All content displayed is free of NDA agreements