Parameter
A secure AI health app that turns complex data into clear insights
Health Tech
AI Integration
Timeline
7 Weeks (Oct 27 - Dec 11 2025)
Role
Tools
Overview
Parameter is a patient-facing genetic health app designed to reduce health data fragmentation. By centralizing daily health metrics, wearable signals, and medical reports into a single experience, Parameter helps people understand how their data connects over time and what it means for their health.
Design Challenge
How might we design a trusted experience that turns fragmented health data into clear, actionable guidance?
Problem Space
Managing your health shouldn't feel like a second job
What's happening today
People have more access to their health data than ever, from wearables, labs, and provider portals. But that data lives across disconnected platforms, often filled with medical jargon and numbers without context. Patients are left to connect the dots themselves, turning scattered information into decisions without clear guidance.
50%
of adults had health records spread across more than one portal or provider type
Solution
From raw metrics to understandable, personalized guidance
How the experience works
Parameter is designed around four connected moments: personalised dashboard, health information in plain language, AI integration, and data aggregation. Together these layers reduce cognitive load, improve comprehension, and help users translate information into practical next steps.
Key Screens
From raw metrics to understandable, personalized guidance
How the experience works
Parameter is designed around four connected moments: personalised dashboard, health information in plain language, AI integration, and data aggregation. Together these layers reduce cognitive load, improve comprehension, and help users translate information into practical next steps.
Key Feature 1
Customizable metrics dashboard
No two patients track the same things. Parameter lets users pin exactly what matters to them, so the first screen they see is built around their condition and goals.
Key Feature 2
Contextual AI health assistant
The AI assistant is a single hub that branches into whatever you need: interpreting a lab result, spotting a trend, prepping for an appointment, or unpacking what a genetic predisposition means for daily life. It's framed as educational, not diagnostic.
Key Feature 3
Reports, labs, and metrics in one view
Most people piece their health picture together from a patient portal, a wearable app, a separate lab results site, and a folder of paper documents. Parameter pulls all of it into a single organized view: reports, lab results, and daily health metrics side by side.
Key Feature 4
Plain-language results and FAQ
Lab values, genetic risk factors, and health metrics are presented with plain-language descriptions, color-coded range indicators, and an FAQ section for deeper context. Instead of showing a number and leaving you to Google it, Parameter tells you what it means and what's healthy.
User Interviews
Health data is everywhere, but understanding it is not
Who I talked to
I interviewed 7 participants: 5 patients managing conditions like diabetes, glaucoma, and bipolar disorder, and 2 medical professionals, to explore the gap between how health information is managed today and how it could be.
The research focused on:
How patients track and manage information across tools
Where digital health tools fall short
What barriers exist between patients and doctors in accessing and communicating health information
1.0 Affinity map of patient and provider interviews, revealing six themes in a fragmented healthcare system
Key Insights
Too many tools, too little clarity, and a gap between patients and their doctors
Health data is fragmented and manually managed
"I use Excel sheets, keep paper documents in a folder, or jot things down in my Notes app."
Most patients only engage with their data reactively
"I only look at my health data when something feels wrong or I have an appointment coming up."
Difficulty understanding health information
"I can see my results, but I don't really know what they mean or what I should do next."
Trust, usability, and privacy concerns
"I don't use most health apps because they feel confusing, and I'm not sure who can see my data."
Competitive Research
Today’s tools are improving, but still fragmented for patients.
Competitive takeaway
Across wearables, portals, and consumer health apps, each tool solves part of the journey, yet users still have to assemble the full story themselves. Parameter positions itself as the connective layer.
2.0 How current health apps structure data: by type, by metric, or chronologically
2.1 Health data is visible but not understandable: dense, decontextualized, and hard to act on
Key Insight 1
Organization doesn't equal understanding
Apps structure data by record type or metric, prioritizing storage logic over how users make sense of their data.
Key Insight 2
Numbers displayed without enough context
Charts and values appear without healthy ranges or context, leaving users unsure what their data means.
Key Insight 3
Information overloads without clarity
Even when data is visible, users get no guidance on what to do next, turning information into noise.
Ideation & Testing
Translating complex healthcare constraints into a usable product flow
Testing approach
I tested four core flows with patients and providers to surface friction in navigation, comprehension, and trust. Findings shaped how Parameter handles context, categorization, flexibility, and guidance.
3.0 User feedback on the home dashboard surfacing needs for visual context and customization
3.1 Testing navigation structures for health data, with users favoring categorization but flagging terminology friction
3.2 Comparing three record views, with users preferring a timeline-first approach with category filtering
3.3 Three AI hub variations tested, showing structured prompts help until choice overload sets in.
Reflection & Next Steps
Key takeaway
Parameter reinforced that trust in health products comes from clarity, not capability. The most valuable shift was moving from "showing more data" to "supporting better decisions," which sharpened my approach to information architecture in sensitive domains where language, hierarchy, and timing shape user confidence.
Paths for further development
Clinical Validation
Integrate directly with additional labs, providers, and wearable ecosystems to reduce manual upload friction.
Personalization Engine
Expand adaptive models that learn from user behavior, preferences, and history to make guidance more relevant over time.
Data Partnerships
Integrate directly with additional labs, providers, and wearable ecosystems to reduce manual upload friction.
Longitudinal Testing
Run longitudinal pilots to measure comprehension, confidence, and behavior change across longer care cycles.