Health Gorilla
Health Gorilla is a digital healthcare interoperability platform that provides real-time access to connected clinical data, enabling scalable, intelligent, and data-driven healthcare products.
Health Gorilla
Health Gorilla is a digital healthcare interoperability platform that provides real-time access to connected clinical data, enabling scalable, intelligent, and data-driven healthcare products.
COMPANY
Health Gorilla
Role
Senior Product Designer
COMPANY
Health Gorilla
Role
Senior Product Designer
COMPANY
Health Gorilla
Role
Senior Product Designer
Product Discovery
Understanding the business goal.
As healthcare organizations across the United States and Canada adopted the platform to access patient data and clinical workflows within ecosystems connected to thousands of healthcare institutions across North America, we focused on understanding how clinicians interacted with fragmented patient information in real-world scenarios.
The goal was to improve usability, reduce cognitive load, and enable scalable adoption across complex healthcare environments.
Objectives
Identify real usage and behavioral patterns among clinical professionals.
Reduce friction when navigating fragmented clinical data.
Improve clarity and speed across critical healthcare workflows.
Enable scalable adoption across clinics, laboratories, and hospitals.
Insight & Pain point
Through UX Research sessions, interviews, and usability testing with healthcare providers and organizations integrated into the healthcare ecosystem, we identified a critical gap in how clinicians interacted with patient data.
While professionals valued having access to comprehensive clinical information, the experience became fragmented and difficult to navigate in high-pressure scenarios, especially during emergency situations or critical diagnoses.
Instead of navigating across multiple modules — each with its own filters and structures — clinicians needed a single, integrated view that allowed them to quickly understand the patient’s context.
One key insight emerged during the research:
Clinicians do not explore information broadly during critical moments — they focus on a specific condition or diagnosis.
For example, during a cardiac event, clinicians need immediate access to all relevant information — including labs, medications, clinical encounters, and vital signs — connected to that diagnosis, without constantly switching contexts.
Approach (Design Thinking)
1. UX sessions and stakeholder alignment.
We collaborated with clinical professionals, Product Managers, and business stakeholders to observe real-world workflows, understand user behavior patterns, and identify key pain points across the experience.
2. Journey Mapping and workflow analysis.
We mapped how providers accessed, interpreted, and navigated across multiple clinical data sources to identify friction points, context switching, and opportunities for optimization.
3. Research synthesis and pain point identification.
Based on research analysis and workflow evaluation, we identified the main friction points:
High cognitive load
Fragmented clinical information views
Lack of visual hierarchy and contextual clarity
Complex navigation across modules and filters
4. Rapid prototyping and iteration.
We explored different interaction models through wireframes, coded interactive prototypes and iterative user validations to evaluate clarity, efficiency and usability within critical clinical scenarios.
Design Solution
Designing the experience.
The solution keeps HumanGraph's longitudinal view and the product's MUI components, and reorganizes the page around the question clinicians bring to it: what in this record matters for the problem in front of me? The view starts from the encounters, and everything else (provider, labs, results) lives inside each visit. The interactive prototype at the top of this page walks through the full flow.
Clarity over density → Show the encounters first and keep the main view clean.
Start from the diagnosis → Let clinicians enter the condition they are working on and see only the visits related to it.
Progressive disclosure → Surface the visit first, and open its provider, labs and results only when needed.
Source first → Abnormal flags come from the source lab or system. Relevance to a diagnosis is inferred and never shown as a diagnosis.
Systems thinking → Build with the fields, tabs, chips and side panels already in the design system.
The flow, step by step
1. Enter a diagnosis
Typing “Chest pain” suggests “Chest pain / Suspected ACS”. The related visits are highlighted on the encounter timeline and listed below it, each with its date, type, provider and how many labs and abnormal values it has.
2. Narrow by provider, location or date
Three simple filters sit next to the diagnosis: provider by name, location by facility, and date with quick presets or a custom start and end date. Active filters show as removable chips with a single Clear filters, and the timeline rescales to the selected dates.
3. Open a visit
Selecting a visit opens a side panel with the provider, the labs and vitals from that encounter with their source flags, and the results available, such as ECG or imaging reports, along with the source document.
4. Review all labs and results
The Labs & results tab lists every result from the related visits, one per row: the test and its type (for example a blood test, imaging or an ECG), the value with the source’s flag and reference range, the facility, the ordering provider and the related visit, the date, and any comment from the lab. Each row links to the full lab report, so comparing values no longer means opening visits one by one.
5. Ask the record (future AI integration)
As part of the roadmap for future AI integrations, the HumanGraph Assistant would let clinicians get to information quickly in plain language. Asking “Any abnormal values?” lists the values flagged by the source, cites the record each one comes from and filters the table to them. A follow-up such as “How has troponin changed?” compares the value across visits. Answers summarize source records only and don't make clinical recommendations.
Before vs. after
Before: every category and test on one dense timeline, filtered only by category and date, with details spread across modules.
After: start from a diagnosis, see the related visits, narrow by provider, location or date, open a visit for its details, or review all labs and results in a single table.
Outcome
Clinicians can now enter the experience with a specific focus and instantly access all relevant information in a single place, enabling faster understanding and more efficient, confident decision-making during critical moments.
Impact
Optimized clinical interpretation → Reduced cognitive load through a more contextual and integrated experience.
High module adoption → Drove 80%+ recurring usage across a network of 1,500+ clinics and healthcare organizations in the US and Canada.
Ecosystem scalability → Supported large volumes of data and multiple interoperable sources.
What's next
The goals for this iteration are to reduce the time to the first relevant finding, cut the steps needed to check a value against its source, and keep the main view readable in high-pressure moments. These are expected improvements to validate with clinicians in usability testing, not measured results.
Product Discovery
Understanding the business goal.
As healthcare organizations across the United States and Canada adopted the platform to access patient data and clinical workflows within ecosystems connected to thousands of healthcare institutions across North America, we focused on understanding how clinicians interacted with fragmented patient information in real-world scenarios.
The goal was to improve usability, reduce cognitive load, and enable scalable adoption across complex healthcare environments.
Objectives
Identify real usage and behavioral patterns among clinical professionals.
Reduce friction when navigating fragmented clinical data.
Improve clarity and speed across critical healthcare workflows.
Enable scalable adoption across clinics, laboratories, and hospitals.
Insight & Pain point
Through UX Research sessions, interviews, and usability testing with healthcare providers and organizations integrated into the healthcare ecosystem, we identified a critical gap in how clinicians interacted with patient data.
While professionals valued having access to comprehensive clinical information, the experience became fragmented and difficult to navigate in high-pressure scenarios, especially during emergency situations or critical diagnoses.
Instead of navigating across multiple modules — each with its own filters and structures — clinicians needed a single, integrated view that allowed them to quickly understand the patient’s context.
One key insight emerged during the research:
Clinicians do not explore information broadly during critical moments — they focus on a specific condition or diagnosis.
For example, during a cardiac event, clinicians need immediate access to all relevant information — including labs, medications, clinical encounters, and vital signs — connected to that diagnosis, without constantly switching contexts.
Approach (Design Thinking)
1. UX sessions and stakeholder alignment.
We collaborated with clinical professionals, Product Managers, and business stakeholders to observe real-world workflows, understand user behavior patterns, and identify key pain points across the experience.
2. Journey Mapping and workflow analysis.
We mapped how providers accessed, interpreted, and navigated across multiple clinical data sources to identify friction points, context switching, and opportunities for optimization.
3. Research synthesis and pain point identification.
Based on research analysis and workflow evaluation, we identified the main friction points:
High cognitive load
Fragmented clinical information views
Lack of visual hierarchy and contextual clarity
Complex navigation across modules and filters
4. Rapid prototyping and iteration.
We explored different interaction models through wireframes, coded interactive prototypes and iterative user validations to evaluate clarity, efficiency and usability within critical clinical scenarios.
Design Solution
Designing the experience.
The solution keeps HumanGraph's longitudinal view and the product's MUI components, and reorganizes the page around the question clinicians bring to it: what in this record matters for the problem in front of me? The view starts from the encounters, and everything else (provider, labs, results) lives inside each visit. The interactive prototype at the top of this page walks through the full flow.
Clarity over density → Show the encounters first and keep the main view clean.
Start from the diagnosis → Let clinicians enter the condition they are working on and see only the visits related to it.
Progressive disclosure → Surface the visit first, and open its provider, labs and results only when needed.
Source first → Abnormal flags come from the source lab or system. Relevance to a diagnosis is inferred and never shown as a diagnosis.
Systems thinking → Build with the fields, tabs, chips and side panels already in the design system.
The flow, step by step
1. Enter a diagnosis
Typing “Chest pain” suggests “Chest pain / Suspected ACS”. The related visits are highlighted on the encounter timeline and listed below it, each with its date, type, provider and how many labs and abnormal values it has.
2. Narrow by provider, location or date
Three simple filters sit next to the diagnosis: provider by name, location by facility, and date with quick presets or a custom start and end date. Active filters show as removable chips with a single Clear filters, and the timeline rescales to the selected dates.
3. Open a visit
Selecting a visit opens a side panel with the provider, the labs and vitals from that encounter with their source flags, and the results available, such as ECG or imaging reports, along with the source document.
4. Review all labs and results
The Labs & results tab lists every result from the related visits, one per row: the test and its type (for example a blood test, imaging or an ECG), the value with the source’s flag and reference range, the facility, the ordering provider and the related visit, the date, and any comment from the lab. Each row links to the full lab report, so comparing values no longer means opening visits one by one.
5. Ask the record (future AI integration)
As part of the roadmap for future AI integrations, the HumanGraph Assistant would let clinicians get to information quickly in plain language. Asking “Any abnormal values?” lists the values flagged by the source, cites the record each one comes from and filters the table to them. A follow-up such as “How has troponin changed?” compares the value across visits. Answers summarize source records only and don't make clinical recommendations.
Before vs. after
Before: every category and test on one dense timeline, filtered only by category and date, with details spread across modules.
After: start from a diagnosis, see the related visits, narrow by provider, location or date, open a visit for its details, or review all labs and results in a single table.
Outcome
Clinicians can now enter the experience with a specific focus and instantly access all relevant information in a single place, enabling faster understanding and more efficient, confident decision-making during critical moments.
Impact
Optimized clinical interpretation → Reduced cognitive load through a more contextual and integrated experience.
High module adoption → Drove 80%+ recurring usage across a network of 1,500+ clinics and healthcare organizations in the US and Canada.
Ecosystem scalability → Supported large volumes of data and multiple interoperable sources.
What's next
The goals for this iteration are to reduce the time to the first relevant finding, cut the steps needed to check a value against its source, and keep the main view readable in high-pressure moments. These are expected improvements to validate with clinicians in usability testing, not measured results.
then open any visit to see its provider, labs and results.
Product Discovery
Understanding the business goal.
As healthcare organizations across the United States and Canada adopted the platform to access patient data and clinical workflows within ecosystems connected to thousands of healthcare institutions across North America, we focused on understanding how clinicians interacted with fragmented patient information in real-world scenarios.
The goal was to improve usability, reduce cognitive load, and enable scalable adoption across complex healthcare environments.
Objectives
Identify real usage and behavioral patterns among clinical professionals.
Reduce friction when navigating fragmented clinical data.
Improve clarity and speed across critical healthcare workflows.
Enable scalable adoption across clinics, laboratories, and hospitals.
Insight & Pain point
Through UX Research sessions, interviews, and usability testing with healthcare providers and organizations integrated into the healthcare ecosystem, we identified a critical gap in how clinicians interacted with patient data.
While professionals valued having access to comprehensive clinical information, the experience became fragmented and difficult to navigate in high-pressure scenarios, especially during emergency situations or critical diagnoses.
Instead of navigating across multiple modules — each with its own filters and structures — clinicians needed a single, integrated view that allowed them to quickly understand the patient’s context.
One key insight emerged during the research:
Clinicians do not explore information broadly during critical moments — they focus on a specific condition or diagnosis.
For example, during a cardiac event, clinicians need immediate access to all relevant information — including labs, medications, clinical encounters, and vital signs — connected to that diagnosis, without constantly switching contexts.
Approach (Design Thinking)
1. UX sessions and stakeholder alignment.
We collaborated with clinical professionals, Product Managers, and business stakeholders to observe real-world workflows, understand user behavior patterns, and identify key pain points across the experience.
2. Journey Mapping and workflow analysis.
We mapped how providers accessed, interpreted, and navigated across multiple clinical data sources to identify friction points, context switching, and opportunities for optimization.
3. Research synthesis and pain point identification.
Based on research analysis and workflow evaluation, we identified the main friction points:
High cognitive load
Fragmented clinical information views
Lack of visual hierarchy and contextual clarity
Complex navigation across modules and filters
4. Rapid prototyping and iteration.
We explored different interaction models through wireframes, coded interactive prototypes and iterative user validations to evaluate clarity, efficiency and usability within critical clinical scenarios.
Design Solution
Designing the experience.
The solution keeps HumanGraph's longitudinal view and the product's MUI components, and reorganizes the page around the question clinicians bring to it: what in this record matters for the problem in front of me? The view starts from the encounters, and everything else (provider, labs, results) lives inside each visit. The interactive prototype at the top of this page walks through the full flow.
Clarity over density → Show the encounters first and keep the main view clean.
Start from the diagnosis → Let clinicians enter the condition they are working on and see only the visits related to it.
Progressive disclosure → Surface the visit first, and open its provider, labs and results only when needed.
Source first → Abnormal flags come from the source lab or system. Relevance to a diagnosis is inferred and never shown as a diagnosis.
Systems thinking → Build with the fields, tabs, chips and side panels already in the design system.
The flow, step by step
1. Enter a diagnosis
Typing “Chest pain” suggests “Chest pain / Suspected ACS”. The related visits are highlighted on the encounter timeline and listed below it, each with its date, type, provider and how many labs and abnormal values it has.
2. Narrow by provider, location or date
Three simple filters sit next to the diagnosis: provider by name, location by facility, and date with quick presets or a custom start and end date. Active filters show as removable chips with a single Clear filters, and the timeline rescales to the selected dates.
3. Open a visit
Selecting a visit opens a side panel with the provider, the labs and vitals from that encounter with their source flags, and the results available, such as ECG or imaging reports, along with the source document.
4. Review all labs and results
The Labs & results tab lists every result from the related visits, one per row: the test and its type (for example a blood test, imaging or an ECG), the value with the source’s flag and reference range, the facility, the ordering provider and the related visit, the date, and any comment from the lab. Each row links to the full lab report, so comparing values no longer means opening visits one by one.
5. Ask the record (future AI integration)
As part of the roadmap for future AI integrations, the HumanGraph Assistant would let clinicians get to information quickly in plain language. Asking “Any abnormal values?” lists the values flagged by the source, cites the record each one comes from and filters the table to them. A follow-up such as “How has troponin changed?” compares the value across visits. Answers summarize source records only and don't make clinical recommendations.
Before vs. after
Before: every category and test on one dense timeline, filtered only by category and date, with details spread across modules.
After: start from a diagnosis, see the related visits, narrow by provider, location or date, open a visit for its details, or review all labs and results in a single table.
Outcome
Clinicians can now enter the experience with a specific focus and instantly access all relevant information in a single place, enabling faster understanding and more efficient, confident decision-making during critical moments.
Impact
Optimized clinical interpretation → Reduced cognitive load through a more contextual and integrated experience.
High module adoption → Drove 80%+ recurring usage across a network of 1,500+ clinics and healthcare organizations in the US and Canada.
Ecosystem scalability → Supported large volumes of data and multiple interoperable sources.
What's next
The goals for this iteration are to reduce the time to the first relevant finding, cut the steps needed to check a value against its source, and keep the main view readable in high-pressure moments. These are expected improvements to validate with clinicians in usability testing, not measured results.
